SECTION 5.4: SOMERFORD KEYNES NEIGH BRIDGE ENVIRONMENTAL REPORTS5.4.1 ANIMAL BONES (Emma-Jayne Evans) 5.4.2 CHARRED PLANT REMAINS AND MOLLUSCS (Mark Robinson) 5.4.1 ANIMAL BONES by Emma-Jayne EvansThis report encompasses animal bones from the site at Somerford Keynes. A total of 6282 fragments (77048g) of animal bone and teeth were recovered from the site (Table 1). MethodologyIdentification of the bone was undertaken at Oxford Archaeology with access to the reference collection and published guides. All the animal remains were counted, and where possible identified to species, element, side and zone (Serjeantson 1996). Also, fusion data, butchery marks, gnawing and burning were noted. Undiagnostic bones were recorded as small (small mammal size), medium (sheep size) or large (cattle size). The separation of sheep and goat bones was done using the criteria of Boessneck (1969) and Prummel and Frisch (1986), in addition to the use of the reference material housed at OA. Where distinctions could not be made, the bone was recorded as sheep/goat (s/g). The condition of the bone was graded using the criteria stipulated by Lyman (1996). Grade 0 being the best preserved bone and grade 5 indicating that the bone had suffered such structural and attritional damage as to make it unrecognisable. Tooth eruption and wear stages were measured using a combination of Halstead (1985) and Grant (1982), and fusion data was analysed according to Silver (1969). Measurements were taken according to the methods of von den Driesch (1976) and withers heights were calculated using Fock (1966), Harcourt (1974) and Kieserwalter (Boessneck and von den Driesch 1974, 334). ResultsThe animal remains excavated were in good condition, with the majority of the bone according to Lyman’s grading being between 1 and 2. However, the fragmentary nature of the remains only allowed for the identification of 1639 bones and teeth to species, 26.1% of the total remains present. A list of all the species identified is shown in Table 1. Late Iron Age/RomanA total of 2281 bone and teeth fragments were dated to the Late Iron Age/Roman period from the site, 662 (29%) of which were identified to species. Cattle Cattle were the best-represented species with regards to the number of fragments identified, giving a minimum number of 7. Age at death using tooth eruption and wear stages was calculated, and although the mandible was the best-represented element, fragmentation and post-mortem tooth loss resulted in only 11 mandibles being used. The age at death suggests that a there is a slight majority (54.5%) of cattle dying within the age range 8–30 months, with the larger proportion aged 18–30 months, but only slightly less were aged as adult or old adult. However, the fusion data suggests that, although cattle were being killed before reaching maturity, a larger proportion were surviving to maturity. Unfortunately, fragmentation has meant that many of the bones were missing the articular ends, so could not be assessed for fusion, which may have compromised the results obtained from this analysis. Chop and cut marks were noted on many of the bones indicating that cattle were processed, probably for the consumption of meat and marrow utilisation. However, the skeletal elements represented do not show a marked preponderance of meat bearing limbs, suggesting the cattle remains are the remains of whole animals. The presence of articulating cattle remains at the site further substantiates this, and the cattle present do not mainly constitute butchery or kitchen waste. Measurements were taken on many of the bones, but fragmentation only allowed for the withers height of one animal to be calculated, giving a height of 1.10 metres. Pathologies were noted on two cattle bones, namely a 1st and 2nd phalanx, exhibiting expansion of the proximal articulations. Sheep/goat The number of sheep/goat bone and teeth fragments recovered, though less than the cattle remains, gave the higher minimum number of individuals of 9. Although it is likely that the remains present are those of sheep, the undiagnostic remains have been recorded as sheep/goat. Four elements have been positively identified as sheep, but due to this small number both sheep and sheep/goat remains have been analysed together. As with cattle, the mandible was the best represented element, with 20 mandibles being used for calculating age at death from tooth eruption and wear stages. This method suggests that there is a slight majority of animals dying as juveniles with a peak in the age at death at 20-34 months, although 35% were surviving to maturity. The fusion data indicates that a much higher proportion of the population were dying before reaching 3˝ years old Cut marks were found on a number of sheep/goat bones, many of which were in positions often associated with dismemberment (Binford 1981). A number of long bones had also been chopped, usually an indicator of marrow extraction, and two skulls had cut marks around the horn core associated with skinning. Anatomical distribution again indicates the remains are of whole animals. Although several sheep/goat bones could be measured, no complete long bones were present in which to estimate withers heights. Pathologies were noted on three sheep/goat mandibles. One mandible had an expanded area of bone around the 3rd and 4th premolar, and a gap where the 1st molar is missing. There is porosity of the bone where the jaw is expanded, and the 3rd and 4th premolar have begun to cross over. A second mandible had a swelling of the lateral aspect around the area of the missing 3rd molar., resulting in the expansion of the root socket. This caused a mal-occlusion, as seen by the unusual wear pattern of the remaining teeth. The third pathological mandible has evidence of an abscess, with swelling of the lateral aspect around the premolars, and pitting of swollen area. This has resulted in ante-mortem tooth loss and bone remodelling of root socket. Horse There were 158 fragments of horse found, only slightly less than the number of sheep/goat bones. However, the minimum number of individuals calculated was lower, giving a total of 3. The age at death using fusion data suggests that the majority of animals were surviving to maturity, with only 11.1% of horses dying before 4 years of age. Cut marks were found on only a small number of bones, an indication that horses were not kept primarily for meat. Several bones were seen to articulate, which, along with the anatomical distribution would suggest that like cattle, the remains are representative of whole animals. Osteophytic lipping of two articulating thoracic vertebra may suggest the use of horses for traction or riding. The horse bones were present in a more complete state than the cattle or sheep/goat, indicating that they were less likely to have been processed. This allowed for withers heights to be calculated on five long bones, giving a range in height from 1.15m to 1.46m, with an average of 1.33m Pig Pig remains were present in much lower quantities than cattle or sheep/goat giving a minimum number of 2, suggesting that pork did not contribute as much to the diet of the local population. The age at death from both tooth eruption and wear stages and fusion data suggests that the pigs present at the site were killed before reaching maturity. Cut marks were noted on several bones, indicating that pigs had been dismembered and processed, probably for consumption as pigs provide very little secondary products and are exploited mainly for their meat. No articulating bones were present, and the anatomical distribution, although based on a small number of fragments, suggests a preponderance of meat bearing limbs, suggesting that the pig remains are those of butchery waste. Dog The dog remains from the site suggest a minimum number of 3. Fusion of the bones suggests that the remains are from adult dogs, with one apparent dog burial in context 210. No pathologies or cut marks were noted on any of the dog bones, but withers heights could be calculated on three bones, giving heights of 0.41m and 0.42m. Domestic fowl The domestic fowl present is only represented by one bone. It is likely that the fragmentary nature of the remains has lead to a lack of small bones being recovered, rather than the actual absence of birds and smaller animals at the site. Roe deer As with the domestic fowl, roe deer is only represented by one bone. However, unlike the birds and small mammals, roe deer bones are of similar size and robustness as sheep/goat bones. Therefore if they had made a greater contribution to the diet of the local population it would be expected that they would survive in greater numbers. It may then be suggested that the presence of only one roe deer bone meant that these animals paid little contribution to the diet of the local population. RomanA total of 3779 bone and teeth fragments were dated to the Roman period from the site, 936 (24.8%) of which were identified to species. Cattle As with the Late Iron Age/Roman period, cattle were the best-represented species with regards to the number of fragments identified, giving a minimum number of 9. The tooth eruption and wear stages suggest that the majority of cattle were dying at an old age, with a smaller peak of animals dying aged 1-8 months, and less again at 18-30 months. The fusion data also implies that the majority of cattle were surviving to maturity. Again, the fragmentation of many of the bones may have compromised the results obtained from this analysis. Cut marks associated with dismemberment, and the chopping of bones associated with marrow utilisation was observed on a number of bones, indicating the cattle were processed for the consumption of food. The anatomical distribution of elements suggests that the remains found on the site are from whole animals, indicating that the remains are not solely from butchery or kitchen waste. Although measurements could be taken on many of the cattle bones, only two complete long bones could be used to calculate withers heights, which gave measurements of 1.12 m and 1.23 m. Pathologies were noted on several bones, the majority of which suggest the presence of osteoarthritis, seen as porosity and eburnation on articular surfaces. One 1st phalanx also has grooving and expansion of the proximal articulation as well as extensive eburnation. Sheep/goat As with the Late Iron Age/Roman period, even though the number of sheep/goat remains was less than the number of cattle bones, a higher minimum number of individuals was calculated, giving a number of 13. One bone was recorded as sheep, and is discussed with all the sheep/goat bones identified from this period. Age at death using tooth eruption and wear stages suggests that the majority of the population survived to maturity, with a lower percentage dying aged 1 – 3 months and 20 – 34 months. Fusion data also suggests that almost half the population were mature animals, although juvenile bones may have been more susceptible to fragmentation. Many of the long bones were chopped indicating marrow utilisation as well as dismemberment cut marks which suggest the processing of animals for consumption. Anatomical distribution again does not show a preponderance of meat bearing limbs, indicating the remains are of whole animals. Although measurements were taken of the bones when possible, no complete long bones were present to estimate withers heights of the animals. Pathologies were noted on a number of bones, one indicating the presence of arthritis evident as eburnation on a distal humerus. Three mandibles exhibited pathological changes, those associated with tooth decay and possible abscesses. Horse Although the number of horse bone recovered from the Roman period was considerably less than the cattle bones, the minimum number is only slightly less at 7. Fusion data suggests that all the horse present were mature animals. Anatomical distribution suggests the remains are from whole animals. Cut marks were only noted on two bones, suggesting that horses were not kept primarily for meat. No pathologies were noted on any of the horse remains. Pig Pig remains were present in considerably lower numbers than cattle, sheep/goat or horse, giving a minimum number of 3. The calculation of age at death suggests that the majority of animals were killed before reaching maturity, with tooth eruption and wear stages implying that only 14.2% reached adulthood. Dismemberment cut marks were noted on several bones, indicating that the animals had been processed, probably for consumption. Anatomical distribution suggests that the remains are from whole animals. No pathologies were noted on any of the pig remains from the site. Dog The dog remains from the site suggest a minimum number of 1. Fusion data indicates that all the remains are from adult dogs, and although measurements were taken, no complete long bones were present in which to calculate withers heights. No pathologies or cut marks were noted on any of the dog remains. Birds A single bone of both domestic fowl and duck were recovered, suggesting that, although birds were present on the site, they did not provide a great deal of sustenance to the local population. However the remains of birds and small mammals may be underrepresented due to the fragmentary nature of the remains as a whole. Deer The small number of remains of both red deer and roe deer at the site suggests that although the hunting of wild animals played some part in the activities of the local population, they did not contribute a great deal to the diet. One red deer bone showed evidence of being chopped. Toad The presence of one toad bone is likely to be due to the intrusion of a wild animal at the site. DiscussionThe minimum number of individuals obtained from the Late Iron Age/Roman period suggests that sheep/goat were kept in larger numbers than cattle, with pig being kept in relative low numbers. This pattern is also observed through to the Roman period, with all animals increasing in numbers slightly. The only major difference in the two periods is the number of horse, which increases to a number only slightly less than cattle during the Roman period. However, if the total fragment count is to be taken into consideration, cattle were present in greater number in both periods, with sheep/goat present in fewer numbers, and pig present in relatively small numbers. Horse bones are present in higher figures than may be normally expected. During the late Iron Age/Roman period it is suggested that cattle and sheep/goat were of relatively similar importance, with pig generally being present in low numbers (Hambleton 1999). This is reflected in the results seen at this site, with no changes in species representation through to the Roman period, with the exception of the greater increase in horse. The remains of cattle from both the late Iron Age/Roman and Roman periods suggests that although cattle were being killed at the optimum age for meat production, many were surviving to maturity. It may be suggested that although many cattle were killed for meat, as indicated by the butchery marks, many cattle were kept into adult hood, probably for secondary products such as traction, milk and manure. The idea that many cattle were used for traction may be supported by the expansion of the proximal articulations of two phalanges from the late Iron Age/Roman period and one from the Roman period, a condition often associated with the pressures exerted on the limbs during traction, and the presence of osteoarthritis on bones from the Roman period. The withers heights of the cattle from both periods suggest heights normally expected for cattle from this time, which is smaller than cattle today, although this is only based on the measurements of three long bones. Age at death of sheep/goat suggests that there in an increase in the number of adult sheep during the Roman period, with more animals being killed below the age of 34 months during the late Iron Age/Roman period. This pattern is also observed with the fusion data. It may be suggested that whilst during the late Iron Age/Roman period a slight majority of sheep/goat were being killed at the optimum age for meat production, many were being kept into adulthood probably for breeding and wool production. During the Roman period there is an increase in the number of adult bones found, which may indicate a change in the use of sheep/goat from both meat and wool production to primarily wool production. However, the fragmentation of the remains must be taken into account, and the results may be biased due to preservation considerations. At Somerford Keynes, as at Owslebury and Winnall Down, Hampshire (Maltby 1985), horse was well represented during both the Late Iron Age/Roman and Roman periods. Cut marks on the horse bones suggest that the inhabitants at Somerford Keynes may have exploited horses for meat as well as traction. Perhaps the presence of osteoarthritis present on articulating horse thoracic vertebra may be in part caused by pressures exerted on the vertebral column, such as those when horses are used for riding and/or traction. The withers heights calculated generally fall into those expected for both periods, which are roughly equivalent in size to small ponies It is likely that the pig remains at the site are the remains of pigs used for consumption. Butchery marks and age at death of pigs may support this. As pigs can produce large litters outside the usual seasonal cycles followed by cattle and sheep, a plentiful supply of pork is always available, therefore pigs are usually killed prior to full maturation (Dobney, Jaques and Irving 1996). The presence of dogs from both periods may be from dogs used as guard dogs or hunting dogs. There is no evidence that the dogs have been butchered. One dog burial is evident from the late Iron Age/Roman period, an adult dog probably disposed of by the inhabitants on its death. Withers heights were calculated on three bones from the dog burial, giving a height of 0.41 – 0.42 cm. Birds apparently provide very little to the diet of the population from both periods, although they may be underrepresented due to excavation and preservation conditions at the site. It is apparent that deer were only rarely exploited for meat. ConclusionThe information gained from this study does not show any major changes in the activity on the site from the late Iron Age/ Roman period through to the Roman period, with the exception of the slight in crease in the number of horses. It is clear that horses were treated differently from cattle, sheep/goat and pig, with presence of more complete adult horse bones and only a few bones bearing cut marks, suggesting they were kept mainly for reasons other than consumption, such as traction and for riding. The evidence from the main domestic species on its own does not point to the site having been one of high status, with the meat from cattle and sheep/goat coming as much from young animals as older animals that had likely served their purpose for farming for their secondary products. BibliographyBinford, L, 1981, Ancient Men and Modern Myths, Academic Press. Boessneck, J, 1969 Osteological Differences in Sheep (Ovis aries Linné) and Goat (Capra hircus Linné), in D Brothwell and E Higgs (eds) Science in Archaeology, Thames and Hudson, 331-358 Boessneck, J, and von den Driesch, A, 1974 Kritische Anmerkingen zur Widerristhöhenberechnung aus Längenmassen vor und frühgeschichtlicher Tierknochen, Saugetierkdl Mitt 22 (4), 325-48 Dobney, K, Jaques, J and Irving, B 1996 The Animal Bones from Lincoln, in Of Butchers and Breeds, Report on vertebrate remains from various sites in the City of Lincoln, Lincoln Archaeological Studies, No 5 Fock, 1966, Metrische Untersuchungen an Metapodien einiger europäischer Rinderrassen, Dissertation, Munich. Grant, A, 1982 ‘The Use of Tooth Wear as a Guide to the Age of Domestic Ungulates’, in B Wilson et al. Ageing and Sexing Animal Bones from Archaeological Sites, BAR British Series 109, 91-108, Oxford Halstead, P, 1985 A Study of Mandibular Teeth from Romano-British Contexts at Maxey, in F Pryor, Archaeology and Environment in the Lower Welland Valley, 1.219-282 Hambleton, E, 1999 Animal Husbandry Regimes in Iron Age Britain, BAR British Series 282 Harcourt, 1974 The Dog in Prehistoric and Early Historic Britain, Journal of Archaeological Science, Vol. 1, No. 2 Lyman, R L, 1996 Vertebrate Taphonomy, Cambridge Manuals in Archaeology, Cambridge University Press, Cambridge Maltby, J, M, 1985 The Animal Bones, in P J Fasham, D E Farwell and R J B Whinney, The Archaeological Site at Easton Lane, Winchester, Hampshire Field Club and Archaeological Society: Monograph 6 Prummel, W and Frisch, H-J, 1986 A Guide for the distinction of species, sex and body size in bones of sheep and goat, Journal of Archaeological Science XIII., 567–77 Serjeantson, D, 1996 The Animal Bones, in Refuse and Disposal at Area 16, East Runnymead: Runnymead Bridge Research Excavations, Vol. 2, (eds) E S Needham and T Spence, British Museum Press, London Silver, I.A. 1969, The Ageing of Domestic Animals, in D. Brothwell and E.S. Higgs, Science in Archaeology, Thames and Hudson. Von den Driesch, A, 1976 A Guide to the Measurement of Animal Bones from Archaeological Sites, Peabody Museum
5.4.2 CHARRED PLANT REMAINS AND MOLLUSCS by Mark RobinsonIntroductionIn 1986, the Oxford Archaeological Unit excavated a rural Roman settlement at Somerford Keynes, Glos, in advance of gravel extraction within the Cotswold Water Park. The Roman remains discovered included a corn-drier and boundary ditches. Bulk samples from these contexts were floated to recover charred plant remains. The settlement was situated on a low-lying area of gravel terrace and some of the Roman archaeological features had a thin layer of alluvial sediment in the top. A sample was therefore taken from one of the Roman ditches for molluscan analysis to investigate the possibility of contemporaneous flooding. Methods and resultsTwenty bulk samples, mostly of around 10 litres, were floated onto a 0.25mm mesh to recover charred plant remains. The dried flots were scanned under a binocular microscope. One sample, from a Roman ditch, was found to contain a high concentration of crop processing debris, so was analysed in full. The results are given in Table 1, nomenclature following Clapham et al. (1987). Otherwise charred seeds etc were virtually absent. Nine samples, from important contexts, the corndrier and the cremations, contained charcoal and were analysed. The results are given in Table 2. The mollusc sample, of 1kg, was floated onto a 0.25mm mesh and the residue sieved to 0.5mm for the recovery of shells. The minimum number of individuals of each species present is given in Table 3, nomenclature following Kerney (1999). DiscussionCharred Plant RemainsEven though five samples were taken from different localities within the corn-drier (Context 167, Samples 1, 2, 3, 5, 6), the only charred remains from it other than charcoal noted in the assessment was a single grain of Triticum dicoccum or spelta (emmer or spelt wheat). These samples did, however, contain Quercus sp. (oak) charcoal, some in great quantity. A fragment of Pomoideae sp. (hawthorn, apple etc) charcoal was also identified. When corndriers are used for cereal processing, either for the parching of spelt wheat spikelets prior to de-husking or for malting grain, this usually results in the presence of much processing waste amongst the ashes. In this case, there were copious quantities of charcoal from the oak used to fuel the corndrier but cereal remains were virtually absent. This raises the possibility that the structure was in fact a kiln with another purpose. Five samples, four of which contained charcoal, were investigated from three pits containing burnt deposits. Two pits, 413 (413/A and 413/B) and 416, contained considerable quantities of oak charcoal. Pit 273 had much less charcoal in it, with only a little oak charcoal in Sample 7. Sample 12 from Context 164/H, a section across a Roman ditch, contained many charred crop processing remains, particularly grain and weed seeds. Triticum spelta (spelt wheat) predominated amongst the identified grain and chaff but hulled Hordeum sp. (hulled barley) was also present. While grain comprised 37% of the assemblage, weed seeds made up 60%. The most numerous weed seeds were from Atriplex sp. (orache), Galium aparine (goosegrass) and Carex sp. (sedge). The first two species are common arable weeds that grow on a range of soils. G. aparine is characteristic of autumn-sown crops. Carex spp. are marsh and wet-ground plants that sometimes spread into crops where the cultivated area extends up to marshy ground or has wet flushes in it. The high proportion of weed seeds in the sample suggested that the assemblage represented waste from a late stage of crop cleaning. MolluscsSample 10 from the Roman fill to a ditch (252/A) contained many shells of terrestrial molluscs, particularly Trichia hispida gp., but including species characteristic of dry open conditions, such as Pupilla muscorum and Vallonia excentrica. They probably lived on the general ground surface. There were also examples of the amphibious to slum aquatic molluscs Lymnaea truncatula and Anisus leucostoma that are likely to have lived in puddles of stagnant water in the ditch bottom. There were no shells of flowing water aquatic species as might be introduced by floodwaters and were present in the alluvial sediment in the tops of some of the Roman ditches. Clapham, A R, Tutin, T G and Moore, D M 1987 Flora of the British Isles, 3rd edition. Cambridge: University Press Kerney, M P, 1999 Atlas of the land and freshwater molluscs of Britain and Ireland. Colchester: |