Kilnwood, Blenheim Hill, Harwell, Oxfordshire

Transcription

1 Kilnwood, Blenheim Hill, Harwell, Oxfordshire An Archaeological Excavation for Westbuild Homes Ltd by Andy Taylor Thames Valley Archaeological Services Ltd Site Code KBH 6/06 November 206

2 Summary Site name: Kilnwood, Blenheim Hill, Harwell, Oxfordshire Grid reference: SU Site activity: Archaeological Excavation Date and duration of project: 5th-8th July 206 Project Coordinator: Tim Dawson Site supervisor: Andy Taylor Site code: KBH 6/06 Area of site: 936 sq m Summary of results: The excavation found further remains of later Roman date, and uncovered four main ditches forming two enclosures. Several pits and postholes, likely to be of contemporary date were also identified, Four inhumation burials of Middle Saxon date were also found, two of which were radiocarbon dated. A few prehistoric struck flints including a microlith and a possible Bronze Age pit point to a little prehistoric activity on the site. Location and reference of archive: The archive is presently held at Thames Valley Archaeological Services, Reading and will be deposited with Oxfordshire Museum Service in due course with accession number OXCMS: This report may be copied for bona fide research or planning purposes without the explicit permission of the copyright holder. All TVAS unpublished fieldwork reports are available on our website: Report edited/checked by: Steve Ford Steve Preston i Thames Valley Archaeological Services Ltd, De Beauvoir Road, Reading RG 5NR Tel. (08) ; Fax (08) ; tvas@tvas.co.uk; website:

3 Kilnwood, Blenheim Hill, Harwell, Oxfordshire An Archaeological Excavation by Andy Taylor Report 6/06b Introduction This report documents the results of an archaeological recording action carried out at the former site of Kilnwood, on Blenheim Hill, Harwell, Oxfordshire, OX 0DS (SU ) (Fig ). The work was commissioned by Mr Matthew Brook of Westbuild Homes Ltd, Hunters Lodge, Rectory Road, Padworth Common, Reading, Berkshire, RG7 4JB. Planning permission (P5/V0324/FUL) has been granted by the Vale of the White Horse District Council to erect four new dwellings with garages on the site of the existing house and garden. The consent is subject to two conditions (5 and 6) relating to archaeology requiring a programme of investigation prior to the redevelopment of the site. This is the second stage of investigation on the site, after initial evaluation trenching uncovered late Roman ditches and pits, indicating a possible occupation site and a lone inhumation (Taylor 206). These works are in accordance with the Department of Communities and Local Government's National Planning Policy Framework (NPPF 202), and the District Councils policies on archaeology. The field investigation was carried out to a specification approved by Mr Hugh Coddington, Archaeology Team Leader with Oxfordshire County Council, advisers to the District on matters relating to archaeology. The fieldwork was undertaken by the author with assistance from Tom Stewart, Cosmo Bacon and Michael Johnson. The site code is KBH 6/06. The archive is presently held at Thames Valley Archaeological Services, Reading and will be deposited with Oxfordshire Museum Service in due course, with the accession code OXCMS: Human remains were excavated under Ministry of Justice licence Location, topography and geology The site is located on the northern side of Blenheim Hill on the eastern edge of the village of Harwell (Fig. ) in Oxfordshire, which lies 2km to the west of the larger urban centre of Didcot. It is a roughly rectangular plot of land and consists of a 0.3ha garden which has had the house known as Kilnwood demolished at its centre. The site is on the eastern end of a row of residential housing with arable fields neighbouring to the north and east (Fig 2). The stripped area covered 936 sq m, and has boundaries with tree protection zones to

4 the east and south (Fig. 3). The underlying geology is mapped as Upper Greensand Formation, which is a mix of calcareous sandstone and siltstone (BGS 206), with the archaeological horizon exposed at the top of this deposit. The site lies at a height of c.75m above Ordnance Datum. Archaeological background The archaeological potential of the site has been identified through previous work at the site (Taylor 206) and is summarized by Oxfordshire County Archaeological Service in the design brief (Coddington 206). The site lies within the Upper Thames Valley which has a wealth of archaeological sites on the gravel terraces, and of particular importance for this region, the Chalk Downland (Booth et al 2007; Pine and Preston 205). Iron Age settlements in the area are characterised by pit clusters, small enclosures and post built structures (Lambrick 2009). Roman settlement tends to be characterized by small farmsteads containing wooden structures within enclosures, which occasionally develop in the later period to 'villa' stone buildings with associated field systems. Extensive Iron Age and Roman settlement has been recorded at Great Western Park on the west side of the ever-developing Didcot. Recent fieldwork for Harwell village during new development, however, has focused on the archaeological evidence in the adjoining fields. On the western side of the village at Grove Road, geophysical survey and trenching evaluation and later excavation covering 2ha has identified settlement evidence of Iron Age and Roman date (Powell 204; Thompson and Andrews forthcoming). Findspots of Roman coins, located around the village, are recorded in the Historic Environment Record (HER). It is also thought that 5th-century Saxon inhumations found south of the village may be situated in proximity to an earlier (prehistoric) funerary monument. On the eastern side of the village, off Blenheim Hill, recent evaluation and excavation has confirmed deposits survive concentrated close to the main road consisting of densely packed later Roman features with ditches, pits, postholes, a corn dryer and a coin hoard (McNicoll-Norbury and Taylor 206; Taylor forthcoming). Evaluation in the garden of Kilnwood itself identified further late Roman boundary features, pits and a truncated human burial (Taylor 206). Immediately to the east of the site a geophysical survey (Thompson 206) and subsequent evaluation confirmed the presence of further linear or enclosure features of probable similar date. Objectives and methodology The general objectives of the excavation were: 2

5 to excavate and record all archaeological deposits and features within the areas threatened by the development. to produce relative and absolute dating and phasing for deposits and features recorded on the site. to establish the character of these deposits in attempt to define functional areas on the site such as industrial, domestic etc; and to produce information on the economy and local environment and compare and contrast this with the results of other excavations in the region. Specific research objectives were to address the following questions: When was the site first utilised and when was it abandoned? What is the nature and origin of the features previously recorded on the site? Are there any further human inhumation burials present on the site? What is the palaeoenvironmental setting of the area? Topsoil and subsoil were removed by a 360 type machine fitted with a toothless grading bucket under constant archaeological supervision. Excavation of the footings for the previous house was also carried out under archaeological observation to restrict damage to underlying deposits. Tree protection was in place during the course of excavation. All archaeological features were to be planned and sectioned as a minimum objective, with subsequent excavation or sampling to an agreed fraction depending on the nature of the feature/deposit. The excavated features are summarized in Appendix. Results The excavation confirmed the remains located in the evaluation (Fig. 3) and revealed further features. In total, two enclosures were identified along with a further three burials in addition to the one identified during the previous trenching exercise. Seven postholes and eight pits were also excavated. Phase : Bronze Age Although no securely dated features from this period were identified, a single, abraded, sherd of Bronze Age urn was recovered from pit. It is possible that this single sherd dates the pit but more likely this is residual material: in either case, it does indicate some Bronze Age activity within close proximity of the site. 3

6 Phase 2.: Roman Enclosure This enclosure is speculative as it only consists of two sides and continues both westwards and southwards. Its northern side comprised ditch 200 with an entranceway and continuing into ditch (204) which turned and was cut by ditch had three slots (36/37, 39, 40) dug across it and measured between 0.80m and.50m wide and between 0.20m and 0.55m deep and contained 0 sherds of pottery from along its length, six pieces of animal bone and two pieces of fired clay. 204 had three slots (20, 30, 34) dug into it and measured between 0.54m and 0.74m wide and between 0.09m and 0.35m deep with 20 sherds of pottery and four pieces of animal bone recovered from its length. Gully 203 probably also forms part of an addition to this enclosure. This was 0.27m wide and 0.2m deep but did not produce any finds. Phase 2.2: Roman Enclosure 2 The second enclosure almost acted as an annexe to the larger enclosure consisting of ditches 20 and 205 with 202 as a possible redefined northern edge. 20 formed most of three sides of the feature and had nine slots (2, 4, 7, 0, 5, 7, 22, 26, 35) dug along its length and measured between 0.7m and.66m wide and between 0.26m and 0.66m deep. 87 sherds of pottery (5 from the evaluation), 39 pieces of animal bone (four from the evaluation), five pieces of fired clay and four nails and an iron awl were recovered along its length. 205 had four slots (9, 29, 3, 32) dug into it measuring between.40m and.65m wide and between 0.36m and 0.360m deep. It produced sherds of pottery, 24 pieces of animal bone, a piece of tile, a piece of fired clay and a nail. Between the terminal ends 26 and 32 was a small stretch of gully (203), which was cut by both terminal ends of ditches 20 and 205: it may also have served as part of the earlier enclosure, suggesting that enclosure 2 was almost wholly a replacement. On the outer western side of the enclosure was post hole group 206, which consisted of postholes forming a rhombus which may represent a four post 'structure' with sides c..5m long. These measured between 0.6m and 0.30m in diameter and between 0.04m and 0.m deep. Only 09 produced dating evidence, a sherd of pottery and a piece of struck flint and as a result this is tentatively given a contemporary date with the enclosure. Discrete Features with Finds Several features could belong to either phase of Roman activity. Pit/terminus 00 measured 0.40m wide and 0.9m deep and contained a sherd of pottery. 4

7 Pit 05 measured.86m wide and 0.24m deep and produced 3 sherds of pottery, four pieces of animal bone and a metal plate. Pits -4 were a series of inter-cutting features with measuring.82m wide, 0.49m deep and containing a piece of residual Bronze Age urn and two pieces of animal bone. 2 measured m wide, 0.32m and 3, which cut 2, measured 0.80m wide and 0.30m deep. Neither of these produced any dating evidence. 4 cut 3 and measured 0.30m wide, 0.25m deep and produced a piece of fired clay and a nail. Pit 28 measured 0.6m in diameter, 0.04m deep and contained an iron nail. Phase 2.3: Roman Ditch 202 formed part of enclosure 2 and was on the same alignment as the northern side of 20, which it cut, and as such may be a re-defining of that edge, It had four slots (3, 6, 2, 23) dug into it and revealed it to be.2m and.4m wide and between 0.45m and 0.5m deep. It contained seven sherds of pottery (two from the evaluation), mostly Roman but also a Saxon sherd, considered to be intrusive, 9 pieces of animal bone (eight from the evaluation), an iron hook and a piece of fired clay. Phase 3: Anglo-Saxon The Anglo-Saxon activity on the site comprised the digging of a single pit and four graves. Pit 02 was shallow measuring 0.40m wide but only 0.05m deep producing two sherds of Saxon pottery and a piece of animal bone. Graves Four inhumation burials, in varying states of completeness and preservation, were recorded from both evaluation and excavation phases of works. Grave cut 5 contained the remains of skeleton 64. The grave measured.77m long, 0.57m wide and 0.24m deep. This contained two sherds of pottery (one Roman, one Iron Age) and a piece of struck flint. None of these seem likely to be deliberate grave goods, but rather accidental inclusions in the backfill. Of note was the remains of an iron blade, which may indicate a later date (knifes are common accompaniments in Saxon graves). Only the legs of the skeleton were present, with the remainder of the body having been truncated by the construction of the former house. Grave 24 contained partial skeleton 79, which consisted of the lower jaw, arms, legs and partial pelvic remains. A tooth from this skeleton returned a radiocarbon date of cal AD (UB32878). It measured.55m long, 0.64m wide and 0.2m deep. An iron blade was located across the right side of the pelvis, as well as a buckle and nail. 5

8 Grave 25 contained the remains of very partial skeleton 82, which consisted of a leg only. The grave measured m long, 0.56m wide and 0.6m deep. The grave fill contained four sherds of pottery (one Roman and three Saxon) a piece of animal bone and an iron hook, although, with the exception of the hook, these are again likely to be accidental finds in the backfill rather than deliberately placed grave goods. Grave 38 contained skeleton 250, which comprised the most complete burial from the site, consisting of skull, arms, legs and partial pelvic remains. A tooth from this skeleton returned a radiocarbon date of cal AD (UB32877), markedly later than skeleton79. Six glass beads were located around the neck of the body along with an iron blade on the edge of the grave cut. Also, two fragmentary silver/silver gilt rings were recovered. Finds Pottery by Malcolm Lyne The excavation of the site yielded 38 sherds (2389g) of pottery from 28 contexts: a single further sherd was retrieved from environmental samples (Appendix 2). Most of the pottery is Roman and can be dated to c One abraded and residual Bronze Age urn fragment is also present as are nine c dated Early Saxon potsherds and a fragment from an 8th or 9th c. bread crock. All of the pottery assemblages were quantified by numbers of sherds and their weights per fabric. These fabrics were identified using a x8 magnification lens with built-in metric graticule in order to determine the natures, forms, sizes and frequencies of added inclusions and those naturally present in the potting clay. Four numbered fabric series were drawn up with the prefixes BA, C, F and ES for Bronze Age, Coarse Roman, Fine Roman and Early Saxon respectively (Appendix 2). None of the pottery assemblages are large enough for further quantification by Estimated Vessel Equivalents (Orton 975). Bronze Age A single abraded urn sherd in coarse calcined-flint tempered fabric came from Pit at the northern end of the site. This feature did not yield any other pottery. Roman Assemblage. From the fills of Ditch 200 (97, 25 and 252). This ditch yielded a mere six sherds (34g) of Late Roman pottery, comprising one sherd each in fine Oxfordshire Greyware and Oxfordshire Oxidised 6

9 ware and two each from an indeterminate Oxfordshire Red Colour-coat beaker (c.ad ) and BB oval dish of Bestwall Class 9 (Lyne 202, c.ad ). Assemblage 2. From the fills of Ditches 20, 205 and 204. The 99 sherds (784 g.) of pottery from these enclosure ditches are, for the most part, similar in date to those from Ditch 200: there is, however, a little late-2nd century pottery from Ditch 20 as well, including fragments from M3 and M0 Oxfordshire Whiteware mortaria (c.ad and respectively). Oxfordshire industry products account for 6% of the sherds and include the above as well as those from a greyware dish of Young s (977) type R6 (c.ad ), an oxidized one of type O38. (c.ad ), a parchment ware jar of type P8 (c.ad ) and a bottle, dish of type C50 and bowl of type C52 in Oxfordshire Red Colour-coat fabric (c.ad , and respectively). Ten sherds from straight-sided dishes in BB (c.ad ) are also present as is one sherd each from a handmade jar in shell-tempered ware (c.ad ), a wheelturned example from the Harrold kilns (c.ad ) and a beaded-and-flanged bowl (c.ad ) in miscellaneous greyware. Early Saxon Assemblage 3. From the fills of Ditch 202 (Contexts 67, 69 and 78). This ditch yielded just 5 sherds (63g) of pottery, of which four are featureless bodysherds from closed forms in Oxfordshire Greyware and the fifth from a handmade vessel in silty brown-black fabric ES with internal polish. The Roman sherds cannot be dated with any precision but it could be that the ditch was late Roman and continued to receive rubbish into the Early Saxon period. It could equally-well be that the ditch was Early Saxon and is part of the conversion of the enclosure into a small cemetery of that period. Assemblage 4. From the fill of Grave 25 (Context 8). The four small sherds from this grave comprise a tiny chip in Oxfordshire Greyware fabric C2D, one in silty black Early Saxon fabric ES and two in silty chaff-tempered fabric ES2. The other three graves lacked pottery with the exception of Grave 5, which yielded another small abraded chip of Oxfordshire Greyware. Assemblage 5. From the fill of Pit 02 north of the enclosure (Context 52). This feature produced two large sherds (273g) making up the lower half of a black Early Saxon cooking-pot in silty chaff-tempered fabric ES2 with both external and internal polish. 7

10 Human Bone by Ceri Falys Human skeletal remains were excavated from three contexts within the excavation area (SK79, 82 and 250) and a fourth from the evaluation. The inhumations were found close together and organized into a row, running approximately SW-NE across the site. The graves were all roughly rectangular in plan, measured between.00m and.70m long, and between 0.56m to 0.70m wide (though grave 5 from the evaluation had been truncated by the construction of the recently demolished building. All graves were shallow, ranging from just 0.2m to 0.20m deep. Grave 25 was exceptionally shallow at the north-west end, likely resulting in the truncation and removal of the majority of the upper half of the body (SK82). The bodies were all orientated NW(head)-SE(feet). Two individuals were in supine and extended positions (SK79 and 250). The third skeleton, SK82, was comprised solely of a left leg, which was in a semi-flexed (crouched) position. Grave 5 was heavily truncated by a modern patio, the rest of the grave contained the lower half of a human skeleton (SK64). Orientated NW-SE (feet), the lower body of this individual was supine, with the right leg slightly bent at the knee, the left leg fully extended, and the feet close together. It was noted at the time of excavation, the bone was poorly preserved with a soft and fragile texture. Additional artefacts were recovered from within each grave. With the exception of glass beads from the neck region of SK250, and a few sherds of residual Roman and Saxon pottery with SK 82, all other associated items were metallic. Four pieces of metalwork were found with SK79, including a knife, a large flat pin, a nail and a copper alloy buckle. A "right angled hook" was found with SK82, and a knife was buried with SK250 and a metal blade was recovered from the pelvic area of SK64. Osteological analysis was undertaken following the guidelines of Buikstra and Ubelaker (994) and Brickley and McKinley (2004) in order to determine the minimum number of individuals present, the extent of preservation and completeness of the skeletons, assessments of the sex of the individuals and their approximate ages at death. Preservation permitting, identification of pathological alterations and presence of non-metric traits were also undertaken. The results of the osteological analysis for each individual, and illustrations of pathologies, can be found in Appendix A-D. Minimum Number of Individuals As expected, each grave was found to contain evidence of a single skeleton. The remains of a minimum of four individuals were present at the time of analysis. 8

11 Preservation Skeletal preservation depends on several factors, both intrinsic (i.e. age, sex, overall size/robusticity) and extrinsic (i.e. factors in the burial environment, post-depositional disturbance and post-excavation treatment) to the skeleton itself. The overall state of preservation was assessed using a generalized grading system of three categories: good, fair and poor. All elements were fragmented and large areas of compact bone were eroded, giving an overall weathered and worn appearance to the skeletal elements. Trabecular bone was not preserved, which resulted in the absence of the axial skeletal elements (i.e. spine, ribs, pelvis, ends of the long bones). Although the teeth were chalky and fragile in texture, they were the best preserved aspects of the skeletons (SK79 and 250). As a result, no elements were deemed to be of "good" preservation (large areas of well preserved cortical and trabecular bone). All three skeletons displayed "poor to fair" preservation, characterised by some erosion of the cortical bone surface (fair preservation), however, the lack of survival of the trabecular bone inhibiting pathological analysis rendered a designation of "poor" preservation. The state of preservation limited the amount of retrievable demographic and pathological information from the remains Completeness The completeness of the skeletons varied between the inhumations, with no individuals 00% complete at the time of excavation. SK79 and 250 comprised approximately 25-50% of the expected number of skeletal elements (portions of skull, arms and legs), and SK82 was <25% complete (left leg only). The absence of elements primarily made up of trabecular bone (i.e. vertebrae, ribs, pelvis, ends of the long bones) contributed to the lack of completeness. SK64 was <25% complete, with no complete elements. Age at Death Age was determined using standard ageing techniques, where possible. The extent of skeletal development identified all three individuals as adult at the time of death (i.e. 20+ years, epiphyses have fully fused ends of the long bones). As the preservation of the postcranial elements were less than ideal, more precise estimations of age relied on the teeth, specifically, the eruption of the third molars (wisdom teeth), and the degree of dental wear to the molars (Brothwell 98). SK79 and 250 each had estimated ages at death of between 26 years and 35 years, based on the amount of wear on the molars present for analysis. It is noted that the right mandibular molar of SK79 had also erupted (the wisdom teeth were not present in SK250). As postcranial remains comprised the recovered skeleton of SK82 (left ilium, femur, tibia, talus and calcaneus), the very broad age category of "adult, 20+ years old" could only be suggested. For SK64, likewise, only the very broad age estimate of adult could be determined based on the completed fusion of the left distal femur to the shaft. 9

12 Sex Where possible, the morphologies of the cranium, mandible (lower jaw) and pelvis were analysed for their sexually dimorphic characteristics, based on the suggestions found in Buikstra and Ubelaker (994). Metric analyses of the diameters of the femoral heads were also performed (Stewart 979), however the femoral heads present all displayed varying degrees of damage. The overall robusticity, or lack there of, of the skeletal elements was also recorded. If sex could not be clearly contributed to male or female, the title of indeterminate sex ( I ) was given. The poor preservation inhibited sufficient analysis of the elements present for SK79, and resulted in a designation of indeterminate sex. The left leg of SK82 was notably robust, with strong muscle markings, suggestive of a male individual. The morphologies of the cranium, mandible and pelvis of SK250 indicated female sex. The sex of the SK64 was only suggested by small portions of the greater sciatic notches located on the left and right ilia, which resulted in a probable male designation. Metric and Non-metric Analyses The poor preservation and subsequent high degree of fragmentation inhibited metric analysis of the long bone maximum lengths (i.e. stature estimation), as well as examination of non-metric traits present. Conclusion Osteological analysis of the skeletal remains recovered from Kilnwood, Harwell, has established the assemblage contains all adult individuals, with two of the three individuals aged approximately years. One female, one male, one probable male, and one individual of indeterminate sex were identified. The graves themselves were uniformly orientated NW(head) to SE (feet), and roughly rectangular in shape. Three of the bodies in the graves were supine and extended (SK64, 79 and 250), while the left leg, SK82, was in a crouched position. Grave goods (including a range of metal objects and beads) were also found in association with the bodies. The generally less than ideal preservation hindered much of the demographic skeletal analysis, and the pathological examination. Health Status Pathological alterations are recorded by the human skeleton most commonly when the conditions are chronic, or are the result of a traumatic event. Observations of disease and trauma can give essential insights into the health status, living conditions and activities of the individuals living (and dying) in the Harwell region in the past. The fair to poor preservation of the skeletal remains themselves, hindered a great deal of pathological analysis. Two types of abnormalities were able to be identified, which are described in the general categories of congenital or developmental abnormality and evidence of antemortem trauma. 0

13 Congenital and Developmental Abnormalities Congenital and developmental abnormalities are present from the time of birth. They may be hereditary, or can result from a number of factors, including genetics (e.g. inbreeding), poor maternal health (e.g. virus infection, poor diet), environmental pollution or trauma in the womb. Most congenital anomalies do not affect the person in any way. The back of the cranium of SK250 (26-35 year old female) displayed an unusual appearance. The superior half of the occipital bone is symmetrically unfused along the transverse midline, just inferior to the transverse sulcus (Appendix C). The unfused edge is smooth and completely remodelled (no evidence of active healing, i.e. active bone formation). The only exception to this is located inferior to the internal occipital protuberance, which has a small area of "normal" bone thickness with exposed diploë. Unfortunately the inferior portion of occipital bone is too highly fragmented for assessment. Two different conditions (aetiologies) could be responsible for such an observation. Differential diagnoses include antemortem trauma (e.g. trepanation or weapon wound) and developmental/congenital defect (e.g. retained mendosa suture). Supposing the unfused appearance of this portion of occipital bone was the result of antemortem trauma, given the extent of the bone affected and nearly complete separation of the superior and inferior portions of the occipital bone and possible damage to the underlying anatomical soft tissues, it is unlikely the individual would have survived long enough for such extensive healing to occur. Given the non-fusion is symmetrical, it is most likely this observation is the result of a retained foetal mendosa suture. This suture "separates the developing membranous interparietal and cartilaginous supraoccipital portions of the occipital bone during development" (Barnes 2008, 333), and is "programmed to disappear as the two bones fuse into one" (Barnes 2008, 333). As a result, an Inca bone is formed (the name given to the unfused superior half of the occipital bone). Antemortem Trauma Two elements of the left leg of SK82 (20+ years, male) displayed evidence of muscle trauma, that occurred a long time before the individual's death. The posterior surface of the proximal end of the tibial shaft displays a rugged and raised ridge along the attachment for the Soleus muscle (Appendix B). The bony projections have the appearance similar to osteophytic development, and are curved and claw-like in places. Measuring 4.8mm at thickest part, the projection of bone extends a total of 58.3mm across the posterior surface of the tibia, and projects approximately 3.2mm off the shaft's surface, at the most projecting point. The pronounced and rugged appearance of the soleal line is unusual, and likely suggests antemortem muscle trauma. The Soleus muscle is responsible for plantar flexion of the foot (Stone and Stone 2002, 92). It is possible

14 repetitive movement of the foot, or an episode of forced flexion of the foot strained the Soleus muscle attachment on the proximal tibia, causing the osteophyte-like enthesis. A second, and more unusual, linear bony projection is present on the anterior surface of the left femur of SK82, located approximately half way down the shaft length, running superior to inferior down the middle of the surface (Appendix B). The projection of bone has a smooth and pinched appearance (mediallateral), measures 28.9mm long by 7.0mm wide, and projects approximately 2.6mm off the shaft's surface. This location on the anterior surface of the femur is unexpected to display an enthesis, as it is not an anchoring place for muscles to attach. Given the unusual location, the aetiology of this occurrence is unknown. Differential diagnoses include antemortem trauma to a fibrous attachment of a Quadriceps muscle, or possibly the site of a healed greenstick fracture (a partially broken bone that occurred during childhood). Nevertheless, despite the poor preservation of Sk64, a well healed oblique fracture is present on the distal third of the right femoral shaft. The shaft is bulbous at the fracture point, although the area has been fully remodelled. There was no evidence that this antemortem fracture hindered the mobility of the individual, as no difference was identified in the overall size and muscle attachments of the femoral or tibial shafts. Dental Health Teeth have the ability to provide information on more than oral hygiene. Dental disease reflects the quality of the diet and the enamel on tooth crowns can record periods of illness or malnutrition during childhood. Dentition was recovered from only two of the Kilnwood individuals (SK79 and SK250). The teeth themselves were generally fragile and chalky in texture, resulting in the post-mortem chipping and exfoliation of the enamel layer from several molars. However, the preservation was such that some dental observations were able to be made. Unfortunately, the small sample size was too small to draw any conclusions to patterns of disease expression. Of the 35 tooth sockets present, a total of 36 teeth were present for analysis (including one loose left mandibular first molar from SK250). All teeth were permanent (adult) dentition, with formed an assemblage comprising just 4 maxillary and 22 mandibular teeth. During analysis, no observations were present of antemortem tooth loss, or some common dental diseases (i.e. caries or abscesses). Evidence of calculus (dental plaque), periodontal disease, linear enamel hypoplasia, episodes of trauma and unusual patterns of dental wear were all recorded. Calculus 2

15 Calculus is the mineralization of dental plaque that collects on the surfaces of the tooth crowns. Daily brushing of teeth removes plaque before it can calcify. The presence of calculus indicates that dental hygiene was not a daily priority. Small patches of plaque, of "slight" severity, were observed on the dentition of SK79 (following criteria by Brothwell 98). Given the sparse patches of plaque, the poor preservation of the enamel in both SK79 and 250, the soil conditions in the burial environment were likely not conducive to the preservation of the calculus deposits. Periodontal Disease Periodontal disease is the chronic inflammation of the gums that results in the gradual destruction (resorption) of the alveolar bone of the maxilla and/or mandible. It can ultimately result in antemortem tooth loss. Both SK79 and 250 displayed "slight" resorption of the alveolar processes (following criteria by Brothwell 98). Linear Enamel Hypoplasia Enamel hypoplasia are linear defects of the enamel that are formed in response to nutritional deficiency and/or pathological events (e.g., acute illness, high fever) during childhood. While the crowns are developing within the jaws (under the age of 7 years), these episodes of disease or malnutrition cause a cessation of growth, resulting in deficiencies of the enamel. These defects can take the form of grooves or pits, broad bands of hypomineralization or honeycombed beds of cup-shaped enamel voids (Langsjoen 998). Faint linear defects were recorded on the dentition of both SK79 and 250 (see photographs in Appendix A and C). The mandibular canines and first premolars were affected in SK79, and the maxillary incisors and canines of SK250 displayed these defects. Trauma Several examples of post-mortem trauma were present on the molars of SK79 and SK250, likely the result of the fragile nature of the tooth crown enamel. In contrast, episodes of antemortem trauma were observed on two of the mandibular incisors (the right central and left lateral incisors) of SK79. Small nicks are located on the occlusal surfaces of the incisors. The nick on the right central incisor is located mid-way along the anterior edge of the crown. Two infarctions are present on the left lateral incisor, which are located on the mesial and distal edges of occlusal surface, spanning the entire thickness of the crown. Dental Wear Unusual Patterns of dental wear were present on the dentition of both individuals. SK79 displayed uneven dental wear on the mandibular molars, with the left side more severe than right. Severe wear was also present on the central incisors. A third location of unusual wear was located on the right canine, which took the form of a semicircular notch worn into the distal half of the tooth crown (see photograph in Appendix A). 3

16 Unusually severe dental wear was present on the maxillary incisors and canines of SK250, in addition to all of the first molars present. Such unusual patterns of dental wear (other than those suspected to be related to the normal ageing process), may reflect habitual use of the teeth as tools (i.e. use of the mouth as a third hand). This is primarily suggested by the semicircular notch in the right canine of SK79. Heath Status Conclusion Preservation inhibited a thorough analysis of the remains (skeletal and dental). The skeletons themselves displayed the probable developmental defect of a retained mendosa suture (and subsequent formation of an Inca bone) on the occiput of SK250, and antemortem muscular trauma of the left leg of SK82. Despite the relative small number of teeth present for analysis, evidence of health (dental and childhood) and habitual activities were identifiable. The dentition displayed some evidence of calculus and periodontal disease, which suggest that oral hygiene was not a daily task. Episodes of childhood stress (nutritional or disease) were recorded by the tooth crowns, and unusual patterns of wear, primarily to the anterior dentition of both individuals, and the right mandibular canine of SK79 suggest the teeth were likely used in habitual activities, possibly related to tasks such as occupation. Animal Bone by Lizzi Lewins A small assemblage of animal bone (94 fragments), weighing a total of 225g, was recovered during the course of the excavation. The bone was in good condition, although fragmentary, with little surface abrasion or erosion noted. The bone was classified according to size (large mammal-cattle, horse; medium-sized mammal-sheep/goat, pig, deer) and where possible by species. Schmid's 972 text was used to confirm identification when necessary. A full inventory of the animal bone by context can be found in Appendix 3, only the identified bone will be discussed here. Details of individual elements identified are in the archive. Deposit (83) contained a right un-fused cattle calcaneus which was noted to have a possible cut mark across the shaft. Ditch 205 contained a fragment of large mammal long bone shaft that had been sliced. Given the small number of duplicated skeletal elements the minimum number of individuals was found to be horse; 3 cattle; pig and sheep/goat. The assemblage is likely to represent small scale domestic consumption, however, the presence of butchery marks suggests that a small amount of processing may have taken place within the bounds of the site. Other than the butchery marks and some burning no further taphonomic processes were identified. 4

17 Flint by Steve Ford Just four stuck flints were recovered from the site during the evaluation and excavation (Appendix 4). One was an oblique blunted microlith which had lost its tip. It is of Mesolithic date. It was recovered from the backfill of the presumed Saxon grave in Trench 3 (5, 63) and is clearly residual. The other pieces comprise a broken blade segment (from ditch 32 (93), a poorly made end scraper on a cortical flake (3,90) and a spall (piece less than 20x20mm) (09, 59) Whilst these finds are of very slight interest for this site, a much larger collection of Mesolithic flintwork was recovered from the excavated site to the south of Blenheim Hill, though lacking distinctive microliths (Taylor in prep). The microlith and blade segment for certain therefore, enhance the evidence for the presence of a Mesolithic site on the more southerly site. Ceramic Building Material by Andy Taylor A single piece of tile, possibly floor tile, was recovered from slot 3 in ditch 205 (Appendix 5). This was a red/orange colour and measured 20mm thick and weighed 204g. Two tiny slivers of probable tile came form ditch 20 slot 26. Fired Clay by Andy Taylor Fourteen pieces of fired clay were recovered during the excavation weighing a total of 59g (Appendix 6). most of which were small undiagnostic pieces. One piece of daub was recovered from slot 7 in ditch 20. This was a cream/buff colour, weighed 53g and had evidence of leaf impressions on its interior surface. Metalwork by Steve Crabb A total of 7 objects were recovered from this site, of which 4 are ferrous objects with a single copper alloy object recovered and two silver rings (Appendix 7). To this can be added an iron blade and a nail from the evaluation. The assemblage can be split into two groups those objects recovered from grave contexts and those recovered from non grave context. The objects recovered from graves are mostly grave goods rather than grave fittings. The objects recovered from outside these contexts are all nails or fittings. 5

18 Objects from grave contexts: Cat. No. 4 is two objects recovered from grave 24. First of these is a knife with a flat back and blade with both angled to meet at the tip. The blade is slightly concave at the heel end of the blade indicating sharpening rather than wear of the blade. It measures 76mm long with the blade 6mm long and the tang 60mm long, the blade is 27mm wide. Also recovered from this grave is a large flat pin with a tear drop shaped flat head. Also recovered from this grave was Cat. No. 5 a nail and Cat. No. 5 which is a small copper alloy buckle. This is made from a single plate folded around a cast loop to form the buckle. The plate is slightly damaged on one side but on the complete side the decoration and fitting points of the buckle are present. Two holes were pierced into the sheet, with the edges of these visible on the damaged side. The back edge of the plate has a sawtooth pattern created by a series of sawn cuts. The pin of the buckle is present but separate from the body of the buckle. Cat. No. 6 was recovered from grave 25, which is a right angled hook. Grave 38 had two objects recovered from its fill, Cat. Nos. 3 and 4 which are fragments of the same small knife. As a whole object it measures 7mm long with the tang 23mm long. It has an angled back and straight blade which the back angles down to meet at the tip. Silver Rings Two silver rings (Cat. No. 6) were recovered from the environmental sample 22 associated with grave 38 (from around the skull of Sk250). The two rings are both fragmentary, this is due to the silver becoming brittle in the burial environment. The first ring is a loop of silver wire it is roughly circular in cross section and while not intact it would have been a complete loop with only a small fragment missing. The largest section is just under half of the full ring it measures 9mm across and just under mm thick. The second ring is a flat ring of silver gilt, it has a domed cross section formed of flat silver sheet which has been coated with a thin layer of gold on the outer surface. This ring is fragmented into six pieces, the largest being 2mm long, all the fragments are a uniform 4.7mm across and the sheet it has been made from is 0.35mm thick at its widest. There is evidence for the outer edges having a scored line of decoration which corresponds to the flange. It is difficult to determine the original size of this ring because it is so fragmented, however the largest fragment has the same curve profile as some of the fragments of wire ring, it is highly likely that both rings were about the same size. The similarity of size of rings suggests that they made have been made specifically for this individual. 6

19 Objects from outside grave contexts Cat. No. is a section of plate recovered from pit 05, it is not clear what this plate would have been used for as it is not a complete piece of plate. Cat. No. 2 is a nail recovered from gully 4. Cat. No. 3 is a hook or fitting from ditch 23 Cat. Nos. 7, 8 and 9 are nails recovered from fill 83 of ditch 26. Cat. Nos. 0 and 2 are nails recovered from fill 84 of ditch 26. Cat. No. is a tool with a square section shaft leading to an increased width head. It appears to be an awl. It was also recovered from fill 84 of ditch 26. The objects from this site are fairly typical for a site of this type with a small assemblage of general finds such as nails and hooks as well as personal items such as knives and the buckle recovered from the grave contexts. Glass Beads by Steven Crabb A total of six glass beads were recovered from grave 38 (Appendix 8). Three of these beads are blue, one green and one red. Cat. No. is a transparent blue glass globular bead formed by winding glass around a central core, therefore it has a central hole which is straight sided and even on either side. It measures 20mm in diameter and 8mm thick, the central hole is 8mm in diameter. Cat. No. 2 is an opaque green glass cylindrical bead formed by winding glass around a central core, it is slightly uneven with one side being slightly wider than the other. It measures 8mm in diameter and between 4.7 and 4mm thick, the central hole is 4mm in diameter. Cat. No. 3 is a clear blue glass globular bead formed by pressing a hole through the centre of a small globule, this is shown by the uneven sides of the bead one being pushed inwards and one more curved. It measures 0mm in diameter, 7mm thick and the central hole is 2.5mm in diameter. Cat. No. 4 is a clear blue glass globular bead formed by winding glass around a central core. The sides are slightly uneven showing it was made on a flat surface. It measures 2mm in diameter and 6mm thick, the central hole measures 3mm in diameter. 7

20 Cat. No. 5 is an opaque red glass cylindrical bead formed by winding glass around a central core. The sides are flat but slightly uneven. It measures 9mm in diameter, 6mm thick and has a central hole which is 3.5mm in diameter. As they have been found in one grave it can be safely assumed that these beads are part of the same necklace. These beads have some similarities to each other but each has been made using different techniques or using different materials. The three blue beads are all globular in form, two of these are wound formed (Cat. Nos. and 3) and the other being formed by piercing a droplet of glass to form the central hole. The remaining two beads have been made using the same method, they are wound cylindrical beads which have been shaped to give them flattened sides. They are however made of different colour opaque glass, Cat. No. 2 is an opaque green glass bead and Cat. No. 5 is an opaque red bead. Work into the origins of Anglo Saxon glass, especially that which was used for glass beads, suggests that most of the glass used in this period was recycled (Peake 203; Sainsbury, Pers Comm). The main source of coloured glass would appear to have been the tessellated pavements and mosaics of Roman high status buildings. This has been ascertained by chemical analysis of the glass showing the presence of elements whose only role would be to add or remove colour but the glass is not the colour it should be. Environmental sampling by Rosalind McKenna A programme of soil sampling was implemented during the excavation, which included the collection of soil samples from 5 sealed contexts, which were floated and wet-sieved using a 0.25mm mesh. Methodological details are in the archive. The full species list appears at Appendix 9. Charred plant macrofossils were present within two of the samples, but only indeterminate cereal grains. Identifiable charcoal was present in only two of the samples, and all was of oak. Isotope analysis TBA Radiocarbon dating Two samples of teeth form the graves were submitted to the Chrono radiocarbon dating laboratory at the Queen s University of Belfast. The results were calibrated using Calib rev 7.0 with data from INTCAL 3 (Reimer et al. 203) and are detailed in Appendix 0. All results are quoted at 2-sigma (95.4% probability). 8

21 Conclusion The excavation revealed a modest amount of archaeology on site mostly consisting of two probable enclosures of late Roman date along with later Saxon burials. A single sherd of Bronze Age pottery and some struck flint of Mesolithic date points to some limited activity from these periods in the vicinity Roman evidence on site consisted of pits and enclosures, which in themselves are a fairly common archaeological feature and as no structural evidence of occupation was revealed it is likely that the enclosures were used for stock rather than habitation or industrial activity. As these features continued outside the excavation area it is uncertain as to the full extent of the enclosures and it is unclear if these relate to the site excavated over the road to the south (Taylor forthcoming) or to the known geophysical anomalies immediately to the east. Both enclosures can be dated to the later Roman period (the first is certainly no earlier than mid 3rd century and may be later, while the re-definition and enlargement belongs to the 4th century and probably the second half of that century). Saxon deposits in the form of four inhumation burials were also evident. The consistent alignment of the graves may indicate that the burials were placed with an area designated as a cemetery, re-using the late Roman enclosure (a single sherd of pottery in ditch 202 suggests that ditch might still have been open in early Saxon times, though it is difficult to see it still being in place when the latest burial was laid out in the 9th or 0th century). The radiocarbon dating for two of these burials gave dates of between and , which may show a small family plot utilized over a long period of time (there is a minimum separation of a century between the two dated burials). The grave goods are, in the main, fairly typical for the early or middle Saxon periods, personal knives being almost ubiquitous amongst furnished graves. However while the glass bead necklace would also be standard for female burials in the early and middle Saxon periods, the 9th- to 0th-century radiocarbon date is unexpected, as most graves by this time were unfurnished. At the same time, the two silver rings with the same skeleton may hint at higher status. References Barnes, E, 2008, Congenital anomalies, in R Pinhasi and S Mays (eds), Advances in Human Palaeopathology, Chichester, BGS, 206, Geoindex Onshore, submit=open+the+onshore+geoindex [accessed 3rd August 206] Booth, P, Dodd, A, Robinson, M and Smith, A, 2007, The Thames Through Time: The Archaeology of the Gravel Terraces of the Upper and Middle Thames: The early historical period AD 000, Oxford Archaeology Thames Valley Landscapes Monogr 27, Oxford Brickley, M and McKinley, J (eds), 2004, Guidelines to the Standards for Recording Human Remains, IFA Pap 7, Reading 9

22 Brothwell, D R, 98, Digging up Bones: The Excavation, Treatment and Study of Human Skeletal Remains, 3rd edn, Oxford Buikstra, J E and Ubelaker, D H, 994, Standards for data collection from human skeletal remains Arkansas Archaeological Survey Research Series, 44, Fayetteville, Ark. Coddington, H, 206, Land at Kilnwood, Blenheim Hill, Harwell, Design Brief for Archaeological Excavation, Oxfordshire County Archaeological Service, Oxford Dils, J and Yates, M, 203, An Historical Atlas of Berkshire, Berkshire Record Society, Eynsham Hey, G and Hind, J, 204, Solent-Thames Research Framework for the Historic Environment: Resource Assessments and Research Agendas, Oxford Wessex Monogr 6, Oxford Gray Jones, A, 2003, Chilton, Rutherford Appleton Laboratory, South Midlands Archaeol 33, 6 Gray Jones, A, 2003, Chilton, Blue Car Park Extension, Rutherford Appleton Laboratory, South Midlands Archaeol 35, 48 9 Langsjoen, O, 998, Diseases of the dentition, in A C Aufderheide and C Rodriguez-Martin, The Cambridge Encyclopedia of Human Paleopathology, Cambridge, Lyne, M, 202 The Late Iron Age and Roman Black Burnished Ware Pottery, in L Ladle, Excavations at Bestwall Quarry, Wareham Volume 2: The Iron Age and Later Landscape, Dorset Nat Hist Archaeol Soc Monogr 20, McNicoll-Norbury, J and Taylor, A, 206, Land at Blenheim Hill, Harwell, Oxfordshire, an archaeological evaluation, Thames Valley Archaeological Services unpubl rep 4/96b, Reading NPPF, 202, National Planning Policy Framework, Dept Communities and Local Government, London Orton, C J, 975 Quantitative pottery studies, some progress, problems and prospects, Science and Archaeology 6, 30-5 Pine, J and Preston, S 205, An Iron Age Roundhouse and Roman villa at Chilton Fields, Oxfordshire, Thames Valley Archaeological Services Monogr 20, Reading Schmid, E, 972, Atlas of Animal Bones, London Stewart, T, 979, Essentials of Forensic Anthropology, Especially as Developed in the United States, Springfield, Ill Stone, R J and Stone, J A, 2002, Atlas of Skeletal Muscles (4th Edition), London Taylor, A, 206, Kilnwood, Blenheim Hill, Harwell, Oxfordshire, an archaeological evaluation, Thames Valley Archaeological Services unpubl rep 6/06, Reading Taylor, A, forthcoming, Land at Blenheim Hill, Harwell, Oxfordshire, an archaeological excavation, Thames Valley Archaeological Services unpubl rep 4/96c, Reading Young, C J, 977, Oxfordshire Roman Pottery, BAR Brit Ser 43, Oxford 20