Radiocarbon Dates from the Oxford Ams System: Archaeometry Datelist 31



This thirty-first list of AMS determinations measured at the Oxford Radiocarbon Accelerator Unit (ORAU) is predominantly made up of those measured since the start of 1996. However, it also includes some determinations measured earlier which have not been published in previous lists. The dates presented here include those measured through the NERC-funded Oxford Radiocarbon Dating Service (ORADS), those funded by English Heritage and Historic Scotland, and those submitted to the laboratory on a commercial basis.


All dates have been measured using the procedures outlined in Law and Hedges (1989), Hedges et al. (1989; 1992) and Bronk Ramsey et al. (Arch. List 30). AMS determinations with OxA numbers greater than 6293 were measured either as graphite (if sufficiently large) or as CO2 (Bronk Ramsey and Hedges 1997).

In accordance with international radiocarbon convention all dates are expressed in radiocarbon years before AD 1950 (years BP) using the half-life of 5568 years. Errors are quoted as one standard deviation (1 σ) and are based on an assessment of all the contributions to the error in the laboratory isotope ratio measurement. Natural fractionation of carbon isotopes is accounted for by measuring the δ13C values relative to PDB (with errors of approximately 0.3‰).

All combining procedures and significance tests are based on Ward and Wilson (1978). Comments composed by the Laboratory on the basis of information supplied by submitters are given without attribution.

All calendrical dates quoted, unless stated otherwise, have been calibrated using the OxCal computer program (Bronk Ramsey 1995; 1998a; 1998b; 2001) using atmospheric data from ‘INTCAL98’ (Stuiver and van der Plicht 1998), and are quoted to 95.4% probability.

Previous Archaeometry datelists are referred to in the form: Arch. List 30.

Details of methods used are described fully in our in-house documentation. This is regularly updated and archived, so the exact method used for any sample is always fully recorded.

Reliability of measurements in this list

Over the 7–8 years the measurements in this list cover, much work has been done to ensure the reliability and reproducibility of measurements made at the laboratory. To a large extent this is to do with good laboratory practice and standardisation of all aspects of the dating process, from calibration of equipment to full traceability of key components and reagents used during the dating process. These aspects of ‘Quality Management’ are often best assessed by those outside the laboratory and, to this end, the lab was accredited to the ISO-9002 standard by the BSI in January of 2000.

On a more scientific level, one important test of a radiocarbon dating laboratory is its ability to determine the correct radiocarbon concentration in material whose true age is known. This means the regular measurement of international standards, known-age material and background material (very old material of different types used to estimate sample treatment blanks) and participation in International Radiocarbon Intercomparisons. Another and equally important test is the ability to reproduce measurements on real archaeological material, since this is usually in a poorer state of preservation than known-age material.

Here we summarise the results relating to measurements of known-age material (generally tree rings supplied to us by the Queen's University of Belfast, with known ages taken from the INTCAL98 calibration curve) and duplicate measurements made on unknown samples.

Measurements on known-age material

The results for about 750 measurements on known-age material shown in Table 1 show that error terms quoted on the radiocarbon determinations do reflect very well the true uncertainty. The number falling within one, two and three standard error terms (σ) is very close to the expected proportion (see cf. values in the table header). They also show that any laboratory offset relative to the calibration curve is well under ten years. Overall, the figures suggest a reliability of about 99% (since the average proportion lying within 3σ is 98.93% rather than the theoretical 99.74%). Here we define reliability to be a measure of the probability that a typical sample will give a result in accordance with the quoted error.

Table 1. Showing summary of measurements made on known age material in the years 1995–2001. This shows the average offset in radiocarbon years from the expected radiocarbon age of the known-age material measured and the proportion lying within one, two and three standard errors
YearNumber acceptedNumber rejected1BiasWithin 3σ (cf 99.74)Within 2σ (cf 95.44)Within 1σ (cf 68.26)
  1. 1 Rejection implies that all unknown material from this batch will also have been re-measured before release

  2. 2 At this time it was the practice to repeat batches but not to transfer the original measurements to our database – so this information is not available

  3. 3 The larger offset for this year is largely due to one measurement which was over 4σ away from the expected value; if this is removed the bias for the year drops to 11.06 ± 5.40 and the proportion within 3σ rises to 99.06%; no explanation for this result was found but results in the same batch did not seem to be affected and were not rejected

  4. 4 Data shown for all measurements, including those from rejected batches

199558na28.26 ± 7.47100.00100.0065.52
199639na2−9.27 ± 9.94100.00100.0058.79
19973107na214.50 ± 5.3598.1390.6559.81
199810202.80 ± 4.5298.0497.0669.61
199911114.78 ± 4.5499.1098.2074.77
19994112 7.34 ± 43298.2197.3274.11
200015097.00 ± 3.8398.6795.3365.33
20004159 13.28 ± 3.6993.7189.9461.64
200117901.38 ± 3.0699.4498.3171.51
1995–2001746104.64 ± 1.7398.9396.6567.69

No process is totally reliable, however, and this can be seen in the details of the data in Table 1. There was one significantly aberrant (well over 2σ) measurement in 1997 and another in 1999. However in 2000 there were nine such measurements. Three of these were due to a known batch-specific problem (the results from these batches were not released). Measurements from batches involving the remaining six (about 4% of the total number) were also rejected, but were unexplained at the time. This intermittent problem with graphitisation was eventually diagnosed and eliminated in November 2000. Having already repeated approximately 150 measurements, we estimated that a further 2–3% of the output from 2000 (20–30 determinations) could have been affected. A further set of repeated measurements on samples especially likely to have been affected identified 21 determinations liable to be in error, which have now been withdrawn (see Appendix I). We therefore estimate, from the data on known-age material, that 99% of measurements for this year should now lie within the expected distribution, as for other years (though see next section).

Duplicate measurements on unknown samples

The results for about 250 measurements made in duplicate on unknown samples are shown in Table 2. These show, as one would expect, that duplication on real archaeological samples is harder to achieve than on selected, well-preserved material. These measurements include both randomly duplicated samples (now performed on ∼5% of submitted samples) as well as requested duplicates (either by submitter or laboratory, e.g. because of technical difficulty). Such duplicated samples can be expected to contain a higher proportion of ‘problem samples’, but are a fair reflection of the actual dating process. (For example, not all samples can be said to be ‘single entities’.)

Table 2. Showing summary of duplicate measurements (including duplicate pre-treatment) made on samples in the years 1997–2001; before this period random duplication was not performed
YearNumber acceptedNumber rejected1within 3σ (cf 99.74)within 2σ (cf 95.44)within 1σ (cf 68.26)
  1. 1 Samples rejected from this analysis all have explained reasons for discrepancies (multiple entity charcoals, anomalous stable isotope values, anomalous elemental ratios, incomplete removal of preservatives or deliberate experimental tests).


Taking these data as they are, we can conclude that the reliability of the dating process, which is about 99% for well preserved standard material, is reduced to about 97% for real samples. It is likely that the distribution is not accurately Gaussian and it would be unwise to compensate for this decrease in reliability by increasing the estimated error (by, in this case, around 10%). In any case, the results vary from year to year, and indeed the latest results are now very close to the theoretical limit. Also, the number of samples rejected from this analysis has gradually fallen over the years in question, reflecting the more careful selection of sample material by submitters in collaboration with the laboratory.

Summary of quality control data

Quality control of the dating process is essential if radiocarbon results are to be used as the basis for assessing archaeological chronologies. The known-age and duplicate data presented here support the determinations published in this datelist. These data show that the error terms quoted on the measurements can be relied upon to provide a realistic assessment of the true uncertainty in the radiocarbon determinations. They also demonstrate that there is no significant laboratory bias relative to the radiocarbon dates used to build the radiocarbon calibration curve (INTCAL98) and so the measurements published here should be directly comparable to those produced in other laboratories with similar quality control.


Obviously much work lies behind the dates published in this list. We must first thank the submitters for their care in sample selection, collaboration during the dating process and for providing comments on the dates for this publication. We would also like to acknowledge all the hard work of past and present members of ORAU who have done so much of the work involved in the dating process and who are not in the author list for this paper: Diane Baker, Angela Bowles, John Foreman, Gillian Hanford, Greg Hodgins, Rupert Housley, Martin Humm, Phil Leach, Neil Murphy, Tamsin O'Connell, Vanessa Pashley, Paul Pettitt, Angie Stokes, Christine Tompkins and G.-J. van Klinken. We also acknowledge that little of this work would ever have been attempted without the generous funding of NERC Scientific Services, English Heritage and Historic Scotland.

Human Remains

Great Britain

Oreston Breakwater Quarry

Fragment of a human right clavicle (catalogue number GS 267), identified by Andrew Chamberlain in 1994 at the British Geological Survey, Keyworth. This previously unidentified bone was among a collection of faunal remains retrieved in 1822 from the third bone cave at Oreston Breakwater Quarry, Plymouth, Devon (NGR SX 502538), submitted by A. T. Chamberlain, Dept. Archaeol. & Prehistory, Univ. Sheffield, through the NERC-funded ORADS facility.

OxA-4777human bone, GS 267, δ13C = −20.2 ‰8615 ± 75

Comment (A. T. C): fauna from the cave include a mixture of Pleistocene and Holocene species (Whidbey and Clift 1823; Chamberlain and Ray 1994). The Early Mesolithic date for GS 267 is similar to direct dates on human remains from other coastal caves in southern Britain including Kent's Cavern, Devon (OxA-1786: 8070 ± 90 BP), Worm's Head Cave, Gower (OxA-4024: 8800 ± 80 BP) and Ogof-yr-Ychen, Caldey Island (OxA-7690: 8290 ± 55 BP; OxA-7691: 8210 ± 55BP; OxA-7741: 8415 ± 65 BP). However the specimen from Oreston Breakwater Quarry may have been treated with Isinglass (a collagen-based preservative) and this radiocarbon determination must therefore be regarded as a minimum estimate of the age of the specimen, as the possible inclusion of modern collagen would bias the determination towards a younger date.


Sample from skull from the site of Skirethorns (NGR SD 965645), North Yorks, submitted by S. Kirrane, District Museums Officer, Craven Museum, Skipton, N. Yorks.

OxA-6109human bone, SKIPM: E110, δ13C = −21.4‰2420 ± 65

Comment (S. K.): this skull came to us as a result of superficial survey work at the Skirethorns site undertaken by the Department of Archaeological Sciences at Bradford University in 1994. The site forms part of an interesting plateau which has provided finds from the Mesolithic to the Medieval period. On this particular site there is evidence of what appears to be a Viking farmstead, and also of Iron Age round huts.

The skull was retrieved from a deep crevice in the limestone, at the rear of one of the huts. Its date confirms Iron Age occupation. The fact that it was discovered without any other human bones has raised interesting questions about ritual Celtic ‘head cults’, however opinions are divided on this point.

The fact that the skull was dated to this period added to its interest and it became the focal point for a major exhibition here at the Craven Museum (1997). This contrasted the archaeological interpretation of the site and the skull, with that of a local artist. This exhibition stimulated a great deal of interest and in this small, local way has contributed to public understanding and support for archaeology.

Taywood Homes

Sample of bone from Taywood Homes, Fulbourn (NGR TL 552257), Cambridgeshire, submitted by C. Duhig, 109 Sturton Street, Cambridge, CB1 2QG.

OxA-7452human bone, FULSR97 (1), δ13C= −19.1‰1900 ± 40

Comment (C. D.): this sample was submitted for the Cambridgeshire Constabulary, being a skeleton of suspiciously-modern appearance, found in a context which had no evidence of antiquity and in a location which might suggest that the bones were of forensic interest. They were extremely pleased to have the antiquity of the remains demonstrated.

Shavards Farm

A sample of bone from Shavards Farm, Meonstoke, Hampshire (NGR 61651065), submitted by N. Stoodley, 207 Farley Lane, Braishfield, Romsey, Hampshire, SO51 0QL.

OxA-9030human bone, SHF87, δ13C = −19.2 ‰2275 ± 39

Comment (N. S.): during work on the early Anglo-Saxon cemetery and settlement at Shavards Farm in 1998 and 1999 it was discovered that excavations on the site in the 1980s by Mike Hughes had recovered a crouched inhumation burial. It was suspected that the burial took place in the Bronze Age and a sample of bone was sent for dating in order to confirm this. However, the Middle Iron Age result was not expected and is further evidence, along with an enclosure, for activity at this time.

Southampton High Street

Samples of bone from Southampton High Street (NGR SU 41991108), Hampshire. Submitted by A. D. Russell, Archaeology Unit, Heritage Services, Civic Centre, Southampton.

OxA-5941human bone, SOU 161, 60, 3933, δ13C = −19.99‰1135 ± 60
OxA-5942human tooth, SOU 266, 2197 SF 4800, δ13C = −19.2‰1450 ± 50

Comment (A. D. R.): OxA-5941 and −5942 were from human remains found within the medieval town of Southampton at the lower end of the High Street. OxA-5941 was from remains found in C. Platt's excavation of 1966. Three partial skeletons were found, one cut by the construction of a twelfth to thirteenth century stone house, but they had no independent dating evidence. Further work on the adjacent site in the 1990s produced a human molar from a late Saxon deposit. This raised the possibility that an earlier cemetery had occupied the site before the late Saxon town was founded c. 950 AD. The two dates would appear to provide proof of this, and raise interesting questions about the sites relationship with Hamwic (AD 650–850), situated less than I km to the northeast. The settlement that the cemetery served has yet to be located.

St. Mary's Church Yard

Sample of bone from St. Mary's Churchyard (NGR SU 42651159), Southampton, submitted by A. D. Russell, Archaeology Unit, Heritage Services, Civic Centre, Southampton.

OxA-7187human bone, SOU 753, δ13C = −19.6‰885 ± 60

Comment (A. D. R.): OxA-7187 was obtained from bones of a human foot encountered during repairs to the nineteenth century churchyard wall of St Mary's Church, St Mary Street, Southampton. Although long considered to be the mother church of middle Saxon Hamwic (AD 650–850) there is little real evidence to back up this assumption. This new date fits well with dates from two skeletons found under the nearby pavement a few years ago (OxA-5447 and −5448). These all go to show that bodies were being buried some distance from the church by the ninth century.

Ysgol Twm O'r Nant

Sample of bone fromYsgol Twm O'r Nant, Denbigh, (NGR SJ 05956662) Wales, submitted via the Univ. Bradford, Dept. Archaeol. Sci., by C. Roberts, Dept. Archaeol., Univ. Durham, through the NERC-funded ORADS facility.

OxA-4942human bone, GIIJI, δ13C = −19.7‰515 ± 65

Comment (C. R.): the site was dated archaeologically to pre-Columbus and therefore, with a skeleton with bone changes suggestive of venereal syphilis, we wanted to see whether it was actually pre-Columbian in date. It seems that the dates provided indicate that it may be but may not be. Of course, dates such as these are important for the discussions of the evolution and development of syphilis in the Old and New Worlds.

Hyde Abbey

A sample of bone from Hyde Abbey, Winchester, England (NGR SU 48323013), submitted by H. Rees, Historic Resource Centre, 75 Hyde Street, Winchester, SO23 7DW.

OxA-9290human bone, HA99 2034, δ13C = −18.9‰231 ± 37

Comment (H. R.): although it was a possibility that the date would be before the Norman Conquest, we were not surprised that it was later, as the disarticulated fragment could have been redeposited from any time in the late Saxon or medieval periods. It was quite surprising, however, that the calibrated ranges were in the seventeenth and eighteenth centuries (1520–1570, 5.4%; 1630–1690, 41.1%; 1730–1810, 40.1%; 1920–1950, 8.8%), as the site should have ceased to be used for human burial at the Dissolution in 1538. The date range AD 1520–1570 therefore fits the archaeological and documentary evidence best.


Cerro Virtud de las Herrerias

Three samples of human bone and a fourth of olive wood from Cerro Virtud de las Herrerias, Cuevas de Almanzora, Almeria (37:17N 47:00W), Spain, submitted by R. W. Chapman, Dept. Archaeol., Univ. Reading, through the NERC-funded ORADS facility.

OxA-6580human bone, CV-4, δ13C = −19.6‰5850 ± 80
OxA-6713human bone, CV-1, δ13C = −17.7‰5765 ± 55
OxA-6714human bone, CV-2, δ13C = −18.8‰6030 ± 55
OxA-6715charcoal, olive, CV-3, δ13C = −23.1‰5895 ± 55

Comment (R. W. C): three samples of human bone and a fourth of olive wood charcoal (with some bark attached) were taken from a Middle Neolithic collective burial in a pit, excavated in 1994: OxA-6713 dates a female aged 35–50 years, OxA-6714 an adult 35–45 years and OxA-6580 a juvenile, probably male, aged 14–15 years. The dates form an internally consistent series, dating to the early-mid fifth millenniumcal BC. Taken overall, the samples for Cerro Virtud provide the first reliably contextualised absolute dates for the Neolithic in lowland south-east Spain, extending its duration to a possible fifteen hundred years. Cerro Virtud is the first example of non-megalithic, Neolithic collective burial to be excavated in this region, although there are parallels in other parts of southern Spain (Montero Ruíz et al. 1999). The site has also yielded evidence for copper metallurgy at a similarly early date (Ruíz Taboada and Montero Ruíz 1999).


Sample of bone from Gatas, Turre, Almeria (37:07N 01:53W), Spain, submitted by R. W. Chapman, Dept. Archaeol., Univ. Reading, through the NERC-funded ORADS facility.

OxA-7764human bone, BA28, δ13C = −18.8‰3520 ± 35

Comment (R. W. C): the sample dates the ribs of a juvenile, probably male, aged 14–16 years, and buried in a sealed stone cist, tomb 39, in a stratified context in the Bronze Age settlement of Gatas. According to the stratigraphy and to multiply-dated samples from other burials at Gatas (see Arch. Lists 15, 17 and 19; Castro et al. 1995), tomb 39 dates to phase 4 of the site's occupation, c. 1750–1500 cal BC. Dating was intended to confirm this and to show the late use of stone cists as burial containers. OxA-7764 predates the expected period of time for tomb 39. In the absence of laboratory error, this suggests that either there is a stratigraphic problem, or that the death of the juvenile pre-dated the construction and use of the stone cist.

Calle de Los Tintes

Samples of bone from Calle de los Tintes, Lorca, Murcia (37:40N 01:41W), Spain, submitted by R. W. Chapman, Dept. Archaeol., Univ. Reading, through the NERC-funded ORADS facility.

OxA-7667human bone, BA24, δ13C = −18.9‰3560 ± 35
OxA-7668human bone, BA25, δ13C = −18.9‰3690 ± 40

Comment (R. W. C): the samples date a double burial, Calle de Los Tintes 2, in a pit in a stratified context in the Argaric Bronze Age occupation of the settlement of Lorca: a primary female, aged 35–40 years was interred on the left side and associated with a copper awl, and then succeeded by an adult male, aged 35–45 years, with the same placing and orientation. The pit was then filled with stones and earth and sealed. There are no signs of later intrusions or disturbance. Dating of both burials was intended to confirm the sequence of interment and the timespan between the burials. The dates suggest the interments possibly took place within 100 years of each other, with the female (OxA-7768) pre-dating the male (OxA-7767). As in other examples of such Argaric double burials in south-east Spain (Lull 2000), these dates suggest the practice of burying individuals related by kinship, but at least two or more generations apart from each other.

Madres Mercedarias

Samples of bone from Madres Mercedarias 11, Lorca, Murcia (37:40N 01:41W), Spain, submitted by R. W. Chapman, Dept. Archaeol., Univ. Reading, through the NERC-funded ORADS facility.

OxA-7671human bone, BA20, δ13C = −18.5‰3435 ± 35
OxA-7672human bone, BA21, δ13C = −18.5‰3510 ± 40

Comment (R. W. C): the samples date a double burial in a cist in a stratified context in the Argaric Bronze Age occupation of Lorca: a primary female, aged 30–35 years, associated with a copper awl was succeeded by a male of the same age and then the stone cist was sealed with no evidence of later disturbance or intrusion. Dating of both burials was intended to confirm the sequence of interment and the timespan between burials. Like Calle de Los Tintes 2 (see above), the dates confirm the order of interment (OxA-7672 is the female and OxA-7671 is the male) and the practice of making double burials out of different generations.

Cova d'es Carritx

Cova d'es Carritx, Menorca, Balearic Islands (39:58N 07:38W), Spain, submitted by R. W. Chapman, Dept. Archaeol., Univ. Reading, Whiteknights, Reading, through the NERC-funded ORADS facility.

OxA-10308human bone, 13, δ13C = −18.9‰3072 ± 38
OxA-10309human bone, 20, δ13C = −18.8‰3175 ± 38

Comment (R. W. C): these two dates are in full agreement with previous dates and confirm the interpretation of the ritual practices carried out in the main funerary chamber (Sala 1) between c. 1450/1400 and 800 cal BC (see Lull et al. 1999). OxA-10309 dates one of the first inhumations in the chamber, which has surprisingly been preserved as a semi-articulated skeleton in a context of successive burial and complex bone rearrangement rituals. On the other hand, OxA-10308 dates a specific burial event in the long-standing sequence of collective funerary use of the chamber.

Cava d'es Carritx

Samples from Cova Des Carritx, Menorca (39:58N 07:38W), Spain, submitted by R. W. Chapman, Dept. Archaeol., Univ. Reading, through the NERC-funded ORADS facility.

OxA-8263human hair, 6C, δ13C = −19.7‰2585 ± 40
OxA-6416human hair, 20, δ13C = −22.5‰3165 ± 45
OxA-7235human hair, 20, δ13C = −20.3‰2935 ± 45
OxA-7803human bone, 10, δ13C = −19.1‰2875 ± 40
OxA-7810human bone, 1, δ13C = −19.7‰3100 ± 40
OxA-7811human bone, 7, δ13C = −18.9‰2915 ± 40
OxA-7812human bone, 9, δ13C = −19.7‰2850 ± 40
OxA-7819human bone, 21, δ13C = −18.9‰2965 ± 40
OxA-7820human bone, 24, δ13C = −19.2‰2380 ± 40
OxA-7821human bone, 26, δ13C = −18.8‰3030 ± 40
OxA-7822human bone, 27, δ13C = −19.2‰2680 ± 40
OxA-7823human bone, 29, δ13C = −19.0‰2805 ± 40
OxA-7888human bone, 28, δ13C = −19.6‰2710 ± 75

Comment (R. W. C): the Cova d'es Carritx is an intact burial cave located in a deep gorge in the southwest of Menorca. It was discovered by potholers in 1995 and excavated in 1995–97 (Lull et al. 1999). Two chambers immediately inside the entrance contained disarticulated burials, while placed deposits of artefacts and human bones were found throughout the 120 m long system of chambers and passages. An inner chamber contained a sealed pit with a hoard of organic and inorganic artefacts, including wood and horn tubes filled with cut human hair. An AMS dating programme was undertaken to establish the range for deposition of bone and artefacts in different areas of the cave system. OxA-6416, −7235 and −8263 are samples of hair from the hoard, while the rest of the samples date human bone in chamber 1. Taken together, the dates show that the cave was used as an ossuary from c. 1500–800 cal BC. The latest date for bone deposition (OxA-7820) is from a sample deposited on top of the dry-stone wall blocking the entrance to the cave system.


Prehistoric cave burials

Human bone samples from twenty-six caves located in the Meuse Basin in the province of Namur, submitted by R. Orban, C. Polet and I. Jadin, Royal Belgian Institute of Natural Sciences, and N. Cauwe, Royal Museums of Art and History, Brussels, Belgium. Analyses were undertaken with the financial support of the ‘Fonds National de la Recherche Scientifique’. The radiocarbon determinations presented here are divided into three chronological groups.

Late Palaeolithic and Early Mesolithic

OxA-6856Waulsort, cave “X”, 22-T4, human bone, δ13C = −21.3‰10 820 ± 80
OxA-4917Abri des Autours, tomb 3, human bone, δ13C = −20.2‰9500 ± 75
OxA-5838Abri des Autours, tomb 2, human bone, δ13C = −19.5‰9090 ± 140
OxA-5679Chaleux Cave, IG 2602, 3, human bone, δ13C = −19.9‰8730 ± 80
OxA-5841Magrite Cave, IG 2426, human bone, δ13C = −20.5‰8645 ± 70

Comment (N. C., R. O. and C. P.): several sites in Belgium are known to contain Early Mesolithic human bones. Most of them are collective tombs, ie. graves used over a long period of time and to which the dead were successively brought. The date of the Tomb 2 of the Autours rock-shelter (OxA-5838) confirms the emergence of this kind of funeral practice at the beginning of the Holocene period. Today, the corpus of Mesolithic collective tombs includes at least ten examples, located in Belgium and the southern part of England (Cauwe 1998; in press). But, in southern Belgium, with the exception of the Autours rock-shelter and the Margaux Cave, the only other discoveries known are from ancient or non-scientific excavations which give less explanation about the exact conditions of the discovery.

Nevertheless, these deposits testify that collective burials in natural caves were frequent during the ninth and eighth millennia BP in north-west Europe. Once more the dates of human bones from the Chaleux Cave (OxA-5679) or the Magrite Cave (OxA-5841) indicate the importance of this phenomenon. We suspect a Late Upper Palaeolithic origin for these funeral practices and this impression first received confirmation with the Late Upper Palaeolithic date of the human bones from the Waulsort X Cave (OxA-6856).

Finally, the western Mesolithic is characterised by a great variation in funeral practices. The individual grave of the Autours rock-shelter (OxA-4917) supports this hypothesis.

The Middle Neolithic

OxA-5837Abri des Autours, tomb 1, human bone, δ13C = −21.0‰5300 ± 55
OxA-9022Hastière, cave D, human bone, δ13C = −20.7‰5235 ± 45
OxA-9021Hastière, cave B, human bone, δ13C = −20.5‰5180 ± 45
OxA-9023Waulsort, cave AB, human bone, δ13C = −20.8‰5130 ± 45
OxA-9088Hastière, cave L, human bone, δ13C = −21.1‰5070 ± 60
OxA-5677Reuviau Cave, IG 2585, 14, human bone, δ13C = −20.2‰5025 ± 65
OxA-9089Anseremme, human bone, δ13C = −20.6‰4945 ± 55
OxA-9025Hastière, Maurenne I cave, human tooth, δ13C = −20.1‰4635 ± 45
OxA-5840Waulsort, cave Q, human bone, δ13C = −20.6‰4620 ± 50
OxA-5314Waulsort, cave Q, human bone, δ13C = −21.9‰155 ± 45

Comment (N. C., R. O. and C. P.): these dates on human bones from several Belgian caves clearly show that the Michelsberg culture also had recourse to the same collective tomb as in the megalithic civilizations. The Autours rock-shelter (OxA-5837) is a good example (Cauwe 1994). The samples used for the eight other dates come from old exavations, but establish the importance of a phenomenon until now only suspected (Toussaint and Becker 1992).

Note that OxA-5314 is clearly anomalous and coincides with the time of the excavation at the end of the nineteenth century.

Whatever the quality of these collections, it is clear now that the collective burial is not exclusive to the megalith builders and plays a large part in the Middle Neolithic of Western Europe. Indeed the same tradition occurred at the beginning of the Middle Neolithic in southern France and Catalonia beyond the megalithic world (Cauwe 1999).

The Late Neolithic

OxA-9024Waulsort, cave R, human bone, δ13C = −20.7‰4365 ± 45
OxA-5315Waulsort cave Y, human bone, δ13C = −20.9‰4355 ± 55
OxA-6558Hastière, cave M, human bone, δ13C = −21.0‰4345 ± 60
OxA-6852Hastière, Petite cave, human bone, δ13C = −20.6‰4300 ± 50
OxA-6851Hastiere, cemetery (?), human bone, δ13C = −20.1‰4280 ± 50
OxA-9467Trou Felix, Falmignoul, human bone, δ13C = −20.5‰4260 ± 40
OxA-9466Trou Felix, Falmignoul, human bone, δ13C = −20.7‰4255 ± 45
OxA-6857Waulsort Piret IV, human bone, δ13C = −21.0‰4250 ± 45
OxA-5313Freyr Cave, human bone, δ13C = −21.3‰4240 ± 50
OxA-6252Spy Cave, human bone, δ13C = −21.8‰4230 ± 70
OxA-5839Trou-des-Blaireaux, human bone, δ13C = −20.6‰4230 ± 55
OxA-6853Hastière, Trou Garcon C., human bone, δ13C = −20.5‰4220 ± 45
OxA-6855Waulsort Cave “O”, human bone, δ13C = — 20.0‰4170 ± 45
OxA-5041Furlooz, Trou Rosette, human bone, δ13C = — 20.1‰4165 ± 70
OxA-9026Maurenne 2 Cave, human tooth, δ13C = — 20.5‰4160 ± 45
OxA-6854Hastière, Trou Fanfan K, human bone, δ13C = — 20.4‰4155 ± 50

Comment (N. C., R. O. and C. P.): the dates presented here concern collective burials and belong to the Late Neolithic, and contain no surprises. They can be added to about thirty dates already known for southern Belgium (Toussaint 1992). From the nineteenth century, about 200 collective graves have been discovered in the Meuse basin and some of them contained burial goods from the Seine-Oise-Marne Culture (De Laet 1982).

The majority of the collective burials so far dated belong to the first half of the third millennium BC and take their place in a large tradition including the passage graves of the Parisian basin and those from the Hesse in western Germany (Masset 1997). Detailed comments on the date from Spy can be found in Semal et al. (1996).

Prehistoric collective burials

These samples of human bone originate from presumed prehistoric collective burials discovered in the cave Trou Madame, Bouvignes-sur-Meuse, province of Namur, (40:16N 4:53E), and Goyet Cave on the bank of the River Samson near Namèche, Mozet, province of Namur (50:27N 4:59E), Belgium. They were excavated by E. Dupont in the 1860s and submitted by Rosine Orban, Caroline Polet and Anne Hauzeur, Royal Belgian Institute of Natural Sciences, Brussels, Belgium.

OxA-6118Trou Madame, human bone, δ13C = — 19.7‰1915 ± 40
OxA-5678Goyet Cave, human bone, δ13C = — 19.1‰1985 ± 70

Comment (R. O., C. P. and A. H): OxA-6118 and −5678 reveal that the skeletons belong to the Roman period. This is, at first sight, surprising because most of the burials found in the caves of the Meuse basin are dated to the Neolithic. However, a recent re-examination of the pottery collected during the excavations of Trou Madame has been performed by A. Hauzeur and A. Cahen-Delhaye, Royal Museums of Art and History. It shows that eight potsherds belong to the Bronze and the Iron Age, and another one (dolium) to the second half of the first century AD. Further, the site of Goyet has been attributed to the Upper Palaeolithic on the basis of the industry (Otte 1984) and the fauna (Germonpré 1996). But Tihon (1895–6) points out the presence of Roman pottery which gives support to the dates obtained here and to the possible occurrence of cave inhumations in Belgium during the Roman period.


Samples of human bone from Stavelot, Belgium (50:23N 5:5E), submitted by M. Otte, Préhistoire, Univ. Liège, Belgium.

OxA-5777human bone, 156–8b, δ13C = −19.5‰1035 ± 55
OxA-5778human bone, 156–4, δ13C = −19.2‰1455 ± 50
OxA-5779human bone, 159–8, δ13C = −19.5‰1085 ± 45
OxA-3445human bone, STV91/L7, δ13C = −20.4‰1150 ± 70
OxA-5273human bone, P159/10, δ13C = −18.6‰1285 ± 45

Comment (M. O.): The position of the burial suggested an attribution to the Abbot Wibald, and the results confirm this attribution.



Samples of bone from the site of Irlich, Neuwied (50:26N 07:28E), Germany, submitted by Dr. M. Baales, Romisch-Germanisches Zentralmuseum Mainz, Forschungsbereich Altsteinzeit, Schloss Monrepos, 56567 Neuwied 13, Germany. Comments by M. Baales and A. von Berg, Archaologische Denkmalpflege, Koblenz.

OxA-9736human bone, IR 4, δ13C = −19.4‰12 310 ± 120
OxA-9848human bone, IR 2, δ13C = −18.9‰11 965 ± 65
OxA-9876human bone, IR 3, δ13C = −18.8‰2660 ± 40
OxA-9847human bone, IR 1, δ13C = −19.2‰11 910 ± 70

Comment (A. v. B. and M. B.): in 1998 A. von Berg found in the Kreismuseum Neuwied a cardboard box filled with sand, two flint (silex) artefacts, a bone point fragment, an animal tooth pendant, as well as human bones and teeth. The last can be attributed to at least three individuals: an adult with red coloured bones (among them a complete femur) and teeth, juvenile (neonate) also with red coloured bones and teeth, and an adult with light brown bone fragments, among them part of a skull cap.

These finds were uncovered on a heap of sand transported to the village of Heddesdorf (town of Neuwied) and found there together with red coloured sand. It is believed that these finds were originally dug out during sand mining in 1957 near the modern Neuwied suburb of Irlich, not far from where the Wied flows into the Rhine.

Samples from the bone point and a red-coloured bone fragment believed to belong to the second individual were sent for dating to the Utrecht AMS-laboratory. The bone fragment gave a date of 12 110 ± 90 BP (UtC-9221), δ13C = −23.9‰ while the bone point did not contain enough collagen to date.

This first age is now confirmed by the ORAU results, dating the adult femora (OxA-9847) and a rip fragment of the juvenile (OxA-9848). While the sample of the brownish skull cap gave a very young date (OxA-9876), a light brown rip fragment (probably belonging to an adult individual) is again of Late Glacial age (OxA-9736), giving the oldest date for all samples.

The new ORAU dates suggest that at least three Late Glacial human individuals were found at Irlich These are believed to represent burials of the period between the Gönnersdorf/Andernach Magdalenian and the Allerød Federmessergruppen so far not represented by radiometric dated sites in the Neuwied Basin.

German Pleistocene human remains series

Samples in this series were submitted by M. Street, Römisch-Germanisches Zentralmuseum Mainz, and T. Terberger, University Greifswald, in cooperation with R. E. M. Hedges, T. F. G. Higham and P. B. Pettitt (ORAU).

Samples of human bone from eight German contexts of proposed late Pleistocene age submitted to ORAU include remains from archaeological contexts (the burial at the Mittlere Klause in Bavaria and human remains from the Urdhöhle in Thuringia) which have been assigned to the Magdalenian, and skeletal material from possibly Late Glacial contexts in the Central Rhineland (Weißenthurm and Niedermendig).

An important aspect of the dating program are human remains recovered from river gravels without an accompanying cultural assemblage, but attributed to the late Pleistocene. A number of them (e.g. Hahnöfersand, Paderborn-Sande) are assigned to the earlier phases of the German Upper Palaeolithic and have contributed to the discussion of the origins and development of anatomically modern humans. In view of the potential importance of this material, it appeared worthwhile to determine the reliability of older results of dating (e.g. Churchill and Smith 2000, 88, 100).


The Hahnöfersand calvarium fragment was discovered in March 1973 together with a number of faunal remains lying on the south bank of the River Elbe near Hamburg (53:33N 9:43E). Since its publication, the Hahnöfersand find has played an important role in the discussion of models of Neanderthal replacement by modern humans. The calvarium has been classed as an anatomically modern Homo sapiens, but with ‘typical neanderthaloid features’(Bräuer 1980), and cited in the context of Bräuer's ‘African hybridization and replacement model’(e.g. Bräuer 1992; cf. Henke and Rothe 1994, 506 ff.) as possible evidence for coexistence and hybridisation of Neanderthals and Anatomically Modern Humans. Of crucial importance to this argument is a much quoted conventional radiocarbon date of 36 300 ± 600 BP (Fra-24) which placed the Hahnöfersand find exactly at the beginning of the Early Upper Palaeolithic in Central Europe (Bräuer 1980, 3). In view of the importance of this find in the discussion of the timing and mechanism of Neanderthal extinction, a new control date for the specimen was seen as an urgent priority.

The Museum suggested that the specimen had not been conserved or treated, however two casts of the specimen have been made. Damage from earlier sampling of the specimen is in clear evidence. A first sample failed at ORAU, but a subsequent attempt produced a reliable determination.

OxA-10306human bone, δ13C = −23.3‰7500 ± 55

Comments (M. S. and T. T.): the result places the specimen firmly within the Mesolithic period. The robusticity of the specimen might at first sight make such a young age for the find seem unlikely, but similarly massive specimens can in fact be demonstrated for the north German Mesolithic, e.g. the calvarium from Drigge on the island of Rügen in Mecklenburg-Vorpommern, radiocarbon dated to 6250 ± 80 BP (UZ-4093). which is characterized by its marked thickness (10 mm) and pronounced brow ridges (Terberger 1999, 25–9). A Mesolithic context for the Hahnöfersand find is rendered even more probable by abundant evidence for Mesolithic settlement along northern European river systems generally and by Mesolithic remains recovered from fluvial deposits specifically (for the Hamburg region see Lübke 1993).

Against the background of uncertain context of find recovery from fluvial deposits, the very good state of preservation of the specimen, the Mesolithic parallels quoted above and the uncertain reliability of the Frankfurt radiocarbon date series observed in other cases (see below, Binshof and Paderborn-Sande), we believe that an interpretation of the Hahnöfersand frontal bone as a further Mesolithic human, rather than as an early Homo sapiens with neanderthaloid features is the more probable scenario (Terberger et al. 2001).


The Paderborn-Sande, Nordrheinwestfalen, hominid skull fragment was recovered in 1976 by a suction dredger from gravel deposits (51:45N 8:41E). The calvarium is largely preserved and can be attributed to an aged adult or even senile male individual. Protsch (Henke and Protsch 1978) obtained a radiocarbon date of 27 400 ± 600 BP (Fra-15) on the specimen, controlled by an amino acid age for the same sample of 26 000 BP, which suggested that the find could be regarded as one of few Upper Palaeolithic hominid remains older than 20 000 BP. In view of the potential importance of this material, it appeared worthwhile to redate the specimen.

After discussion between B. Rüschhoff-Thaler (Westfälisches Landesmuseum), and TT, c. 1 g of material from the left parietal of the skull, stored at the Museum Schloss Neuhaus, was removed by the conservator of the Landesmuseum adjacent to an earlier sample. Details of the conservation treatment of the specimen are unknown, but a cast of the cranium had been made.

OxA-9879human bone, δ13C = −20.1‰238 ± 39

Comments (M. S. and T. T.): the very young result for Paderborn-Sande can be accepted as credible on a number of grounds. The robusticity of the Paderborn specimen can probably be explained simply by the sex (male) and the advanced age of the specimen (Henke 1984, 12). The broad skull-type was already regarded as a “progressive” characteristic and a possible indication that the specimen could be assigned to the final part of the Upper Palaeolithic (Henke and Protsch 1978, 102). This study had also in fact established that the greatest similarity of various features of the Paderborn-Sande specimen were with Neolithic to medieval populations. A possibly healed injury in the region of the forehead, which might represent an attempted trepanation, was considered unusual for a specimen of proposed Pleistocene age (Henke 1984, 6).

The ORAU result for the Paderborn-Sande calvarium can be reconciled with the poor or absent contextual information and the relatively good preservation of the specimen and on methodological grounds is certainly more reliable than the Frankfurt conventional date (see Hahnöfersand and Binshof-Speyer, this Arch. List). The authors believe that the ORAU result suffices to remove the Paderborn-Sande calvarium from the Pleistocene human record.


The exceptionally well preserved cranium, including most facial bones and a practically intact dentition, was recovered in 1974 during sand extraction by a suction dredger in Binshof close to Speyer (49:21N 8:28E), Rheinland-Pfalz. Information on the context of the specimen was derived at third hand and there is only a dubious association with Pleistocene faunal remains.

The specimen appears to represent a female; morphological comparison showed similarities with other Upper Palaeolithic human remains from central and western Europe, specifically with younger material of this period (Henke 1980, 292: 1982, 172). The cranium was radiocarbon dated by R. Protsch with a result of 21 300 ± 320 BP (Fra-40; Henke 1980, 277; 1982, 150), while amino acid measurement of the sample gave an age of 22 000 BP (asparagin-dating Fra-30), and it was, therefore, included in the present dating program.

After discussion between K. Kell (Museum Schwarzenacker), M. S. and T. T., sampling was carried out in the Bad Homburg Clinic, removing material adjacent to the foramen magnum (1.86 g).

OxA-9880human bone, δ13C = −19.3‰3090 ± 45

Comments (M. S. and T. T.): The Oxford AMS result (calibrated c. 1420–1290 BC at a probability of 64.8 %) places the specimen firmly in the Middle Bronze Age/ ‘Hügelgräberbronzezeit’(Joachim 1997, 5), so that the plausibility of the result must be discussed. The laboratory did not recognize any methodological problems with the sample. A younger context seems feasible from an anthropological point of view. The initial assessment of the Binshof calvarium by W. Henke (Mainz University) was only carried out after the first results of absolute dating were known and he, therefore, accepted it as an Upper Pleistocene specimen, although his analysis showed that the cranium lies within the range of variation of recent Central European material (Henke 1980, 276).

A Bronze Age date for Binshof can be perfectly reconciled with all attributes of the find, the unclear circumstances of discovery, its state of preservation and the fully modern cranial morphology. Against the ORAU result stands only the Frankfurt conventional radiocarbon determination, the validity of which must be questioned. The date (Fra-40) was obtained during a stage of trials in radiocarbon dating by the Frankfurt University laboratory, which was in fact never fully commissioned. To this can be added the known problems of conventional dating of bone during the early years of radiocarbon dating. The authors therefore recommend that the Binshof cranium be removed from the corpus of European Pleistocene human remains (Terberger and Street 2001).


The Emsdetten (52:11N 7:33E), Northrhine-Westphalia, calvarium was probably found by private collectors during sand-dredging in the 1970s, but precise information on the circumstances of retrieval is unavailable. The specimen, now in the Münster Geological Museum, is almost complete and in a better state of preservation than Paderborn-Sande (pers. coram. B. Rüschoff-Thale, Westfalisches Museum für Archäologie, Münster). The Emsdetten calvarium has been only sporadically mentioned in the literature and has not been formally published.

Preliminary anthropological examinations of the fossil have been conducted by B. Herrman (University of Göttingen) in the 1980s and by W. Henke (University of Mainz). Correspondence between the two anthropologists in 1984 suggests that the best correlation of the Emsdetten skull is with female Upper Palaeolithic specimens. AMS-dating of the skull was intended to control the proposed Pleistocene age and sampling was undertaken by B. Rüschhoff-Thale in March 2000.

OxA-10048human bone, δ13C = −17.4‰2460 ± 37

Comments (M. S. and T. T.): calibration of the result for Emsdetten places the specimen firmly within the first millennium BC. The broad cultural context of the specimen would therefore be Late Bronze Age/Early Iron Age. The site is located just outside the geographical distribution of the Hallstatt complex and at the west of the region of the Jastorf Culture (Brandt 2001).

Mittlere Klause

Excavations in 1914 at the Mittlere Klause cave near Neuessing (48:13N 11:20E), Kelheim District, Bavaria, revealed the burial of a man aged c. 30–40 at death. The remains can be attributed to an anatomically modern Homo sapiens and ochre staining of the bones is still recognizable.

The scant records of the Mittlere Klause stratigraphy suggest that the skeleton lay above a Middle Palaeolithic horizon and immediately below a Magdalenian layer. On the evidence of a (final Middle Palaeolithic?) leaf point, misinterpreted at the time, the excavators favoured a ‘Solutrean’ age for the skeleton, an interpretation which has subsequently been repeatedly criticised (eg Narr 1977; Schröter 1979; Wüller 1999).

In the 1970s a first radiocarbon date (UCLA-1869) taken on a tibia fragment yielded a result of 18 200 ± 200 BP (Protsch and Glowatzki 1974, 143), surprisingly close to the age proposed on typological grounds for the grave, but believed to be unreliable in view of the apparent desertion of Germany during this part of the Last Glacial. In the light of recent ORAU dating results (Wiesbaden-Igstadt: Arch. List 26; Street and Terberger 1999, 2000) a context for the grave close to the Last Pleniglacial no longer appears to be impossible. It was, therefore, considered desirable to obtain a second date for the Mittlere Klause skeleton.

With the authorisation of G. Grupe (Anthropologische Staatssammlung) sampling took place at Munich in February 2000. Following discussion between P. Schröter and T. T. a fragment of a vertebra with ochre staining (c. 3 g) was chosen. On information provided by P. Schröter, the entire skeleton had been treated with consolidants which made sample pre-treatment particularly important for this specimen, a factor taken into account by the laboratory.

OxA-9856bone, Homo sapiens, MKLAUSE 1, δ13C = −18.4‰18 590 ± 260

Comments (M. S. and T. T.): the ORAU date is very close to the UCLA date for the skeleton, overlapping at one standard deviation, and placing the Mittlere Klause burial close to the Pleniglacial, and also in accordance with the reported stratigraphic position of the burial.

Urdhöhle (Döbritz)

The Urdhöhle cave is located close to the better known Kniegrotte near Pöneck in Thüringen (50:42N 11:39E). The site was discovered and excavated by M. Richter from 1946–1959 (Feustel et al. 1971). It was not possible to identify a clearly defined Palaeolithic layer, but several flint artefacts and Pleistocene faunal remains recovered from a loessic deposit containing dolomite scree shows that humans were present during the later Palaeolithic. The excavator recovered skeletal remains of one adult male, two adult females and one juvenile from a 4 m broad chamber above the 26 m long main gallery. The remains were scattered across c. 20 m and associated by the excavator with the Palaeolithic artefacts (Grimm and Ullrich 1965, 50; Feustel 1989; Bach 1974). The conditions of deposition are poorly known and disturbance of the skeletons is considered possible.

Radiometric dating was intended to test whether the human remains can indeed be assigned to the Magdalenian. With the authorisation of S. Dušek (Director, Thüringisches Landesamt, Weimar) sampling was carried out in 2001 by T. T. in the presence of M. Kiißner (University, Halle), T. Schüler (Thüringisches Landesamt, Halle) and M. S., In addition to the published finds, a second crate of probably unrecognized finds from the Urdhöhle was viewed. These comprised faunal remains, many of them typically Pleistocene species, and the greater part of a human skull. According to associated labels this was recovered from the ‘SW-Graben’ (trench), some 10 m northwest of the other human remains.

Two samples were removed from the published Urdhöhle human remains (skull and humerus), while a third sample is from the unpublished calvarium. The published specimens are probably treated with consolidants (which was communicated to the Oxford laboratory), whereas the new specimen is untreated. Approximately 1 g of material was removed from each element.

OxA-10827human bone, Urdhohle-1, δ13C = −19.5‰8470 ± 50
OxA-10828human bone, Urdhohle-2, δ13C = −19.3‰8400 ± 50
OxA-10829human bone, Urdhohle-3, δ13C = −20.0‰1290 ± 34

Comments (M. S. and T. T.): OxA-10827 and −10828 do not confirm the expected Magdalenian age of the specimens. They possibly document use of a chamber inside the Urdhöhle cave as a burial site during the earlier Mesolithic, which would be an interesting parallel to situations known from the British Isles, where caves occupied as living sites during the late Palaeolithic were re-used for a different function during the early Holocene (Arch. List 4 and 6). The early medieval age of sample OxA-10829 warns that there is a probability of quite drastic mixing of contexts across relatively small distances.


In the 1920s parts of a human skeleton were recovered at Weißenthurm (50:25N 7:27E) in the Neuwied Basin below pumice from the Laacher See eruption, which is now known to date to 13 000 BC. The find was interpreted as a Late Glacial human killed by the eruption and has since enjoyed a certain local notoriety as ‘Der Mensch von Weißenthurm’. The remains have been missing since World War 2 and since then several attempts to relocate the finds have failed, so that it is generally believed that they were destroyed. References in the published literature (Guenther and Musil 1993) to Pleistocene faunal remains in the store of the Koblenz Mittelrhein Museum led M. S. to search for this material. Apart from faunal and lithic material it was possible to find human remains labelled as the missing Weißenthurm find. An associated photocopy of the original entry of the finds into the museum catalogue describes 23 human remains with specific reference to the published Weißenthurm specimen, stating that the human remains described by A. Günther (1924) were transferred on 10.5.1963 to the Koblenz museum from the collection of F. Michels (a local quarry owner and antiquary/geologist) following his death. A problem presented by the Koblenz skull fragments is that they demonstrate at least two and possibly three individuals, whereas Günther only refers to one skeleton, and it is clear that the equation of the Koblenz human remains with the ‘Mensch von Weißenthurm’ described by A. Günther is ambiguous. What does seem certain is that the finds were taken into the collection of the Mittelrhein Museum in 1968 under this description (so that any confusion of context occurred before this date). In order to clarify the status of these potentially Pleistocene human remains both of the definitely identified individuals were sampled.

OxA-10900human bone, 9, δ13C = −21.0‰1945 ± 70

Comments (M. S. and T. T.): The result on the first sample clearly showed that the ‘Weißenthurm’ finds stored in Koblenz are of Roman age. The second sample was not dated. Whether the remains might indeed be Günther's material, but in a chronological context much younger than he believed (he never saw the bones in situ), or whether unrelated human remains have erroneously been stored with a misleading label, cannot be finally decided. An identification of the Koblenz material with a late Pleistocene hominid can, however, be definitively ruled out.


A human skull recorded as deriving from Pleistocene deposits near Niedermendig (50:22N 7:16E), Neuwied Basin, in the Central Rhineland has, to the knowledge of the authors, never been officially described. The two refitting fragments of a calvarium were found by M.S. in the storerooms of the Rheinisches Landesmuseum in Bonn. Their entry under catalogue number S 446 reads: ‘Remains of animal bones found in Niedermendig in loess during the sinking of a shaft below the pumice in 1884. Donated by the owner of the pit Frz. Xaver Michels in Andernach’ (authors’ translation).

The two dark brown-stained human skull fragments are the os occipitalis and the left os temporalis, which refit along the unfused suture. Their surface is in places polished and reveals very fine cut/scrape or abrasion marks, possibly anthropogenic, but equally possibly due to scouring (in sediment or in water?). Associated faunal material comprises three pale, chalky pieces of reindeer antler (1 section of beam and 2 refitting tine fragments), some of them with what appear to be modern ‘engravings’, a skull fragment of a sheep with both horn cores, a cervical vertebra fragment, possibly of a carnivore and two undetermined weathered bone fragments. A fragment of the human os temporalis was removed for radiocarbon dating to clarify the chronological context of this fragmentary but very well preserved specimen.

OxA-10821human bone, 11, δ13C = −18.9‰2313 ± 36

Comments (M. S. and T. T.): that the Niedermendig material contains a mixture of contexts was already clearly shown by the presence of Pleistocene (Rangifer) and recent species (Ovis). OxA-10821 proves that the human skull is of appreciably younger age than the Pleistocene fauna, although it is not demonstrated beyond question that they are associated with the domestic species. Possibly this material from old collections is simply unreliably archived, alternatively, the different state of preservation of the finds might suggest that they were found at different depths during the sinking of the shaft from which they were supposedly recovered. In the view of the La Tène age of the specimen it might prove worthwhile to investigate the possible modifications to the skull more closely.

Concluding comments (M. S. and T. T.): samples of anatomically modern human remains from eight German localities, all believed to be of late Pleistocene age, were submitted for dating at the ORAU facility. A number of the specimens were regarded as important for specific problems: Hahnöfersand for the question of Neanderthal-Modern hybridisation, Binshof-Speyer and Mittlere Klause for the question of Pleniglacial human presence in Central Europe.

Of four specimens recovered during the 1970s from fluvial deposits, three had been previously dated to the late Pleistocene by the former Frankfurt radiocarbon laboratory using conventional techniques. The intention of the present dating program was, in this case, to confirm their exact chronological position within the Upper Palaeolithic sequence using more modern techniques.

All four ORAU results on the dredged finds deviate greatly from expectations and, in the case of the previously dated specimens, show a massive discrepancy compared with the Frankfurt results. The authors conclude that, in all four cases, the accelerator dates and also a range of further indications warrant the removal of the Hahnöfersand, Paderborn-Sande, Binshof-Speyer and Emsdetten ‘fossil hominids’ from the Pleistocene hominid record.

The fact that all three conventional Frankfurt results (for what can now be seen to be a very heterogeneous group of human calvaria) placed the specimens into the Pleistocene, must cast great doubt on the general validity of the dates produced by this now defunct laboratory. In view of the demonstrated unreliability of the Frankfurt date series, we recommend that the conventional date for the Kelsterbach calvarium, which was unavailable for sampling by the authors of the present study, should also be rejected unless confirmed as valid by re-dating the specimen.


Samples of human bone from the Middle Neolithic cemetery of Trebur (49:56N 08:24E), district of Gross-Gerau, south-west Germany. Excavated in its entirety in 1988–9 by the Landesamt für Denkmalpflege Hessen (H. Goldner), the cemetery at Trebur consists of 79 graves of the Hinkelstein (HST) and 58 graves of the Grossgartach (GG) cultures of the Middle Neolithic period. The arrangement of GG burials around a central area of HST graves clearly indicates that the GG interments were made later (Spatz 1999; Spatz in press). Submitted 1994 and 1995 by H. Spatz, Landesamt fur Denkmalpflege Hessen.

OxA-5598human bone, grave 127, δ13C = −19.7‰6065 ± 70
OxA-5322human bone, grave 132, δ13C = −19.4‰5980 ± 90
OxA-5321human bone, grave 68, δ13C = −19.1‰5945 ± 55
OxA-5595human bone, grave 52, δ13C = −19.9‰5840 ± 55
OxA-5597human bone, grave 107, δ13C= −19.1‰5835 ± 55
OxA-5593human bone, grave 17, δ13C = −20.1‰5910 ± 90
OxA-5320human bone, grave 58, δ13C = −18.9‰5840 ± 55
OxA-5594human bone, grave 21, δ13C = −19.5‰5770 ± 55
OxA-5599human bone, grave 130, δ13C = −19.3‰5760 ± 55
OxA-5596human bone, grave 106, δ13C = −19.8‰5685 ± 55

Comment (H. S.): the 14C dates of the Trebur necropolis were made with three objectives: (1) to produce the first series of dates for the earlier Middle Neolithic period, (2) with the aid of high precision dates, to verify that the use of the cemetery was interrupted between the HST and GG phases, a hiatus which has been postulated in view of the almost complete absence of graves from the early phase of the GG culture, and (3) to obtain absolute dates for the length of use of the cemetery.

The span of time allowed for the cultural sequence of HST-GG-Rossen within the Middle Neolithic can be delimited by means of dendrochronological dates: the Late Early Neolithic (Erkelenz-Kückhoven) and the beginning of the Late Neolithic (Aichbühl group and Egolzwil culture). Evidence for contact between HST and the latest phase of the Linienbandkeramik culture is provided by finds from Köln-Lindenthal, Bad Nauheim-Rödgen, Metz and Vikletice (Spatz 1999).

A series of 25 conventional 14C dates made in Heidelberg did not comply with the expected results in three aspects. (1) The temporal span of dates is far too broad (fiftieth to forty-third centuries BC). This assigns the latest graves from the earlier Middle Neolithic to the Late Neolithic period. (2) The spectrum of dates for both HST and GG is coeval to a great extent, yet this is convincingly contradicted by the planigraphy context in Trebur as well as stratigraphic evidence in Mühlhausen and Hilzingen (district of Konstanz) (Dieckmann 1987). (3) The relative chronological sequence of the graves in Trebur does not correspond to the absolute dates.

Human bone samples from 10 of the 25 graves dated in Heidelberg were analysed with AMS at Oxford. Comparison of both series of dates shows that the dates from Heidelberg are generally later than those from Oxford. Although the AMS results approach the archaeological expectations in all points listed above, they nevertheless still tend to be too recent. The late date of some samples is probably due to their contamination by exogenous organic material. The contamination seems to be more effectively reduced when using pretreatment methods for AMS analysis than conventional methods.

Moreover, the generally higher age, compared with conventional dates, the better differentiation between both cultures and the more limited spectrum of the Oxford AMS-dates are probably attributable to the fact that seven of the ten AMS samples (graves 17, 21, 52, 106, 107, 127, 130) were taken from the femur. Namely, it has been demonstrated that spongiosa is more susceptible to contamination than compacta (Steppan in press). Samples from which the conventional dates derive were composed of varying amounts of spongiosa and compacta, which explains the overly broad spectrum of the Heidelberg dates as well as the seemingly extensive coevality of the Hinkelstein and Grosspartach use of the Trebur necropolis.

Elbe-Kieswerk Magdeburg

Human remains found in a gravel pit near Magdeburg (52:20N 11:40E), Germany, together with Pleistocene faunal remains, submitted in 1990 by T. Weber, then of Landsmuseum für Archäologie. Halle, Saale, Germany.

OxA-3998human skull bone, os frontale, δ13C = −17.3‰2575 ± 70

Comment (T. W.): these remains are not, as expected, from the Middle Palaeolithic age, based on fluorine tests and morphology, especially of the os frontale (Lange and Weber 1991; Weber 1989), but are apparently from the Iron Age (Funke in press).


A sample of bone from a human child skeleton excavated at the famous east gate of the Celtic Oppidum of Manching, near Ingolstadt (48:43N 11:32E), Germany, submitted by R. Gebhard, Prahistorische Staatssammlung, Munchen, Germany. The find was interpreted as a sacrifice during the building of the east gate at the end of the second century BC.

OxA-5658human bone, 1963, 1059a. δ13C = −19.4‰2435 ± 55

Comment (R. G.): on restudying the stratigraphy of the excavation once more some problems were detected, one being the stratigraphic position of the skeleton. That was the reason for dating this sample. The date now strongly supports the idea that the stratigraphy of the excavation has to be re-interpreted.

Harthausen, Baden-Wurttemburg

Samples were taken from early collections in the British Museum in the hope of dating two important Continental grave groups with diagnostic artefacts, thus to lend further support to the Bronze Age chronological revisions of recent years. Submitted by S. Needham, Dept. Prehist. & Romano-British Antiquities, Brit. Mus., London. See also Courtavan, Aube (this Arch. List).

Excavated by Edelmann in 1901, an inhumation grave (48:15N 09:10E) under a barrow yielded two decorated limb-bands of a classic Tumulus type. Edelmann described them as having been on the upper arms (Zürn and Schiek 1969, 13). They were acquired by the British Museum in 1908 (registration 1908 8–1 207–208) along with two pieces of human ulnae; green staining on the bones, presumed to derive from copper, seemed to confirm the association and one was sampled for dating.

OxA-6222human bone, 1908 8–1 207, δ13C = −19.2‰2400 ± 50
OxA-6722human bone, δ13C = −18.8‰2460 ± 45

Comment: (S. P. N.): the original determination and the double-check gave the same result, but again are much later than the date of the ornaments, expected to be in or around the fourteenth century BC. The Early Iron Age dates for the bone now suggests either the re-use of the ornaments in that later period (which could explain the unconventional body position—the form normally adorned the leg) or the conflation of two separate contexts prior to arrival at the Museum. In the latter case, however, the arm bones would need to have been associated with some other copper alloy object to account for the staining.

Slovak Republic

Moèa, Komárno District

Sample of human cranium from Moèa, near the Danube River and Komárno, South Slovakia (47:45N 18:25E), excavated 1990 by L. Katona and submitted by A. Šefèáková, Dept. Anthrop., Slovak Nat. Mus., Bratislava, Slovakia.

OxA-7068human calvarium, P30/95 7369, δ13C = −21.7 ‰11 255 ± 80

Comment (A. Š.): in April 1990 an excellently preserved calvarium was found during gravel extractions from Danube River bottom at Moèa (district of Komárno). The skull was excavated by a dredger from a depth of c. 3 m. Neither animal nor archaeological remains were associated. The site is located in a small dead river arm on the 1742.3 riverine km, about 50 m from the riverbank, on the Slovak side of the Danube River. The gravels near the Danube and the Váh rivers yielded a large number of Late Glacial faunal remains. Thus, seemed likely that the human specimen found in one of these gravels also derived from this period. The position of the primary skull in the Danube River was reconstructed to an area at most 100–150 m upstream from its secondary position (1100 m downstream from the boundary of Moèa). It is located at the surface of the present edge of the ‘low’ Danube terrace T 1. The terrace (Kravany Danube step) is created by fluvial sandy gravels of Würm (W2 and W3) and surface planar cover of bright and sandy loams (W3). Stratigraphically the skull is placed to late Wûrm-Allerõd and Younger Dryas phases (according to the rest of youngest terrace fluvial sediments in the near Štúrovo town area), ie 11 800–10 200 yr BP, in agreement with OxA-7068. We suppose that this is the first find of Late Upper Palaeolithic human remains in the territory of Slovakia. Among the middle-European finds, a little older (final Wûrm 3) skeletal remains from Staré Mìsto (dpt. Uherské Hradištì, Czech Republic) are geographically the nearest ones. From at least 150 individuals of this geological age found in Europe up till now, only a few are as well preserved as this skull from Moèa. For more information see Šefèáková (1997), Šefèáková et al. (1999).


Mondeval de Sora

Sample of human bone from the Mondeval de Sora, Site 1, sector 1(46:27:60N 12:05:38E), Province of Belluno, northern Italy, submitted by F. Fontana and A. Guerreschi, Dipartimento di Scienze Geologiche e Paleontologiche, Universita di Ferrara, Corso Ercok I d'Este, 32–44100 Ferrara.

OxA-7468human bone, δ13C = −19.5‰7425 ± 55

Comment (F. F.): the bone comes from a burial which was found in 1987 in the site VFI sector I of Mondeval de Sora, and which is now on display at the Museo della Val fiorentina (Selva di Cadore – BL). VFI is one of the most important high altitude sites (c. 2150 m altitude) of Mesolithic age in Italy (see Aliciati et al. 1992), where a complex stratigraphic series has been brought to light revealing the presence of well-preserved dwelling structures and of this burial. The burial was accompanied by several grave goods, including flint, bone and antler artefacts. The typology of these objects has allowed an attribution of the burial to the Castelnovian culture (Late Mesolithic). The dating of the bone has confirmed the cultural attribution of the grave.



Samples of bone from the site of Candia, Herakleion, Crete (35:20N 25:08E), Greece, submitted by C. Stringer, Dept. Palaeontol., Natur. Hist. Mus., London. Comments by C. B. Stringer and T. Shaw, Natur. Hist. Mus.

OxA-6021human bone, δ13C = −21.4‰3225 ± 65

Comment (C. B. S. and T. S.): in 1627, a child's skull encrusted in travertine was presented to Sidney Sussex College. It had been found during the sinking of a well near Candia, in Crete. Through the physician William Harvey it was subsequently shown to Charles I as a curiosity. In 1995, Thurston Shaw arranged with the Master of Sidney Sussex College for the specimen to be loaned to CBS for study and possible dating. A sample of cranial bone was removed for accelerator dating, the result demonstrating that the specimen is Holocene in age. The δ13C value indicates that the determination was not significantly contaminated by the calcareous encrustation. Although not Pleistocene in age, the Candia skull provides an interesting example of rapid ‘fossilisation’.



Samples of bone from a Cave near Safed (Zefat), (03:00N 35:05E) Israel, submitted by P. Mitchell, 2 Milton Mansions, Queens Club Gardens, London, W14 9RP, through the NERC-funded ORADS facility.

OxA-10107human bone, 5111, δ13C = −18.5‰595 ± 40
OxA-10108human bone, 5129, δ13C = −17.6‰294 ± 35
OxA-10129human bone, 5164, δ13C = −11.4‰445 ± 33

Comment (P. M.): the skulls were recorded in the University of Cambridge records as Crusader Period (twelfth to thirteenth century AD). This was based on the observations of those who recovered the bones in 1912, but we have no idea who they were or on what information they based this suggestion. The three dates span 1290–1670 AD, which shows that they were not crusader, but were remains of people who lived under Mamluk and then Ottoman rule.

The case of trepanemal disease is still highly significant as it is dated to well before the voyage of Columbus to the New World, confirming the disease this side of the Atlantic before that time. It is also the oldest case so far identified in the Middle East.

While the dates were not what were expected, they do confirm how important it was to check their provenance with a radiocarbon determination.

South Africa

South of Umgababa

Sample of bone from South of Umgababa, Natal (30:09S 30:50E), South Africa, submitted by J. Sealy, Dept. Archaeol., Univ. Cape Town, South Africa.

OxA-6220human bone, UCT 5616, δ13C = −10.6‰1025 ± 35

Comment (J. S.): this sample was submitted as part of a project to identify the remains of early farmers from shell middens along the east coast of South Africa, especially those dating to the first millennium AD. It has been suggested that marine protein was an importance resource for the first agriculturalists spreading southwards along the coastal corridor; dating and stable isotope analyses of skeletal remains should help us to explore this issue.

Elands Bay, Steenboksfontein Caves and Watervalsrivier

Samples of bone from Eland Cave (32:03S 19:01E), Steenboksfontein Cave (32:09S 18:20E) and Watervalsrivier (32:00S 18:58E) South Africa, submitted by J. Sealy, Dept. Archaeol., Univ. Cape Town, South Africa.

OxA-6217Eland Cave, human bone, UCT 5666, δ13C = −18.7‰2145 ± 50
OxA-6219Steenboksfontein Cave, human bone, UCT 5669, δ13C = −12.5‰2445 ± 50
OxA-6218Watervalsrivier, human bone, UCT 5668, δ13C = −18.7‰1985 ± 50

Comment (J. S.): these three samples, all from juvenile skeletons, come from the Elands Bay/Clanwilliam area of the Western Cape of South Africa, where stable isotope analyses of human skeletons have contributed to a debate about settlement patterns among Later Stone Age hunter-gatherers. These individuals expand the database, especially for the Clanwilliam area (Eland Cave and Watervalsrivier). The date for Eland Cave was somewhat older than expected, given the excellent organic preservation at this site. The results are, however, consistent with previous dates for skeletons from this region, many of which are c. 3000 to 2000 BP (Sealy et al. 2000).

Old World Palaeolithic and Mesolithic

Great Britain

Wye Valley Caves

Samples from King Arthur's Cave (NGR SO 54581558), Merlin's Cave (NGR SO 55661533) and Cavall's Cave, (NGR SO 54621533), Wye Valley, submitted by R. N. E. Barton, Dept. Anthropol., Oxford Brookes Univ., Hcadington, Oxford, C. Price, Scarab Research Centre, Univ. Wales College, Newport and A. P. Currant, Dept. Palaeontol., Natur. Hist. Mus., London, through the NERC-funded ORADS facility.

King Arthur's Cave

OxA-6838Microtia, gregalis bone, KAC 580, δ13C = −20.3‰12 300 ± 100
OxA-6839Rangifer tarandus bone, KAC 624a, δ13C = −18.6‰9930 ± 90
OxA-6840Lepus timidus bone, KAC 624b, δ13C = −21.0‰12 240 ± 100
OxA-6841Dicrostonyx torquatus bone, KAC 672, δ13C = −20.8‰10 120 ± 90
OxA-6842Microtus gregalis bone, KAC 722, δ13C = −21.9‰10 390 ± 90
OxA-6843Lemmus lemmus bone, KAC 756, δ13C = −20.5‰10 380 ± 90
OxA-6844Cervus elaphus bone, KAC 854, δ13C = −19.3‰12 250 ± 100
OxA-8997Microtus gregalis bone, KAC 1336, δ13C = −22.8‰11 250 ± 130

Comments (R. N. E. B., C. P. and A. P. C): eight determinations from a well-stratified sequence of Late Glacial deposits on the west side of the cave entrance (Barton 1996, figs. 2 and 3). OxA-6839 to −6841 are all from Unit 3 (yellow scree). The assumed Younger Dryas age of this unit has been broadly confirmed by the new AMS results. They are amongst the first direct dates on small vertebrates which include species now extinct in the British Isles. The assemblage reflects the cool and generally dry conditions of the stadial. The bones were collected from small pockets of microfauna accumulated within the scree and are believed to be the remains of owl pellets (Barton et al. 1997). No archaeological finds were recovered in this unit. The presence of OxA-6840 in a higher stratigraphic position than expected is confirmed by AMS dating, which implies some degree of reworking of earlier sediments. OxA-6844, −6838 and −8997 are from Unit 4 (grey scree) and immediately beneath Unit 3. OxA-6844 is on a smashed red deer bone and dates human presence at the site. It also overlaps statistically with two previously published AMS dates on red deer from the earlier collections, OxA-1562 (12 120 ± 120 BP) and OxA-1563 (12 210 ± 120 BP) (Arch. List 9). Excavation of this unit, at the same level as the red deer, revealed a small assemblage of Late Upper Palaeolithic artefacts including backed material (Barton 1996, fig 4). The AMS dating of Microtus gregalis (OxA-6842, −6838 and −8997) indicates a continuity in this species throughout the Late Glacial Interstadial and subsequent stadial.

Merlin's Cave

OxA-8071Lepus timidus bone, MER 21A, δ13C = −20.7‰10 270 ± 65
OxA-8072Lemmus lemmus bone, MER 40A, δ13C = −21.3‰9685 ± 60
OxA-8073Ochotona pusilla bone, MER 40B, δ13C = −21.0‰9915 ± 60
OxA-8146Arvicola terrestris bone, MER 23, δ13C = −20.8‰10 160 ± 65
OxA-8167Microtus oeconomus bone, MER 21B, δ13C = −22.1‰450 ± 80

Comments (R. N. E. B., C. P. and A. P. C): the five specimens were collected from a large remnant of deposit rich in small vertebrate remains on the west wall of the cave near its entrance (Barton 1996, figs. 6 and 7). The deposits are quite heavily cemented and there appear to be traces of two stalagmitic floors. The upper floor is associated with human remains of presumed Late Bronze Age date (Parry 1994). The microfaunal samples all derive from beneath the uppermost stalagmitic floor from owl pellet accumulations and have no apparent archaeological association. Except for OxA-8167, all of the dated samples occur in the correct stratigraphic order. The latter was not surrounded with matrix when collected so it could have fallen from a higher position within the sequence. This apparently late record of Microtus oeconomus is not unexpected given its well-attested presence in Holocene deposits elsewhere in SW Britain. The date on Ochotona pusilla is the second recorded for this species in the cave. It agrees well with the date of OxA-516 (10 020 ± 120 BP) obtained on a combined sample of steppe pika mandibles (Arch. List 3). The microfaunal specimens compare closely with those from the King Arthur's Cave sequence and reflect the generally dry, open steppic conditions prevailing in this region during the Younger Dryas Lateglacial stadial.

Cavall's Cave

OxA-8168Equus sp. tooth, CAV D5 (116), δ13C = −21.4‰7440 ± 70

Comments (R. N. E. B., C. P. and A. P. C): OxA-8168 is on a single un-modified horse incisor with no direct archaeological associations. It was stratified in a grey-brown silty clay, 10 cm below two well-separated archaeological horizons (Romano-British and Early Bronze Age), in a context suggesting an earlier Postglacial age. The date is slightly younger than expected and confirms that horse was present, albeit as a rare element, during the developed woodland phase of the Holocene.

Mother Grundy's Parlour

Teeth of wild horse excavated from in front of Mother Grundy's Parlour, Creswell Crags, Derbyshire (NGR SK 536743), by A. L. Armstrong (1923–5) and in the collection of the Natural History Museum, London. Submitted by R. M. Jacobi, Dept. Palaeontology, Natur. Hist. Mus., London, through the NERC-funded ORADS facility.

OxA-8738Equus ferus tooth, Laura Kaagan 7, δ13C = −20.6‰11 970 ± 75
OxA-8739Equus ferus tooth, Laura Kaagan 8, δ13C = −19.6‰12 170 ± 80

Comment (R. M. J.): both are lower cheek teeth and both have been transversely broken. These ‘culture breaks’ (West 1997) have been suggested as resulting from blows to the lowest margin of the mandible powerful enough to shear the teeth within the jaw. Marrow extraction is assumed to have been the goal. These fractured teeth, therefore, directly date when wild horses were being processed by humans at Creswell Crags. For further dates on horse teeth from Mother Grundy's Parlour see Arch. Lists 18 and 22.

Lynx Cave

One of two fragments of bone point(s) excavated by J. D. Blore (1962–1981) at Lynx Cave, near Llanferres, Denbighshire, North Wales (NGR SJ 199594). Made available by J. D. Blore and submitted by R. M. Jacobi, Dept. Palaeontology, Natur. Hist. Mus. London, through the NERC-funded ORADS Facility.

OxA-8164bone, NW2, δ13C = −21.5‰11 700 ± 90

Comment (R. M. J.): the excavator recognizes seven layers, labelled A-G from top to base. This fragment of point is from layer C and, despite the likely Post-Glacial age of other items from this layer (Blore, 1981), is now clearly dated to about the middle of the Late Glacial Interstadial. The result is of particular interest for directly dating Palaeolithic human use of the cave. Some of the flint and chert artefacts from Lynx Cave appear, from their typology, also to be Upper Palaeolithic. Notably, these include straight-backed or partially backed blades/bladelets of a type seemingly absent from “Creswellian” contexts which, on present radiocarbon evidence, appear to be older than this result from Lynx Cave. However, a connection between the bone point and these lithic artefacts remains necessarily speculative as the majority of the latter are from layer D or from disturbed contexts. Burrowing by animals, of which there is recent evidence, may have been responsible for this stratigraphic separation.

Upper Kendrick's Cave

Perforated canine of badger (Meles metes) collected in December 1978 during excavation directed by Melvyn Davies at Upper Kendrick's Cave, Great Orme, Llandudno, Conwy, North Wales (NGR SH 78008284). Submitted by R. M. Jacobi, Dept. Palaeontol., Brit. Mus., London, through the NERC-funded ORADS facility.

OxA-5862Meles meles tooth. δ13C = −20.8‰9945 ± 75

Comment (R. M. J.): this is the left upper canine of badger, perforated and with transverse incisions on all four aspects of its fang which has also been thinned to facilitate perforation. This tooth was found during excavations in Upper Kendrick's Cave at 2.3 m below site datum and in layer 6—angular limestone blocks with a little loose brown cave-earth (Davies 1979). Similarly perforated and incised teeth of red deer and wild cattle had been found in 1880, but it is uncertain whether these are from Upper or Lower Kendrick's Cave.

Presumably a decorative item, this tooth may not be evidence for a local presence of badger at this time. However, together with OxA-111 (10 000 ± 200 BP, Arch. List 2) for a wild horse mandible decorated with zigzags from Kendrick's Cave (Sieveking 1971), this date confirms human activity in this part of Wales at about the boundary of the Pleistocene and Holocene.

Dog Hole fissure

Upper canine of lynx (find 73) recovered in 1978 during rescue excavation directed by R. D. S. Jenkinson of a fissure filling exposed by rock fall west of the west entrance to Dog Hole (West Pin Hole) in the northwest corner of Creswell Crags, Derbyshire (NGR SK 532 742). From the collection of the Creswell Crags Museum and Education Centre (DH.418) and submitted by R. M. Jacobi, Dept. Palaeontology, Natur. Hist. Mus., London, through the NERC-funded ORADS facility.

OxA-8737DH418 73, lynx tooth, δ13C = −20.7‰9570 ± 60

Comment (R. M. J.): this canine refits into a left maxilla (find 46: DH.181). Excavation of Dog Hole fissure yielded molluscs, fish, amphibians, large and small mammals (including bats), birds, charcoals and flint flakes (Jenkinson et al. 1982). The presence of the immigrant mollusc Oxychilus cellarius would indicate an age for the fissure filling after about 8500 BP (Kerney, 1977) and it was assumed that the other fauna was similarly more recent. However, OxA-8737 demonstrates that at least part of the fauna is older and suggests that there may have been a time-lag between death and incorporation in to the fissure fill. The result is of further interest as indirect evidence for an early local re-establishment of woodland at the beginning of the Holocene.

Wilton Hall

Sample of charcoal from Witton Hall, Birmingham (NGR SF 089917), submitted by J. D. Hurst, Worcestershire Archaeological Service, Woodbury Hall, Univ. Coll., Worcester.

OxA-6720charcoal, WM 1671, 514, δ13C = −26.2‰6195 ± 65

Comment (J. D. H.): this determination was requested in order to date deposits at the base of an archaeological sequence for which there was no other associated dating. The radiocarbon result provided was earlier than expected, but tended to confirm the view that these deposits were prehistoric based on their general character. This fieldwork was carried out as part of a site evaluation prior to redevelopment.


Sample of bone from Sherborne, Dorset, England, submitted by C. B. Stringer, Dept. Palaeontol., Natur. Hist. Mus., London, through the NERC-funded ORADS facility.

OxA-5239bone, ?horse, E.5305, δ13C = −20.9‰610 ± 45

Comment (C. B. S.): the “Sherborne Bone” is a mammalian rib fragment carrying an engraving of a horse's head, curated at The Natural History Museum, London. It was reportedly found in 1911, and published by Smith Woodward in 1914 as a genuine example of Upper Palaeolithic art. However, several workers have since questioned its authenticity, and therefore a sample from the obverse surface was collected for accelerator dating. The age determination confirms the view that a recent bone was engraved in the style of a known example, perhaps the depiction from Robin Hood's Cave published by Boyd Dawkins in 1877.



Samples of bone from the site Skottemarke, (54:42N 11:32E) Denmark, submitted by A. Fischer, Gl. Rosnaesuej 27, Kalundborg, Denmark 4400.

OxA-4864bone, A20371, δ13C= −21.3‰9570 ± 100
OxA-4865bone, A20364, δ13C = −21.5‰10 350 ± 110
OxA-5528bone, NM. A20364, δ13C = −21.8‰9310 ± 90

Comment (A. F.): OxA-4864 and −5528 represent two leister sponges from the early Mesolithic settlement site of Skottemarke, Denmark. OxA-4865 was also supposed to represent one of these bone artefacts. There was, however, a chance that it was actually taken from a third specimen, a harpoon of supposedly Late Palaeolithic date. The outcome of the AMS analysis supports this latter possibility.



A horse bone (Equus germanicus) from Zemst, O Vlaanderen (50:59:23N 04:23:45E), collected and submitted by V. Hendrix, KIKIRPA, Brussels. Comments by M. van Strydonck, KIKIRPA, Brussels and M. Germonpré, KBIN, Belgium.

OxA-7763Equus germanicus bone, BA285, δ13C = −20.7‰38 100 ± 1000

Comment (M. G. and M. van S.): This sample was submitted for a comparative measurement with an experimental series of determinations (see below) prepared at KIKIRPA and AMS dated at the Utrecht Laboratory. Electron spin resonance methods previously gave a date of 129 000 ± 10 000 BP (GRUN) for the bone.

KIK-753/UtC-5894collagen from hard part of Equus germanicus bone, BA285H/C, δ13C = −21.5‰40 300 + 900 −1100
KIK-777/UtC-5904tripeptides from hard part of Equus germanicus bone, BA285H/T, δ13C = −20.8‰27 400 ± 400
K1K-754/UtC-5895collagen from porous inner part of Equus germanicus bone, BA285Z/C, δ13C = −21.5‰36 500 ± 700
KIK-763/UtC-5900tripeptides from porous inner part of Equus germanicus bone, BA285Z/T, δ13C = −20.7‰24 320 ± 220

The outer hard parts of the bones give better results than the softer inner part. The tripeptide samples give a younger date than the gelatin samples. This aberration can be caused by a minor infiltration of young carbon in sample during HPLC preparation or by a conservation treatment of the bone with a collagen-based product.

Comment: OxA-7763 was measured on tripeptides (GlyProHyp and GlyProAla) following collagenase digestion (van Klinken and Hedges 1992) and should be compared with the collagen and tripeptide determinations above, prepared at KIKIRPA and Utrecht.


A sample of bone from Goyet (50.26N 05.00E), Belgium, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, because no dates were available for the small Gravettian assemblage to which this bone belongs.

OxA-4926bone, δ13C = −20.5‰24 440 ± 280

Comment (M. O.): the Gravettian attribution is confirmed, in an “average” chronological position with respect to the Belgian Gravettian, clearly more recent than its earliest phase (Maisières-Canal: 28–30 000 BP). The attribution to a more evolved phase of the Belgian Gravettian was anticipated by the presence of numerous truncated elements in the lithic industry. The assemblage is anterior to the last Pleniglacial (Eloy and Otte 1995).

Grotte du Docteur

Three samples of bone from Grotte du Docteur (50:33:45N 05:10:50E), Belgium, from a recent re-excavation of intact deposits on the terrace yielded bone samples which could be used to date the Mousterian and Aurignacian levels. Submitted by M. Otte, Préhistoire, Univ. Liège, Belgium.

OxA-8369Equus bone, GDD L9 fi10 YCS/1, stratum 3, δ13C = −20.9‰36 650 ± 1000
OxA-8370Equus bone, GDD L9 fi10 YCS/2, stratum 3, δ13C = −20.5‰27 740 ± 340
OxA-8371herbivore bone (?), GDD L10 fi5 RCS/3. stratum 4, δ13C = −20.4‰28 340 ± 360

Comment (M. O.): stratum 3 was attributed to the Aurignacian on the basis of the presence of carinated endscrapers and a bone polisher, stratum 4 to the Mousterian with Levallois technology. The conflicting dates from stratum 3 could have resulted from mixture of different industries (possibly Gravettian as well as Aurignacian) due to redeposition on the sloping edge of the terrace. The date for stratum 4, obtained from a bone sample near the top of the layer, is clearly intrusive (Miller et al. 1999).


Samples of bone from Huccorgne (50:33:45N 05:10:50E), Belgium, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium to corroborate the series of dates previously obtained.

OxA-3886bone, HU-JU-28, δ13C = −20.3‰26 300 ± 350

Comment (M. O.): the date supports other dates clustering at 26–27 000 BP and confirms the attribution of the lithic industry to the Gravettian (Straus et al. 2000).

Place Saint-Lambert, Liège

A sample of bone from St. Lambert Square, Liège (50:40N 05:35E), Belgium, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, to date the occupation of the local Mesolithic.

OxA-4781cervid bone, fi 4248, δ13C = −21.7‰7850 ± 75

Comment (M. O.): the result is in accordance with the attribution to the final Mesolithic (Gustin et al. 1994).


Samples of bone from Mégarnie, Liège, Belgium (53:35N 05:26E), submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, because no radiometric dates had previously been obtained for this small archaeological assemblage, characteristic of the Late Upper Palaeolithic (Tardiglacial).

OxA-7427Cervus elaphus antler, Mégarnie-1, δ13C = −17.1‰2470 ± 55
OxA-7428Equus sp. bone, Mégarnie-2, δ13 = −21.3‰ 22 160 ± 360 

Comment (M. O.): OxA-7427 could correspond to vestiges of material which is contemporary with the Neolithic burials found in the upper part of the stratigraphy. OxA-7428 clearly does not correspond to the Tardiglacial remains studied (Otte et al. 1997).

Trou Magrite

Sample of charcoal from Trou Magrite (50:13N 04:55E), Belgium, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium to obtain a date for the summit of Stratum 2, dated to the Aurignacian.

OxA-4040charcoal, STV91/L7, stratum 2 (top), δ13C = −23.2‰17 900 ± 200

Comment (M. O.): the sample was composed of small flecks of charcoal and was probably contaminated by fragments percolating down from overlying archaeological layers (corresponding to Gravettian, Magdalenian and even Mesolithic, today absent) (Otte and Straus 1995).

Trou da Somme

Samples of bone from Trou da Somme, Hastière (50:13N 04.50E), Belgium, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium to corroborate earlier dates obtained.

OxA-8308Ovibos moschatus bone, TDS 88–2 fiCHE 23 P21–8, δ13C = −19.1‰12 815 ± 75
OxA-8309Rangifer bone, TDS 97–022–9, δ13C = −18.7‰23 420 ± 220

Comment (M. O.): OxA-8308 confirms the early Magdalenian attribution of the lithic industry. OxA-8309 appears to correspond to faunal material in a non-archaeological level (Miller et al. 1998).


Las Fuentes de San Cristobal

Samples of charcoal from Las Fuentes de San Cristobal (42:20:03N 00:34:25W), Spain, submitted by J. Rosell Ardevol, Area de Prehistoria, Universitat Rovira I Virgili, Tarragona, Spain.

OxA-8524charcoal, FSC98 P3 NivO E6 2, δ13C = −22.4‰36 050 ± 550
OxA-8589charcoal, FSC98 P3 NivO E6 1, δ13C = −21.7‰27 200 ± 1000
OxA-8590charcoal, FSC98 P3 NivP F6 5, δ13C = −23.4‰36 000 ± 1900
OxA-8591charcoal, FSC98 P3 NivM F6 11, δ13C = −23.6‰20 220 ± 380

Comment (J. R. A.): the dating results are consistent with the stratigraphic succession. The archaeological level O has yielded two different dates separated by a wide time span. This may be related to post-depositional modifications affecting the site's stratification, namely bioturbation and secondary carbonation, which could have caused the infiltration of charcoal fragments from other units. One of the dates is likely to be associated with the archaeological level, while the other should be rejected on stratigraphic grounds. Due to the preliminary nature of the data, we cannot confirm which one represents the true age. Future excavations should provide further information on this level.

Parco and Guineu Caves

Samples of bone and charcoal from Parco Cave, Also de Balaguer (41:54:31N 00:56:31E) and Guineu Cave, Font-rubi (41:26:52N 01:34:53E), Spain, submitted by J. M. Fullola i Pericot, Dept. Prehist., Ancient Hist. and Archaeol., Univ. Barcelona, Baldiri Reixac st. s/n, E-08028, Barcelona.

OxA-10796charcoal, UBOX22-Parco3, δ13C= −23.512 605 ± 60
OxA-10797charcoal, UBOX24-Parco5, δ13C= −23.812 460 ± 60
OxA-10798charcoal, UBOX25-Parco6, δ13C= −23.313 175 ± 60
OxA-10835charcoal, UBOX23-Parco4, δ13C= −23.412 560 ± 130
OxA-10799human bone, UBOX26-Guineu1, δ13C = −18.54195 ± 40
OxA-10800human bone, UBOX27-Guineu2, δ13C = −18.54500 ± 40

Comment (J. M. F-P.): OxA-10835, −10797 and −10798 date level II. The coherence of these determinations is very important, because previously level II had only one date (ICEN-501: 10 390 ± 300 BP), which did not fit with the Parco cave sequence. These results confirm the age as the very final Upper Magdalenian, in a level with the bone industry (fragments of needle, a complete one and fragments of other bone and antler elements) and many of the hearths representing the space where people lived. The three determinations come from charcoal samples of three different combustion structures: numbers 17, 18 and 19. It is the first time in north-east Iberia that we have a precise excavation with absolute dates for this transitional moment. OxA-10799 and −10800 date the Chalcolithic phase of Guineu cave accurately. Level Ic of this site was a sepulchral level with a lot of human bones in a paradolmenic structure, inside the cave (a fact that is not very normal); Bellbeaker pottery appeared during the excavation, associated with the human remains. We have very few dates of this period in north-east Iberia, so these results are all the more important.


Gibraltar Caves Project

Samples from the adjacent sites of Gorham's and Vanguard Caves, Governor's Beach, Gibraltar (36:09N 05:21W) submitted by C. B. Stringer, Dept. Palaeontol., Natur. Hist. Mus., London, J. C. Finlayson, Gibraltar Mus., Bomb House Lane, Gibraltar, F. Giles Pacheco, Museo Municipal de El Puerto de Santa Maria, Cadiz, Spain, R. N. E. Barton, Dept. Anthropol., Oxford Brookes Univ., Oxford and P. B. Pettitt, ORAU, through the NERC-funded ORADS facility.

The following determinations come from long, well-stratified sequences with extensive Middle and Upper Palaeolithic human occupation evidence and were collected during recent excavations of the Gibraltar Caves Project (Barton et al. 1999, Stringer et al. 1999). Charcoal identification, were by Rowena Gale, fauna by Andrew Currant. The AMS programme, initiated in 1996, was part of a wider dating project which also involved OSL, U-Series and ESR dating undertaken by laboratories at Oxford and McMaster Universities (Pettitt and Bailey 2000; Rink et at. 2000). A major goal of the AMS radiocarbon dating programme was to establish the date of the terminal Middle Palaeolithic on Gibraltar, in addition to timing the presence of modern Upper Palaeolithic humans in this region of southern Iberia.

Gorham's Cave

OxA-6075charcoal, GOR95 240, context 22, δ13C = −25.2‰45 300 ± 1700
OxA-6997burnt bone, GOR96 526, context 7, δ13C = −21.2‰25 680 ± 280
OxA-7074Pinus sp. charcoal, GOR96 511, context 9, δ13C = −24.2‰30 200 ± 700
OxA-7075Pinus sp. charcoal, GOR96 512a, context 9, δ13C = −27.3‰29 800 ± 700
OxA-7076Pinus sp. charcoal, GOR96 512b, context 9, δ13C = −25.2‰30 250 ± 700
OxA-7077Pinus sp. charcoal, GOR96 514, context 9, δ13C = −24.7‰29 250 ± 650
OxA-7110Pinus sp. charcoal, GOR96 528, context 13a, δ13C = −24.4‰29 250 ± 750
OxA-7388burnt longbone fragment, GOR96 517, context 11, δ13C = −23.1‰29 100 ± 340
OxA-7790charcoal, GOR97 131, context 22d, δ13C = −24.5‰51 700 ± 3300
OxA-7791charcoal, GOR97 132, context 18, δ13C = −23.9‰42 200 ± 1100
OxA-7792charcoal, GOR97 119, context 24 (combustion zone in context 16), δ13C = −22.8‰32 280 ± 420
OxA-7979charcoal, GOR97 167, context 18, δ13C = −21.7‰23 800 ± 600
OxA-8525Juniperus/Tetraclinus charcoal, GOR98 951, context 19, δ13C = −20.9‰43 800 ± 1300
OxA-8526Pistacea charcoal, GOR98 952, context 19, δ13C = −23.7‰46 700 ± 1900
OxA-8541charcoal, dicotyledonous shrub, GOR98 928, context 19, δ13C = −24.9‰31 900 ± 1400
OxA-8542Juniperus/Tetraclinus charcoal, GOR98 1099, context 19, δ13C = −24.4‰42 800 ± 2100
OxA-10183charcoal, tuber or rhizome, GOR00/B10/125, δ13C = −22.3‰2530 ± 40
OxA-10230Pinus sp. charcoal, cone scale, GOR00/BB1/, δ13C = −26.9‰32 330 ± 390
OxA-10295Pinus sp. charcoal, GOR00/C23/47, δ13C = −21.6‰34 600 ± 900

Comment (P. B. P., R. N. E. B., C. B. S. and J. C. F.): the uppermost diagnostically Middle Palaeolithic assemblage from this cave was recovered from the light sands of Contexts 18 and 19. The samples dated from these horizons take the form of discrete charcoal lumps in apparent association with diagnostic lithics or of charcoal taken from well within a remnant combustion zone at the contact of Contexts 16 and 18. Given the nature of the sediment and visual indications that soft sediment loading had occurred, it is not surprising that there is something of a spread of resulting ages from Contexts 18 and 19. Excluding OxA-7979 which clearly stands out from the other dates from these contexts, the results indicate an age of between c. 29 000 and c. 51 000 BP (2σ) for these contexts. In view of the dates for the underlying Context 22 of between c. 42 000 and c. 56 000 BP (2σ) it is conceivable most of the charcoal fragments dated from Contexts 18 and 19 derive from lower down the sequence, moved up by the soft sediment loading. In view of this, OxA-8541 at c. 34 000–29 000 BP (2 sigma) may most likely reflect the age of the latest Middle Palaeolithic occupation of the cave. In view of this, the result for a sample of charcoal found in a well-stratified remnant combustion zone (OxA-7857) is interesting given that it is stratigraphically reliable and statistically the same age as OxA-8541. Together, these results suggest that the latest Middle Palaeolithic at Gorham's dates to between 33 000 and 29 000 BP. A parsimonious reading of the age ranges of the two would suggest that this is most likely to be around 31 000 BP, which agrees well with estimates for the latest Middle Palaeolithic elsewhere in southern Iberia.

Samples dated from above Contexts 18/19 and which have yielded small archaeological assemblages lacking in diagnostic pieces, have been dated to c. 28 000–31 000 BP and c. 25 000–26 000 BP. Of these, the former (Context 9 combustion zone, is statistically the same age as dates Waechter obtained for his Layer D, which might suggest that our Context 9, although it yielded few typologically diagnostic lithics, is the same as Waechter's early Upper Palaeolithic ‘Aurignacian-like’ layer. OxA-6997 could reflect subsequent Upper Palaeolithic (‘Gravettian’) use of the cave.

OxA-10295 comes from a small excavated trench about 7 m further inside the cave than the main series of dates. The pine charcoal, from a well-stratified context, was in close proximity to undiagnostic flakes but in raw materials commonly used in the Middle Palaeolithic. The date overlaps with those for the latest Middle Palaeolithic.

OxA-10183 derives from layer III, a loose, heavily bioturbated, humic rich deposit near the back of the cave which contains a mixture of archaeological materials of historic and prehistoric types. The cave was used as coastal sanctuary by Phoenicians, Iberians and Carthaginians between the seventh and third centuries BC (Guiterrez Lopez et al. 2000). The date appears to fit the sanctuary's main period of use and would correspond to a late Phoenician presence.

OxA-10230 is from layer IV, an apparently in situ deposit underlying layer III, with only Middle Palaeolithic artefacts and fauna. The date confirms that latest Middle Palaeolithic activity in Gorham's extended to the back of the cave.

Vanguard Cave

OxA-6891Juniperus sp. charcoal, VAN-S 96 285a, beneath shell midden, δ13C = −22.1‰54 000 ± 3300
OxA-6892Pistacea charcoal, VAN-S 96 285b, beneath shell midden, δ13C = −22.6‰46 900 ± 1500
OxA-6998Olea charcoal, VAN-S 96 245, context 52, δ13C = −25.1‰41 800 ± 1400
OxA-7078Pinus sp. charcoal, VAN-N 96 351, hearth in northern alcove, δ13C = −23.9‰>44 100
OxA-7127Olea charcoal, VAN-S 96 347, context 55, δ13C = −24.4‰>49 400
OxA-7191Sus sp. bone, VAN-S 96 230, context 52, δ13C = −15.1‰10 170 ± 120
OxA-7389charcoal, VAN-S 96 377, base of spit 3, δ13C = −25.5‰45 200 ± 2400

Comment (P. B. P., R. N. E. B., C. B. S. and J. C. F.): it appears likely that much of the sedimentary fill of Vanguard Cave had accumulated before the limits of radiocarbon dating, thus only the uppermost levels yielded samples for dating. This view is also supported by preliminary OSL dating of the Vanguard sequence (Pettitt and Bailey 2000). Six AMS radiocarbon determinations exist for the uppermost levels. Most are close to the upper limit of the method and in two cases (OxA-7127 and −7078) are effectively infinite ages. Both OxA-6998 and −7191 were in close association, the latter had a very low collagen content and the age should be considered suspect. OxA-6891 and −6892 derive from beneath a Middle Palaeolithic shell midden (Barton 2000; Fernández-Jalvo and Andrews 2000) but the sedimentary context of the finds suggests the dating may be considerably earlier. OxA-7078 was obtained from an isolated circular hearth feature in a northern side chamber of the cave, but it lacked any clear archaeological associations.


Buraca Escura

Buraca Escura (Dark Cave), is a small cave, situated on the Poio Novo canyon (39:58N 08:43W), a relief of Sicó; Buraca Grande, excavated 1991–7, is a cave situated in the northern slope of the Poio Novo valley, a Jurassic limestone mountain in the southern region of the Mondego river (Central Portugal). Submitted by H. Moura, Rua da Fontinha, 6 Fala, 3040–168 Coimbra, Portugal.

OxA-5523Caprid phalanx, BE4, δ13C = −20.1‰22 700 ± 240
OxA-5524Equus phalanx, BE2, δ13C = −20.4‰21 820 ± 200
OxA-5522antler, BG2, δ13C = −20.5‰13 050 ± 100

Comment (H. M.): excavations in this cave between 1991 and 1995, revealed three archaeological levels: recent prehistory, Upper and Middle Palaeolithic. All along the sequence several erosion episodes can be detected. The scarce lithic industry in the different levels is associated with numerous faunal remains. Among the six sub-divisions made in the Upper Palaeolithic sequence (2a to 2f), three archaeological levels have been dated and they correspond to the results expected.

OxA-5524: this Proto-Solutrean date agrees with the typology and technology of lithic tools found in the same level, that are mostly hunting tools - scrapers, quartz bladelets and flint and four perforated teeth. This phase characterized by the same lithic assemblage was dated between 21 000 and 22 000 BP by conventional 14C dating at Lagar Velho, level 6.

OxA-5523: in the final Gravettian archeological level at the top of this geological level containing the ibex bone dated, we found seven backed truncated or bi-truncated bladelets which are projectile points. This kind of microlith is typical of the final Gravettian phase, dated between 23 000 and 21 500 BP by 14C in other open air sites of the Rio Maior region and contemporary with the Proto-Magdalenian of France dated to c. 22 000 BP.

A third date was obtained from the Gif-sur-Yvette laboratory on the 2f level and a result around 26 500 BP was expected, according to associated Gravettian lithic points. The dates obtained agree with lithic assemblages found together.

OxA-5522: the sequence shows different degrees of preservation in the entrance, the middle and the rear. Although the bedrock was not found, the oldest artefacts are from the Middle Palaeolithic. Due to a lack of the Upper Palaeolithic sedimentary layers it is difficult to isolate the levels. Therefore, the position of tools of various ages could have been disturbed. Inside the group of these tools one semicircular fragment of a decorated baguette (osseous) was used for the AMS date. This tool is a Pyrennean style bone found with other artefacts attributed to the Magdalenian. The date is in accordance with the features of these tools.


Absolute chronology of the German Upper Palaeolithic series

Samples of bone, antler and ivory submitted by M. Street, Römisch-Germanisches Zentralmuseum Mainz and T. Terberger, University Greifswald, in cooperation with R. E. M. Hedges, T. F. G. Higham and P. B. Pettitt (ORAU).

German Upper Palaeolithic samples from western and central Germany previously submitted to the Oxford Radiocarbon Accelerator Unit have yielded consistent results for the Aurignacian at Wildscheuer III and Breitenbach B (Arch. List 26), the Magdalenian in the Rhineland (Arch. List 6 and 25) and the Thuringian Basin (Arch. List 25). Research by M. S. and T. T. in collaboration with ORAU on the dating of the Wiesbaden-Igstadt site unexpectedly demonstrated a human presence in the Rhineland close to the last Pleniglacial (Street and Terberger 1999; 2000; Terberger 1998; 2001; Street and Terberger in press). Following up this new evidence for human presence in Central Europe soon after the glacial maximum, three sites potentially belonging to this period have been sampled: Abri Stendel XVIII, in Lower Saxony, the Swiss site Kastelhöhle-Nord, Middle Horizon, and the burial from the Mittlere Klause, which is dealt with together with the series of further German human remains submitted by the authors.

A period of the western and central German Upper Palaeolithic almost wholly undated by modern radiometric methods is the Gravettian: the single exception is a conventional radiocarbon date of dubious context from the Magdalena Cave in the Eifel uplands (Weiß 1978). Another period for which the absolute chronology is not clear is the transition from assemblages of Upper Palaeolithic (Magdalenian) type to others assigned to the final Palaeolithic (Federmessergruppen/Azilian sensu lata) already showing many features traditionally considered to be characteristic of the Mesolithic. An exception is the double burial at Bonn-Oberkassel (map ref), previously interpreted as Middle Magdalenian (Wüller 1999), which was placed into this time range by accelerator dating at ORAU and Kiel (Street and Baales 1999).

The present samples were intended to address these periods and provide a more or less continuous radiometric time scale for the western and central German Upper Palaeolithic cultural sequence.

Wildscheuer IV

Questions of secure context and of unsuitability of material (ivory) for finds believed to originate in the Gravettian layer (IV) at the Wildscheuer cave (50:25N 8:8E), rendered the selection of samples very difficult. Only one sample was therefore submitted, a fragment of an ivory artefact (Landesmuseum Wiesbaden, S 2137), with incised cross-hatching to both of the flattened faces. The sample was pretreated at Oxford (sample code P12065) but collagen yields proved to be so low that the sample failed and could not be dated.

Mainz-Linsenberg and Sprendlingen

The Gravettian context of the material from the open air sites Mainz-Linsenberg (50:1N 8:15E) and Sprendlingen (49:52N 7:59E) is well established. The documentation of the 1921–1923 excavation at Mainz-Linsenberg (Bosinski 1995a) is quite detailed and confirms the association of bones of the selected species with the other categories of archaeological material recovered, while the Sprendlingen site was only discovered and excavated in 1978 (Bosinski 1995b; Bosinski et al. 1985). It was possible to obtain three samples from Mainz-Linsenberg (Landesmuseum 447, 451, 454) and one from Sprendlingen (Landesmuseum Sp Mz 1), which include three species of animal (horse, reindeer and rhinoceros). The samples were pretreated at Oxford (sample codes P12066-P12069). In all cases the collagen yields proved to be so low that the samples failed and were not dated.


Faunal remains and artefacts have been recovered from the site of Koblenz-Metternich (50:22N 7:34E) since at least the end of the last century (Hofer 1937; Schaaffhausen 1882, 1883). Although the major archaeological horizon recognised at the site is assigned to the Gravettian (Hahn 1969), the most recent fieldwork at this important Last Glacial section has discovered, and concentrated on, a deeper-lying, Middle Palaeolithic horizon (Conard et al. 1995).

Although the faunal material from the site was believed to be lost or destroyed, M. S. discovered faunal remains attributed to Koblenz-Metternich in both the collection of the Rheinisches Landesmuseum at Bonn and in the Mittelrhein Museum Koblenz. The Bonn collection possibly derives from the activities of H. Schaaffhausen at the end of the nineteenth century while the Koblenz finds probably represent part of the collection of A. Günther from the early twentieth century (Günther 1907). Some of the Bonn horse ribs were relatively well preserved and showed cut marks, certainly documenting human activity. The Koblenz finds form two unequal groups. One horse bone and four cervid teeth are catalogued as from the Wegelau pit in Metternich, while the exact provenance of a much larger number of bones and teeth of several species is less clear. Four samples of material, none of which appeared to have been conserved, were submitted.

OxA-10492Equus sp. bone, RGZM6/1RLMB/S378, δ13C = −21.3‰13 500 ± 90
OxA-10493Equus sp. bone, RGZM7/2RLMB/S378, δ13C = −21.3‰13 185 ± 80
OxA-10651Equus sp. bone, RGZM8/3RLMB/S378, δ13C = −21.0‰13 270 ± 180

Comments (M. S. and T. T.): OxA-10492 and −10493 are both cutmarked ribs while OxA-10651 is a phalanx. The single Koblenz specimen failed, but the three results on the Bonn samples are in good agreement, although at least 10 000 years too young for the expected Gravettian context. Since there is no reason to doubt the validity of the results two possible solutions to this problem can be proposed. Either the samples relate to a younger (and hitherto undescribed) occupation of the Metternich site, or they are not in fact from here but have at some time become associated with erroneous labelling in the Bonn museum.


Excavations at the Kastelhöhle cave in the Kaltbrunnen valley (47.26N 7:35E), north-western Switzerland, at the end of the 1940s and during the 1950s established a stratigraphy with three clearly separated cultural layers. The lowest of these was definitely Mousterian and the highest represents an Upper Magdalenian. On typological grounds it has been argued that the middle assemblage of c. 250 lithic artefacts represents an ‘earliest Magdalenian’ or ‘Badegoulian’ (Höneisen et al. 1993, 155 f; Le Tensorer and Sedlmeier 1993, 263).

J. Sedlmeier recently identified a small complex of finds with the original labelling of the excavator as belonging to the middle archaeological horizon. Three samples identified as Rangifer tarandus, including specimens with cut marks, were intended to clarify the chronological status of this level.

OxA-9737Rangifer tarandus bone, KA 1, δ13C = −18.5‰18 530 ± 150
OxA-9738Rangifer tarandus bone, KA 2, δ13C = −18.3‰19 620 ± 140
OxA-9739Rangifer tarandus bone, KA 3, δ13C = −18.4‰19 200 ± 150

Comments (M. S., T. T. and J. S.): all three determinations lie around 19 000 BP, confirming the identification of faunal material relevant to the lithic artefacts of interest, and providing credible radiocarbon results for the dating of this assemblage. The Kastelhöhle results open up interesting lines of questioning with regard to the extent and duration of the Pleniglacial desertion of northern Europe and the nature of the subsequent reoccupation of this region (Terberger 2001; Street and Terberger in press).

Abri Stendel

Prospection of rock shelters close to Göttingen (Abri Stendel is located at 51:26N 9:55E) has discovered important evidence for Mesolithic and final Palaeolithic presence (Grote 1994). A test trench between two such rock shelters uncovered a loess deposit containing a concentration of humanly modified Upper Palaeolithic bone and reindeer antler with traces of groove and splinter technique (Grote 1998; 1999), although no diagnostic lithic artefacts were excavated.

The stratigraphic position of the material and the presence of arctic microfauna suggested to the excavator that the Stendel XVIII accumulation dates to an early phase of the Late Glacial and three samples of large mammal remains were selected to test whether the finds derive from an early phase of Magdalenian expansion.

OxA-10470Equus sp. bone, RGZM11/2Lkr/AS 5/009, δ13C = −20.6‰13 105 ± 70
OxA-10471Rangifer antler, RGZM12/3Lkr/AS 5/024, δ13C = −19.8‰12 860 ± 75
OxA-10494Equus sp. bone, RGZM10/1Lkr/AS 5/021, δ13C = −20.9‰12 970 ± 70

Comments (M. S., T. T. and K. G.): the results from Abri Stendel XVIII establish the age of the humanly modified faunal remains as c. 13 000 BP and find contemporary parallels at a number of sites representing the main phase of Magdalenian expansion known from the Rhineland and Thuringia c. 13 000 BP.


It has been suggested on lithic typological grounds that Magdalenian material from the Lahn Valley caves Wildscheuer and Wildweiberlei (50:22N 7:59E) (Terberger 1993) is younger than that from the major Neuwied Basin open air sites Andernach-Martinsberg (Arch. List 6) and Gönnersdorf (Arch. List 25, 231). Uncertainties of context preclude an attempt to date the Wildscheuer V assemblage by absolute methods, but three samples of humanly modified horse bone and reindeer antler from the Wildweiberlei (Landesmusem Wiesbaden 47,3; 57,1; 74,1), where a Magdalenian level is the only one present, were submitted to test the hypothesis. The samples were pretreated at Oxford (sample codes P12077-P12079). In all cases the C/N ratios were considered unsatisfactory and the samples were not dated.

Seedorf 1995

Sample of tooth from the site Seedorf 1995, (54:01N 10:30E) Germany, submitted by K. Bokelmann, Archaologisches Landesmuseum, Schlob Gottorf, 24837 Schleswig, Germany.

OxA-6583wild pig tooth. LA 296, δ13C = −20.9‰6735 ± 65

Comment (K. B.): the dated tooth belongs to the lower jaw of a wild pig showing a geometric ornamentation which was incised just after the bone had been broken for extraction of marrow, thus showing that the “ornamentation” was incorporated in a chain of manipulations with prey in the Late Mesolithic.


Samples of bone from Kartstein, Germany, submitted by M. Baales, Forschungsbereich Altsteinzeit des Romisch-Germanisches Zentralmuseum (RGZM), Mainz, Neuwied, Germany. Comments by M. Baales, O. Jöris and B. Weninger, RGZM.

The Middle Pleistocene Kartstein travertine in the upland zone of the north Eifel region some 60 km south-west of Cologne (distr. Euskirchen) is situated 400 m asl. and is well-known for its Palaeolithic finds uncovered in two caves c. 100 years ago. In 1977, H. Löhr examined Late Glacial and Holocene deposits in front of the northern cliff of the travertine. Layer 2 below the Holocene deposits can be attributed to the Younger Dryas on the base of characteristic angular travertine frost debris, large amount of bones of several typical subarctic animal species (above all Lagopus, Rangifer, and Lemmus and Dicrostonyx), and a few distinct Ahrensburgian lithic artefacts (Ahrensburgian tanged points or Stielspitzen). The faunal remains represent a mixture of natural as well as anthropogenic death assemblages. Seasonal indicators confirm spring as the time of human occupation, when Ahrensburgian hunters would be awaiting the arrival of reindeer migrating into their upland summer stands. Lagopus remains, on the other hand, show no indications of human predation but were accumulated by bird raptors, most likely over a longer period of time within the Younger Dryas (Baales 1996). The site is dated by two AMS and seven conventional 14C determinations on animal bones (all except KN-4023 from the same metric stratum within layer 2) which were sampled by P. Pettitt (ORAU) in 1999, and by M.B. in 1989. The Cologne series was measured in the late 1980s by J. Freundlich (Baales 1996) and, more recently, in 1999 by ORAU, through the NERC-funded ORADS facility.

OxA-9031Rangifer tarandus bone, δ13C = −17.6‰10 220 ± 75
OxA-9032Lajopus lajopus bone, δ13C = −19.2‰9995 ± 65

Comment (M. B., O. J. and B. W.): even today, the site of Kartstein and the classical site of Stellmoor north of Hamburg, which was dated in the AMS laboratory in Copenhagen in the 1980s (Fischer and Tauber 1987), remain the only Ahrensburgian sites which have been extensively dated by radiocarbon. For both sites we must assume a Younger Dryas age. The minimum expected age for all reindeer bone samples would thus be c. 10 070 BP, which is the precise 14C-age for the onset of the Holocene (Stuiver and Van der Plicht 1998). The surprisingly young dates KN-4072 (9950 ± 90 BP) and KN-4074 (9530 ± 90 BP) apparently gave Lanting and van der Plicht (1996) reason to propose a Preboreal survival of Ahrensburgian tradition in the Eifel uplands and in northern Germany. However, other Ahrensburgian sites such as Remouchamps in the Belgian Ardennes or Melbeck in Lower Saxony have supplied the expected Younger Dryas 14C ages. We have long suspected the samples KN-4072 and KN-4073 to be contaminated, but until now this has remained an unproved assumption.

With OxA-9031 we have the first clear indication of a late Younger Dryas age for the Kartstein Ahrensburgian layer (although the δ13C value is surprisingly high). In comparison to the earlier dates, which were both measured on bulk samples, OxA-9031 is from a single Rangifer bone. The other determinations (all measured on bulked reindeer bones) range until c. 9500 BP while the two determinations on Lagopus bones are statistically identical within 1σ error range. Putting OxA-9031 aside, all other ages on the reindeer bones can be separated, using statistical methods, into two distinct groups. The first group has a weighted mean of 9630 ± 39 BP (KN-4072, KN-4073, KN-4254C) while the second gives a weighted average of 9965 ± 29 BP (KN-4252, KN-4254A, KN-4254B). If we calibrate these ages according to the INTCAL98 data set (Stuiver and van der Plicht 1998) the mean age of the younger group falls within the late Preboreal, while the older group dates to the very end of the mid-Preboreal radiocarbon plateau around 10 000 BP. However, we cannot simply accept these results at face value. The results contrast strongly with the sedimentological, ecological, and archaeological setting of the Kartstein Ahrensburgian find horizon.

Only c. 60 km north of the Kartstein cave, the early Mesolithic open-air site of Bedburg-Konigshoven dates undoubtably into the mid-Preboreal, reflecting totally different typical Holocene ecological conditions (Street 1998). Although the available dates on Aurochs bone spread all the way from 10 670 ± 100 BP (KN-4138) to 9740 ± 100 BP (KN-4135), at least the two determinations on wood samples from the archaeological horizon are reproducible: the age values are (KN-3998) 9600 ± 100 BP and (KN-3999) 9780 ± 100 BP, giving a weighted average of 9690 ± 71 BP. Furthermore, seven ORAU determinations on worked red deer antlers and a resin cake of the famous early Mesolithic site of Star Carr also range around 9500 BP (Mellars et al. 1998), again demonstrating a mid-Preboreal date for a site, even more to the north, with a clearly different ecological setting than Kartstein. According to the detailed biostratigraphic analysis, the initial occupation at Friesack 27a (northwest Berlin, Brandenburg) represents a typical Early Mesolithic as well. The determinations from Friesack 27a, which range even earlier in the Preboreal (Gramsch 1991 and pers. comm. 2001) are also in accordance with our model expectations, and underline our conclusion that Kartstein must have been occupied even earlier, i.e. not during the Preboreal.

On the base of this evidence, we thus conclude, the Kartstein site as well as the Stellmoor Ahrensburgian must necessarily predate all the above discussed important early Preboreal (Mesolithic) sites in the northern Lowlands. To explain the apparent 14C age discrepancies we must, therefore, call into question the chemical integrity of all the 14C dates from Kartstein (excepting OxA-9031) and most of the Stellmoor dates. An alternative explanation would be to postulate that some (presently unrecorded) changes in atmospheric 14C-levels, with extremely high-frequency and amplitude (c. 300 14C years) occurred during the Younger Dryas (Jöris and Weninger 2000). Indeed, and taking into account the still rather limited density of the available Late Glacial tree-ring, marine varve, and U/Th-coral 14C calibration data, the possibility of such a massive distortion of the radiocarbon time-scale during the Younger Dryas—although hardly to be expected—can at least not be ruled out.


Langenlois A

Sample of charcoal from a fireplacè, collected during excavations in 1961 by F. Felgenhauer at Langenlois (48:28:07N 15:41:17E), Austria. The fireplace is associated with the main concentration of artefacts (Langenlois A). A smaller cluster of artefacts (Langenlois B) is located at a distance of 25 m. The sample was made available by T. Einwögerer, St. Pölten, Austria and submitted by A. Verpoorte, Faculty of Archaeol., Univ. Leiden, Netherlands.

OxA-10266charcoal, Langenlois A-3, δ13C = −23.7‰25 700 ± 150

Comment (A. V. and T. E.): three recently obtained radiocarbon determinations, OxA-10266 and two AMS results from Groningen (GrN-25603: 25700 ± 400 BP on charcoal from the same fireplace, and GrA-16564: 25 340 ± 170 BP on ibex metatarsus), fall within 1σ of each other. OxA-10266 fits perfectly with the stratigraphic position in the Pleniglacial loess. The Gravettian assemblage contains stone artefacts, worked mammoth ivory and faunal remains dominated by ibex.


Grotta di Fumane

The Fumane Cave lies in the Lessini Mountains (Venetian Pre-Alps) at 350 m altitude, (45:25:43N 12:32:05E) and contains a fill deposit 19 m thick constituted by residual colluvial sand, stones and loess, sealed at the top by a landslide that occluded the cave entrance. The Upper Pleistocene lithostratigraphic series includes well-preserved archaeological evidence of recurrent Mousterian and Aungnacian occupations. Late Mousterian (units A6, A5, A4) and early Aurignacian (unit A2) palaeo-living floors are clearly stratigraphically distinct, the Aurignacian ones characterized by dwelling structures and various evidence (bladelet lithic and formal bone tool production, ornamental and decorated objects, paintings and ochre). The samples were submitted in the framework of an interdisciplinary research project co-ordinated by A. Broglio and M. Peresani, Dept. Sci. Geol. Paleontol., Univ. Ferrara and M. Cremaschi, Univ. of Milano I, Milan, Italy.

Several 14C determinations have been obtained from various laboratories: three AMS series of charcoal collected from Aurignacian and Mousterian (1991: UtC-1774, UtC-1775; 1992: UtC-2044, UtC-2045, UtC-2046, UtC-2047, UtC-2048, UtC-2049, UtC-2051; 1993-UtC-2688, UtC-2689, UtC-2690); one AMS series of marine mollusc shells from Aurignacian (1999: OS-5871, OS-5872, OS-5999); one conventional measurement on charcoal (Ly-9920) from the Aurignacian fire-place S16; one AMS date on charcoal (Ly-12860xA) from the Aurignacian structure S19 and one AMS measurement on collagen (GrA-16231) from the Aurignacian structure S19. Those sent to ORAU are listed below:

OxA-6462charcoal, 1/85/A4II, δ13C = −24.3‰33 150 ± 600
OxA-8021charcoal, 11/A4II, q.95, δ13C = −24.7‰33 300 ± 400
OxA-6463charcoal, 2/85, 86, 95, 96, A5, δ13C = −22.1‰33 700 ± 600
OxA-6464charcoal, 3/85, 86, 95, 96, A6, δ13C = −24.0‰34 950 ± 700
OxA-6465charcoal, 4/A2, 14b, δ13C = −23.5‰31 620 ± 500
OxA-6566charcoal, 5/A2, 514, δ13C = −23.9‰31 900 ± 1100
OxA-8050charcoal, 6/D3ba, 15, δ13C = −24.4‰30 320 ± 320
OxA-8051charcoal, 7/D3ba, 15, δ13C = −24.8‰32 020 ± 340
OxA-8052charcoal, 8/A2, 1h alto, δ13C = −22.9‰34 120 ± 460
OxA-8053charcoal, 9/A2, 1h base, δ13C = −20.6‰33 640 ± 440
OxA-8054charcoal, 10/A2, q.90, δ13C = −24.0‰33 160 ± 400
OxA-8022charcoal, 12/A5+A6, q.20, δ13C = −23.8‰38 800 ± 750
OxA-8023charcoal, 12/A5+A6, q.20, δ13C = −24.2‰38 250 ± 700
OxA-11331Sample 2, Layer A6, charcoal, δ13C = −25.5‰34400 ± 800
OxA-11346Sample 1, Layer A9, charcoal, δ13C = −22.3‰39950 ± 550
OxA-11347Sample 3, Layer A2, charcoal, δ13C = −25.2‰30650 ± 260
OxA-11348Sample 5, Layer D1d, charcoal, δ13C = −23.4‰31490 ± 250
OxA-11360Sample 4, Layer A2, charcoal, δ13C = −23.3‰31830 ± 260

Comment (A. B., M. C. and M. P.): six determinations correspond with the late Mousterian units, four to units A5-A6, and two to unit A4II. The former reveal a divergence between measurements performed in 1997 (OxA-6463 and −6464) and the results of 1998 (OxA-8022 −8023) that belong to one single charcoal divided in two parts at the laboratory. The two dates of A4II are practically coincident (OxA-6462 and −8021), and are stratigraphically congruent with the four dates from units A5-A6, although they contrast with the results of the overlying unit A2.

The complex of determinations defined for unit A2 reveals some problematic aspects in relation to the chronological acquisition of the radiometric measurements and to the intervention of various laboratories. As a whole, excluding UtC-1774, the results span the interval from 36 800 BP (UtC-2688) to 31 900 BP (OxA-6566), with an average value at 33 297 BP. Also in this case there is a divergence between the OxA-8052, −8053, −8054 sets of dates and the OxA-6565. −6466 set. Except for OxA-8054, all the samples belong to the same hearth. A further divergence between the set of dates from the Utrecht laboratory and the first set of Oxford measurements can be correlated to the spatial distribution of the samples, where UtC-1774, UtC-2044, UtC-2047, UtC-2051 record the most recent dates in the zone presently positioned outside the cave. Equally, GrN-16231 and UtC-2049 dates (the former only partially) fall within this group; the second measurement refers to unit A1, lying some centimetres above A2.

Except for OxA-8050, results from the overlying units D6 and D3b do not mark important temporal variation throughout the series.

Considered within the chronological framework depicted by other sites of northern Italy (Paina Cave in the Berici Hills and Mochi Rockshelter in Liguria), the Aurignacian appears earlier than in the Italian peninsula. This finding suggests the spread of modern humans south of the Alps, towards the western Mediterranean regions (Provence, Catalonia).



Sample of charred bone from Badanj, Herzegovina (43:05N 17:54E), submitted by G. N. Bailey, Dept. Archaeol., Newcastle Univ., together with samples from Epirus, north-west Greece (see this Arch. List), through the NERC-funded ORADS facility, to increase understanding of this site which had already produced AMS determinations in good stratigraphic order (OxA-2197: Level 6 at 12 380 ± 110; OxA-2196: Level 13 at 13 200 ± 150). Comments by R. Whallon, Museum of Anthropology, Univ. Michigan, Ann Arbor, USA.

OxA-5859charred bone, J9D4, δ13C = −21.7‰13 200 ± 100

Comment (R. W.): Level 4 is above Level 6, and a result later than 12 380 BP was expected, following on these previous results. It was estimated that it should be something between c. 10 000 BP, a guess for Level 2a/b, probably representing the local Mesolithic, and the date of 12 380 BP from Level 6. All levels from Level 3 on down are Epigravettian, although there is a typological shift allowing a division into an earlier and a later phase at about Level 6. Therefore, it was a surprise to find OxA-5859 giving a date identical to the lower Level 13, rather than closer to, if not actually later than, Level 4. There is some mixing of materials among levels, of course, and this sample is perhaps a piece of bone turned up from a lower level and intrusive into Level 4. This will become clear only if an opportunity for further dating presents itself in the future.


Iron Gates, The Balkans

Samples of human and ungulate bone from Mesolithic burials at Schela Cladovei, Romania (44:39N 22:41E), submitted 1999 by C. Bonsall, Dept. Archaeol., Univ. Edinburgh, V. Boroneant, Inst. Archaeol., Bucharest, Romania, and G. T. Cook, SUERC, East Kilbride, through the NERC-funded ORADS facility.

OxA-8502human bone, IG(D)1.1, δ13C = −18.6‰8300 ± 60
OxA-8547human hone, IG(D)2.1, δ13C = −19.3‰8240 ± 60
OxA-8548human bone, IG(D)5.1, δ13C = −18.2‰8200 ± 70
OxA-8549ungulate bone, IG(D)5.2, δ13C = −20.4‰7905 ± 60
OxA-8550ungulate bone, IG(D)5.3, δ13C = −21.2‰7805 ± 70
OxA-8579ungulate bone, IG(D)1.2, δ13C = −20.6‰7790 ± 100
OxA-8580ungulate bone, IG(D)2.2. δ13C = −20.8‰7320 ± 130
OxA-8581human bone, IG(D)3.1, δ13C = −19.5‰8330 ± 75
OxA-8582ungulate bone, IG(D)3.2, δ13C = −22.0‰7380 ± 150
OxA-8584ungulate bone, IG(D)4.2, δ13C = −21.5‰7915 ± 65
OxA-8585ungulate bone, IG(D)4.3, δ13C = −20.9‰7780 ± 75

Comment (C. B. and G. T. C): the human bone samples are from a late Mesolithic population whose dietary protein was derived mainly from freshwater fish, judging from stable isotope (δ13C, δ15N) values. The 14C results for the human bone samples are always significantly older than the associated ungulate samples by approximately 300–500 years, indicating a freshwater reservoir effect. Therefore, a reservoir correction must be made to these and previous radiocarbon results for human bones from sites in the Iron Gates (cf. ArchList 26, p. 439). The results and their implications are discussed in detail in Cook et al. (2001, 2002).


Samples of charcoal from Lapos, Prahova County (44:23N 27:28E), Romania, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, because this was a non-attributed open-air Palaeolithic site.

OxA-4779charcoal, 1/SVII/A2–0.40, δ13C = −24.5‰4620 ± 60
OxA-4780charcoal, 3/SIII/B2–0.45, δ13C = −22.8‰3225 ± 60

Comment (M. O.): there appears to have been contamination with the superimposed Neolithic layers (or the technical attribution was completely wrong).


Brynzeni I

Samples of tooth and bone from the site of Brynzeni, Kichinau (48:05N 27:15E), Moldova, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium. The dating was undertaken because the industry of Stratum 3 is of Early Upper Palaeolithic type, with foliate pieces and sidescrapers, endscrapers, burins, abruptly backed blades and an ivory amulet, but the available determinations, between 26 000 and 14 000 BP, correspond to a chronological range that is too broad.

OxA-6999Equus germanicus tooth, 1, stratum 3, δ13C = −19.4‰20 300 ± 160
OxA-7001bone, stratum 3, δ13C = −19.5‰23 400 ± 220

Comment (M. O.): the two results do not resolve the problem. The dispersion of the 14C results could be due to a mixture of distinct industries, which could explain the heterogeneity of the industry of Stratum 3 (Otte et al. 1996a).

Corpach Mas

One of two bone points of Mladec type from Corpach Mas (48:06N 28:03E), Moldova, similar to that discovered in the Aurignacian level at the nearby site of Mitoc-Malu Galben, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium because the Aurignacian attribution was not certain, due to the presence of fragments of foliate pieces and sidescrapers in the lithic industry (types absent in the Aurignacian of Mitoc).

OxA-7000bovid bone, 1, δ13C = −19.1‰24 020 ± 220

Comment (M. O.): the result is clearly posterior to those obtained for Mitoc (dated to between 31 000 and 29 000 BP), which suggests either the existence of a mixture in the industry or the existence of a late, eastern Aurignacian territory (also including Kostenki 1/3(?) and especially Climautsy II, also in Moldavia, and dated to between 20 000 and 24 800 BP) (Otte et al. 1996a).


Samples of bone from Cosautsi, nr Sorocka (48:11N 28:20E), Moldova, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium. This is an Epi-Gravettian site composed of 21 distinct archaeological levels which had not all been dated at the time that the samples considered here were selected. The goal of this series was to obtain a homogeneous dating of the entire sequence, on bone alone, chosen on the basis of their clear positioning in an archaeological level.

OxA-5233bone, 1, δ13C = −18.7‰17 900 ± 200
OxA-5234bone, 2, δ13C = −18.7‰17 900 ± 180
OxA-5235bone, 3, δ13C = −19.2‰18 000 ± 180
OxA-5236bone, 4, δ13C = −19.1‰17 840 ± 180
OxA-5237bone, 5, δ13C = −19.4‰18 000 ± 180
OxA-5238bone, 6, δ13C = −19.4‰18 060 ± 180
OxA-5247bone, 7, δ13C = −19.1‰18 140 ± 200
OxA-5248bone, 8, δ13C = −19.4‰18 780 ± 200
OxA-5249bone, 9, δ13C = −19.8‰18 940 ± 220
OxA-5250bone, 11. δ13C = −20.4‰18 980 ± 220
OxA-5251bone, 12, δ13C = −19.6‰19 060 ± 220
OxA-5252bone, 13, δ13C = −18.8‰19 060 ± 200
OxA-5253bone, 14, δ13C = −19.2‰19 080 ± 220
OxA-5254bone, 15, δ13C = −19.3‰18 980 ± 200
OxA-5255bone, 16, δ13C = −20.4‰18 860 ± 200
OxA-5256bone, 17, δ13C = −19.4‰18 560 ± 200
OxA-5257bone, 18, δ13C = −19.1‰17 840 ± 180

Comment (M. O.): the results extend very regularly between 17 900 and 19 100 BP, suggesting a dense series of occupations over a period of around 2000 years, and associated with the Last Glacial Maximum. The Epigravettian lithic industry is similar in all of the archaeological levels considered (Otte et al. 1996a).


Karain B

Samples of charcoal from Karain B, Yakca, Antalya (37:04N 30:34E), Turkey, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium to obtain a new series of dates from recent excavations (Yalçinkaya and Otte 2000).

OxA-8362charcoal, KB98-F13–22, δ13C = −24.8‰27 980 ± 240
OxA-8406charcoal, KB98-F12–16, δ13C = −24.3‰14 600 ± 100
OxA-8407charcoal, KB98-F12–18, δ13C = −23.9‰15 860 ± 100
OxA-8446bone, KB98-F13–22, δ13C = −19.3‰18 960 ± 180
OxA-9264charcoal, (1) KB99 F14/AH19 P.1, δ13C = −24.3‰15 920 ± 100
OxA-9265charcoal, (2) KB99 F14/AH18 P.II.1, δ13C = −22.5‰16 740 ± 100
OxA-9266charcoal, (3) KB99 F14/AH19 P.II.1, δ13C = −22.9‰16 750 ± 90
OxA-9267charcoal, (4) KB99 F14/AH21 P.III, δ13C = −22.5‰17 025 ± 80
OxA-9268charcoal, (5) KB99 F14/AH22 P.III, δ13C = −22.2‰18 360 ± 90

Comment (M. O.): the determinations for Layers P.1 and P.II.1 support the attribution to the Epi-Palaeolithic. Layer P.III, however, was typologically attributed to the Aurignacian, which is not supported by the two dates obtained. The geological provenance of the two samples will be evaluated during the 2000 field season. As the dates suggest early Epi-Palaeolithic, it is possible that the identification of the stratum as P.III was incorrect.


Samples of charcoal from Okuzini 93, nr. Antalya (37:04N 30:34E), Turkey, submitted by M. Otte. Préhistoire, Univ. Liège, Belgium, in order to obtain a homogeneous series of dales on charcoal for all archaeological (and geological) levels identified in the Epi-Palaeolithic sequence, in order to remedy certain inconsistencies and imprecisions in certain results previously obtained.

OxA-5175charcoal. K5d/24. δ13C = −24.2‰14 610 ± 150
OxA-5176charcoal. K5d/25, δ13C = −24.5‰14 820 ± 150
OxA-5177charcoal, K5d/26, δ13C = −24.6‰15 460 ± 160
OxA-5178charcoal. K5d/27, δ13C = −24.4‰16 420 ± 180
OxA-5179charcoal. K5d/30, δ13C = −23.9‰16 440 ± 160
OxA-5213charcoal, 18B/8, δ13C = −24.7‰10 150 ± 90
OxA-5214charcoal. 18B/9, δ13C = −24.6‰12 130 ± 100
OxA-5215charcoal. 18B/10, δ13C = −25.6‰12 500 ± 110
OxA-5216charcoal. 18B/12, δ13C = −24.9‰12 390 ± 110
OxA-5217charcoal. 18B/13, δ13C = −25.2‰12 540 ± 110
OxA-5218charcoal, 18B/14, δ13C = −24.9‰12 580 ± 110
OxA-5219charcoal, 18B/15, δ13C = −24.5‰12 700 ±110
OxA-5220charcoal, 18B/16, δ13C = −25.1‰13 060 ± 120
OxA-5221charcoal, 1 KB/17, δ13C = −25.2‰13 210 ± 120
OxA-5222charcoal, 1 KB/18, δ13C = −24.4‰14 200 ± 130
OxA-5223charcoal, 1 KB/19, δ13C = −25.1‰14 320 ± 130
OxA-5224charcoal, 18B/20, δ13C = −25.1‰14 550 ± 130
OxA-5225charcoal, 1 8B/22, δ13C = −24.6‰14 940 ± 140

Comment (M. O.): cf. Otte et al. (in press).


Buran Kaya III

Two samples of bone from Buran Kaya III, Simferopol, Crimea (45:00N 34:25E), Ukraine, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium. Crimea is located at the gates of Europe. Its study is essential for understanding the origin of the Upper Palaeolithic cultures.

OxA-6882bone, 1, δ13C = −19.6‰30 740 ± 460
OxA-6990bone, 2, δ13C = −19.0‰34 400 ± 1200

Comment (M. O.): the results correspond well to the early phases of the Aurignacian and the eastern Gravettian (for general comment on the chronology of this topic see Pettitt 1998).


Sample of bone from Mejygortsi (59:03N 24:53E), Ukraine, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, because only two radiometric dates (c. 17 500–17 800 BP) were available for this unique Gravettian site in the Ukrainian Carpathian mountains.

OxA-6882bone, 1, δ13C = −19.6‰30 740 ± 460
OxA-6990bone, 2, δ13C = −19.0‰34 400 ± 1200

Comment (M. O.): lithic and faunal comparisons of the Mejygortsi assemblage are strongest with Stratum VII of Molodova V (dated to 23 000 BP). The lack of shouldered pieces place this industy in a phase immediately posterior to Stratum VII of Molodova V, confirmed by OxA-7429 (Kulakovska and Otte 1999).

Siuren I

Samples from Siuren 1, Crimea (44:39N 33:48E), Ukraine, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, to date the old phases of the Upper Palaeolithic in Crimea.

OxA-5154bone, 2, δ13C = −19.2‰28 450 ± 600
OxA-5155bone, 3, δ13C = −19.1‰29 950 ± 700
OxA-8249bone, Siuren 98/4, δ13C = −17.8‰28 200 ± 440
OxA-6987charcoal, N2, δ13C = −27.1‰171.9 ± 2.3% mod

Comment (M. O.): OxA-5154, −5155 and −8249 are acceptable considering the archaeological provisions (Otte et al. 1996). OxA-6987 is clearly intrusive modern charcoal.


Samples of charcoal from Skalistiy, Simferopol, Crimea (44:50N 38:00E), Ukraine, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium to define the chronology of the last phase of the Upper Palaeolithic in Crimea.

OxA-5164charcoal, 1.III/2;A-gamma (5), δ13C = −25.4‰11 620 ± 110
OxA-5165charcoal, 2.III/3;(A-B)-5, δ13C = −25.1‰11 750 ± 120
OxA-5166charcoal, 3.IV:E-7, δ13C = −23.9‰14 570 ± 140
OxA-5167charcoal, 5.VI;A-B(5), δ13C = −23.4‰15 020 ± 150
OxA-5168charcoal, 6.VII;A-B(5), δ13C = −24.6‰14 880 ± 180

Comment (M. O.): the results correspond well to the assumed evolution of this tradition (Cohen et al. 1996).


A sample of bone from Viazovok (54:02N 38:10E), Ukraine, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, to date an unknown phase of the Late Palaeolithic in Crimea.

OxA-6988bone, 1, BOX, δ13C = −18.8‰11 440 ± 90

Comment (M. O.): the result confirms the attribution and provides precision for the culture.


Samerzchle KLD Cave

Sample of bone from Samerzchle KLD Cave (42:19N 43:19E), Georgia, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, from a previously undated site.

OxA-7854bone, 6E-SAMERZ 154–970:295, δ13C = −18.8‰20 160 ± 160

Comment (M. O.): the determination obtained places this industry in the Gravettian period; however, the lithic assemblage presents characteristics proper to both the Gravettian (reduction techniques, straight-backed points) and the Aurignacian (carinated burins, bone points). A mixture of the two industries cannot be excluded (Nioradze and Otte 2000).

Gwardzilas Cave

Two samples of bone from Gwardzilas Cave (42:14N 43:18E), Georgia, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, a previously undated site.

OxA-7855bone, 6E-GWARD 16–54–123, δ13C = −19.0‰15 960 ± 120
OxA-7856bone, 6E-GWARD 16–54–113, δ13C = −18.8‰15 010 ± 110

Comment (M. O.): the two determinations confirm the attribution of the laminar lithic industry (with backed points and armatures) to the beginning of the final Palaeolithic (Nioradze and Otte 2000).

Sakazhia Cave

Sample of bone from Sakazhia Cave (42:19N 42:48E), Georgia, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, a previously undated site.

OxA-7853bovid bone, 6E-SAKAZ/13–45–223, δ13C = −21.6‰11 700 ± 80

Comment (M. O.): the lithic industry is of Upper Palaeolithic type, with traces of Aurignacian and Gravettian, but the date is too recent (Nioradze and Otte 2000).

Dewischwreli Cave

A sample of bone from Dewischwreli Cave (42:19N 43:18E), Georgia, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium, a previously undated site.

OxA-8020Cervus elaphus bone, 6E-DEWIS 21–32/22, δ13C = −20.4‰10 025 ± 55

Comment (M. O.): the result confirms the attribution of this lithic industry (bladelet, with microlithic armatures) to the Late or final Upper Palaeolithic (Nioradze and Otte 2000).


Kostenki 4

A sample of bone from Kostenki 4 (50:30N 39:01 E), Russia, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium to locale the middle phase of the local Gravettian.

OxA-8310bone, δ13C = −20.1‰20290 ± 150

Comment (M. O.): the determination corresponds to what was expected.

Kostenki 1

Sample of charcoal from Kostenki 1 (50:30N 39:01E), Russia, submitted by F. Damblon, Anthropol. Préhist., Institut Royal des Sciences Naturelles de Belgique, Brussels, Belgium.

OxA-6436Picea charcoal, KOST.1-C1/K11, δ13C = −24.3‰17 200 ± 110

Comment (F. D.): with regard to the whole set of determinations obtained from Cultural layer I, the present result gives the youngest date. Obviously, it is much too young. This is confirmed by two AMS dates performed in the Groningen laboratory on two other samples of Picea charcoal (K-10 and K-13) collected by the Praslov team (GrA-5243: 24 030 + 440 −410; GrA-5244: 23 600 + 410 −400). These results correspond to the oldest dates previously obtained on bone and tooth material.

OxA-6436 also comes from well-selected Picea charcoal and the result argues in favour of pleniglacial material. Contamination by thin rootlets hidden in tracheids is not impossible in this layer situated just below the cherno/em at the top of the sequence, but the selection of the charcoal pieces led to the elimination of suspect material.

Another explanation could be a mixture with more recent intrusive (Late Glacial?) Picea charcoal but we have at present no means to prevent this possibility. Indeed a lot of krotovines were visible in the upper part of the local sequence, and their position in the excavation plane would have been hardly discernible during the archaeological work.

Malaya Syia

A sample of bone from Malaya Syia, Altaie Siberia, Russia (45.10N 40:16E), submitted by M. Otte, Préhistoire, Univ. Liège, Belgium to date the transitional phase between Middle and Upper Palaeolithic in Siberia (Russia).

OxA-6078bone, 1, δ13C = −18.2‰28 180 ± 420

Comment (M. O.): the result corresponds well to what was expected (Otte and Dervianko 1996).


Samples of bone from Shlyakh, Middle Don River (45: 10N 40:16E), Russia, submitted by M. Otte. Préhistoire, Univ. Liège, Belgium to date the transitional phase between Middle and Upper Palaeolithic.

OxA-8306bovid bone, Level 8, δ13C = −19.2‰46 300 ± 3100
OxA-8307Equus sp. bone, Level 8, δ13C = −19.4‰45 700 ± 3000
OxA-8405bovid bone, Level 8, δ13C = −18.8‰42 100 ± 1900

Comment (M. O.): the results completely confirm the technical position of the industry.


Taramsa 1

The site of Taramsa 1 (26:05N 32:41E), Egypt has also been dated with OSL by S. Stokes (Dept. Geog., Univ. Oxford). Only a selection of the dates have been published because Stokes is still working at refining the results. The youngest dates that have been obtained by OSL have been published in Vermeersch et al. (1998), Submitted by P. M. Vermeersch, Laboratorium voor Prehistoire, Redingenstraat 16 bis., B-3000, Leuven, Belgium.

OxA-4017charcoal, ME91/355, δ13C = −26.5‰18 820 ± 200
OxA-4018charcoal, ME91/405, δ13C = −23.1‰7530 ± 130
OxA-4019charcoal, ME91/424, δ13C = −25.1‰7290 ± 80
OxA-4038carbonate, ostrich eggshell, ME91/164, δ13C = 0.8‰>49 000
OxA-4039protein, ostrich eggshell, ME91/164, δ13C = −15.6‰18 480 ± 290

Comment (P. M. V.): two charcoal samples, a first (ME91/424) from 61N 50E, just above the ostrich eggshell fragment ME91/164, but still above the lowest limit of fossil rootlets, gave an AMS-date of 7290 ± 80 (OxA-4019) and a second one (ME91/405) stratigraphically a little bit lower but still above the lowest limit of fossil rootlets, gave an AMS-date of 7530 ± 130 (OxA-4018). We presume that the “upper aeolian sandlayer” belongs to a Holocene dry period. As there is clear bioturbation (rootlets at the same level) we can accept that post-depositional processes are responsible for the buried state of the charcoal. In that case it is intrusive. However, it also seems possible to accept that the two charcoal dates, which are very near one another, are coeval with the sand deposit and should, therefore, be taken as an indication that at that time a dry period gave the possibility of aeolian accumulation on the lee side of the Taramsa hill, where the trench was cut. This time period seems to correspond to a minimal human occupation of the Egyptian western desert (probably because it was too dry).

The ostrich eggshell fragments were collected at the lower limit of the fossil rootlets in the upper aeolian sand. The sample (ME91/164) was submitted for AMS dating at the Oxford facilities. R.A. Housley (ORAU) provided the following comment on the date 7 May 1993): ‘The ostrich egg shell was dated using two fractions, the carbonate and the protein. Which is the more reliable depends on a number of factors, like the presence of secondary carbonates in the deposit and the degree of diagenesis of the protein following burial. You are going to have to assess this yourself.’

Fragments of the same ostrich eggshell sample have been submitted to Gifford H. Miller, University of Colorado at Boulder. His comment (May 1, 1993) was the following:‘The sample (ME91/164) consisted of 4 large angular fragments of ostrich eggshell. All four fragments have essentially identical D/L ratios. averaging 1.017 ± 0.004. This ratio is essentially indistinguishable from the oldest Grey Lake at Bir Tarfawi from which ostrich eggshell U-series mass spectrometer dates group tightly at about 130 ka. By correlation. I expect your samples are equivalent in age. 130 ka. or isotope substage 5e. They are certainly much older than the radiocarbon age. We have AMS radiocarbon dated about 10 eggshells that were more than 50 ka old; the apparent ages range from 35 ka to >45 ka, but most gave apparent ages of about 40 to 42 ka. My opinion is that in most instances, 14C ages of ostrich eggshell that are 39 ka are probably minimum ages.’ I feel that the racemisation date of Miller is a good fit with the general stratigraphy of the site. For that reason I prefer to accept the date > 49 000 for the ostrich eggshell fragments.

Charcoal sample OxA-4017 originates from another sector of the same site situated at a depth of about 50 cm beneath the surface in an area where three charcoal patches of rare small fragments were observed. The charcoal was scattered within the human in-fillings of a Middle Palaeolithic extraction trench. It was not certain if the charcoal is coeval or intrusive. OSL dating by Stokes (Vermeersch et al. 1998) of several artefact concentrations belonging to the youngest extraction phase of the site gave a weighted average of 55.5 ± 3.7 ka. I feel that considering the other OSL dates (not yet published) the charcoal sample should he considered as intrusive. R.A. Housley (28.03.95) had pointed out: ‘It may just be a coincidence but it is worth me pointing out that OxA-4017 and −4039 are, in radiocarbon terms, identical in age. Whether this is significant will have to rest on the rest of the evidence from the site.’ I do indeed think that it is a coincidence that both dates are identical in age.

Publication of the dates will be included in Vermeersch et al. (2001).

Taramsa 2

The samples of charcoal were collected in the dump that filled a Middle Palaeolithic chert extraction pit at the site of Taramsa 2 (26:05N 32:41E) in the lower desert near Qena, Upper Egypt, submitted by P. M. Vermeersch, Laboratorium voor Préhistoire, Leuven, Belgium.

OxA-5230charcoal, ME94/19/8, δ13C = −26.1‰> 45 100
OxA-5231charcoal, ME94/19/7, δ13C = −25.7‰> 44 800

Comment (P. M. V.): the samples come from two different filling layers, which are not contemporaneous but it is unknown how much time evolved between the two fillings. It was uncertain if the samples were contemporeaneous with the chert extraction activity; intrusion was a possibility. The dating results prove that the extraction pit is indeed very old and our attribution of the recovered artefacts to the Middle Palaeolithic is sound. It is also important that the technology and typology of the recovered artefacts is somewhat peculiar within the Egyptian Middle Palaeolithic (Vermeersch and Paulissen 1997).

Taramsa 4

The site of Taramsa 4 is nearby Taramsa 2 (26:05N 32:41E) in the lower desert near Qena, Upper Egypt, submitted by P. M. Vermeersch, Laboratorium voor Préhistoire, Leuven, Belgium.

OxA-5232charcoal, ME94/18/6, δ13C = −27.3‰43 600 ± 3400

Comment (P. M. V.): here also several chert extraction pits have been excavated. Charcoal was very rare and the sample was taken from the fill of an extraction pit. Here also we were not sure that it was not intrusive. The date of 43 600 BP is too young for the Middle Palaeolithic but a comment by R.A. Housley (28.03.95), ORAU, stated the following: “It is important to bear in mind that the difference between a finite and an infinite age is probably not all that much in terms of 14C activity, and that very small amounts of modern contamination will modify a greater than’ age into a finite one. I would not, therefore, make too much of an issue over the fact that the first two samples from Taramsa 2 have given ‘greater than’ results whilst the third sample is finite. It could well be that the samples are at least c. 45 kyr old, and possibly much older. View them as minimal estimates”. I fully agree with that comment (Vermeersch and Paulissen 1997: Vermeersch et al. in prep).

South Africa

Blydefontein rockshelter

Samples of sediment from Blydefontein rockshelter (31:09:55S 25:04:00E), South Africa, submitted by B. Bousman, Center for Archaeological Studies, Southwest Texas State Univ., San Marcos, Texas, USA.

OxA-7876sediment, C150-b, δ13C = −25.0‰9400 ± 100
OxA-8530sediment, BLY-CY-Brown 2, δ13C = −21.6‰11 850 ± 150

Comment (B. B.): OxA-7876 is from the older fills at the geological site of Blydefontein Section, a terrace deposit in front of Blydefontein Rock Shelter. This sample is associated with early Holocene pollen samples that demonstrate a significantly cooler environment than at present (Bousman et al. 1988; Bousman and Scott 1994).

OxA-8530 is from the Robberg component at Blydefontein Rock Shelter. Radiocarbon determinations from Robberg occupations in most of southern Africa span the range from 18–12 kya. The Blydefontein sample falls within the younger portion of this range. This sample was selected in order to date the initial occupation at Blydefontein. This is one of the few Robberg samples dating in the Interior Plateau of southern Africa. This occupation is stratified below a Lockshoek occupation dated to 8541 BP (Bousman 1989; 1991).

Old World Neolithic and Later

Great Britain


Samples of charcoal from the double pit alignment at Thornborough, North Yorkshire (NGR SE 286788), submitted by J. Harding, Dept. Archaeol., Univ. Newcastle, through the NERC-funded ORADS facility.

OxA-11009Quercus sp. charcoal, CTX 90207,86, δ13C= −24.4‰3385 ± 38
OxA-11010Cnrylus sp. charcoal, CTX 92703,122, δ13C= −23.7‰2716 ± 37
OxA-11033Cnrylus sp. charcoal, CTX 92703,122, δ13C=-24.6‰2761 ± 35

Comment (J. H.): OxA-11009 was from a small fragment of Quercus charcoal in the post ‘void’ of a large pit to the north of the double timber avenue. It indicates that the primary construction of the monument, or the deliberate extraction of this particular upright (an event which also occurred in many of the other pits), may date to the latter part of the Early Bronze Age. This is supported by the small pottery assemblage of collared urn and food vessel in the upper fills of two of the other pits along this alignment.

OxA-11010 and −11033 were from a small fragment of Corylus charcoal in the top of a pit's recut fill. It is assumed that the recut was dug to facilitate the removal of the timber post originally inserted into the feature. The early first millennium BC date thereby provides a terminus post quern for an event which must have occurred some centuries earlier. It appears likely that the recut was left open to infill naturally.

Aves Ditch

Sample of bone from Aves Ditch, Upper Heyford (NGR SP 5185 2465), submitted by E. Sauer, Sch. Archaeol. & Ancient Hist., Univ. Leicester, through the NERC-funded ORADS facility.

OxA-10773cattle bone, 1998/trl/bl/l, δ13C= −21.7‰2480 ± 45

Comment (E. S.): the sample was taken from a cattle metatarsal; it derives from the bottom fill (context 23) of a section of an enclosure ditch which underlies the bank of Aves Ditch, a 4.2 km long straight linear earthwork in Oxfordshire. It provides a probable terminus ante quern for the enclosure and. more importantly, a firm terminus post quern for Aves Ditch.

A study of the pottery from various sections through Aves Ditch as well as other indications suggest that this earthwork is of Late Iron Age origin. If so, it would be the straightest Iron Age earthwork of comparable length anywhere in Britain and would shed significant light on the sophistication and organisation of Late Iron Age culture in this part of Britain.

The AMS determination has been successful in confirming that a Late Iron Age date is indeed possible, but would equally be compatible with the assumption of a Roman construction. Further samples need to be analysed in order to allow a decision. Interestingly, an archaeomagnetic sample from the same context was dated by Patrick Erwin to c. 500–325 BC. The question arises as to whether this indicates that the bottom fill of the enclosure ditch dates to the short period of overlap between the two independent dating methods in the later fourth century BC or whether the archaeomagnetic date is too early.

Dunaverny and Feltwell fleshhooks

Samples of wood from fleshhook/bog finds from Dunaverny, County Antrim (55:00N, 5:30W), Ireland, and Feltwell, Norfolk (NGR TL 6947 9085). Alan West of Norwich Castle Museum facilitated the sampling of the Feltwell hook. Submitted by J. Ambers, Dept. Scientific Research, Brit. Mus., London.

OxA-10004Dunaverny, Quercus wood, 56, 12–22,1, δ13C= −25.9‰2839 ± 37
OxA-10005Dunaverny, Quercus wood, 56,12–22,1, δ13C = −26.19‰2818 ± 37
OxA-10859Feltwell, wood, Carpinus NCM373.961/BMRL7213–1-Y, δ13C = −25.2‰3013 ± 36

Comment (J. A.): the results obtained for two of this rare type of object help confirm the developmental sequence for fleshhooks. The samples were in direct functional association, being wooden components of the objects, but there is the possibility of some age-offset between growth of the rings dated and manufacture/use of the assembled instruments.

The Feltwell fleshhook (Norfolk) is a simple, single-pronged and socketed example taken to be early in the sequence, a point which drew strength from its association with an early class A cauldron (Gerloff 1986). Contemporaneity with Penard metalwork had been suggested, now believed to be at its floruit between about 1275 and 1140 BC (Needham et al. 1997). The new calibrated date (1390–1120 cal BC, 95% prob.) is neatly coincident with this time bracket. It has the important additional ramification of confirming the Penard dating of early cauldrons.

The dating of the much more elaborate fleshhook from Dunaverny complete with an array of seven bird models along its length, has been the subject of much debate since its discovery in the early nineteenth century. Although recognised to belong to Bronze Age traditions early in the twentieth century, there was still a tendency to place it at the very end of the period. The dated sample (1050–900 cal BC, 95% prob.), however, now allows it to be accepted as falling earlier in the Late Bronze Age, confirming some stylistic links with Wilburton-Blackmoor phase metalwork.

Billown Down

Samples of charred seeds and pottery residue from Billown Quarry Site (NGR SC 268702), Isle of Man, submitted by T. Darvill, School of Conservation Sciences, Bournemouth Univ., under the NERC-funded ORADS facility.

OxA-10127 Hordeum vulgare charred seeds, 7/M, 607, 136, F475, δ13C = −2.3.2‰3150 ± 50
OxA-10128 Triticum dicoccum charred seeds, 9/K, C725, 276, F541, δ13C = −23.3‰70 ± 45
OxA-10140 Hordeum vulgare charred seeds, 4/1, C656, 192, F526, δ13C = −23.2‰4495 ± 40
OxA-10141 Hordeum vulgare charred seeds, 8/K, C24, 266, F541, δ13C = −25.3‰4650 ± 39
OxA-10182 Hordeum vulgare charred seeds, 3/M, C603, 138, F472, δ13C = −23.9‰4930 ± 55
OxA-10202 pottery residue, 12/K, C491, F376, δ13C = −27.8‰4600 ± 45
OxA-10203 pottery residue, 14/C214, pit, 1993, δ13C = −27.3‰4510 ± 45
OxA-10222 Hordeum vulgare charred seeds, 11/K, 491, 132, F376, δ13C = −23.2‰4575 ± 55
OxA-10244 Hordeum vulgare charred seeds, 1/M, C603, 122, F472, δ13C = −22.4‰4465 ± 45
OxA-10245 Hordeum vulgare charred seeds, 2/M, C603, 133, F472, δ13C = −23.5‰4440 ± 45
OxA-10300 Hordeum vulgare charred seeds, 10/K, 491, 132, δ13C = −23.1‰4570 ± 65
OxA-10301 Hordeum vulgare charred seeds, 13/J, C425, 177, F360, δ13C = −23.5‰3120 ± 55

Comment (T. D.): in general the results fit well with the overall stratigraphic sequence and at a smaller scale with the fill-sequences of individual features (Darvill 2001, 14–16). Three points should, however, be made.

First, all the samples submitted to the Oxford programme derive from single-year events, mainly individual cereal grains that will have grown and been harvested within one season. These determinations are generally younger than marker-dates taken from multi-year material in the form of charcoal. F526, for example, has a charcoal date of 4542–4464 BC (Beta-125767: 5680 ± 40 BP) against a cereal grain date of 3501–3350 BC (OxA-10140). Both were determined using AMS. The difference is about a thousand years and serves to provide another example of the difficulties extensively discussed by Ashmore (1998) with reference to Scottish Neolithic and later sites.

Second, the depositional nature of the contexts from which the samples derive is critically important when considering the dates. The three determinations from F376 are all very close and calibrate to the third quarter of the fourth millennium BC. All derive from what can be taken to be a short-lived event: building a fire whose ashes contain carbonized plant remains and broken pottery with carbonized residues (Darvill 1998, 13–14 and 17–18). This contrasts with F472 which is a shallow pit or scoop that stratigraphically underlies an early phase of the Neolithic enclosure ditch, although it is only partly cut by the ditch (Darvill 1998, 12). Here a sample of grain from the lowest part of the fill (OxA-10182) is significantly older than two samples from upper parts of what, in the field, appeared to be the same fill.

Third, there is the inevitable problem of intrusive material when using small samples. This is especially common in the upper fills and late recuts. In the field it is often extremely difficult to accurately gauge the likely depth of penetration of contaminating material. The two dates for F541 illustrate the point, where a barley grain from the lowest fill dates to the mid fourth millennium BC (OxA-10141) while a seemingly intact horizon immediately above yielded what appeared to be an ancient grain of wheat that has been dated as modern (OxA-10128). Recognizing such contamination is still more difficult when it took place in ancient times as the presence of Middle Bronze Age cereals in F360 (OxA-10301) and F475 (OxA-10127) show.

Overall, the new set of determinations establishes some important horizons for the Manx Neolithic in particular and the earlier Neolithic of Atlantic Europe in general. Direct dating of early cereals is rare in the British Isles yet fundamental to an understanding of Neolithicization; considering the point raised above in relation to dating cereals by association with mixed samples, hypotheses based only on indirectly dated samples must be treated with extreme care. Thus the certain presence of barley at Billown from at least the early part of the fourth millennium BC firmly establishes the existence of an agricultural element to the economy of these island communities as early as other parts of the British Isles. The two determinations relating to decorated pottery (OxA-10203 and −10202) are also important as they provide the first direct dates for bowl-style ceramics from the Island, and clearly place this tradition in the later fourth millennium BC, earlier than the well-known essentially third millennium BC Ronaldsway ceramics (Burrow and Darvill 1997).

Findhorn, Moray

Fragment of non-oak charcoal from pyre material in pit fill of a Cordoned Urn burial associated with faience beads, found during work to a house at No. 102 Findhorn (NGR NJ 03976443), Moray, Scotland. Excavated in 1988 (Shepherd and Shepherd forthcoming). Submitted 1996, on behalf of J. A. Sheridan, National Museums of Scotland, through the offices of S. Needham, J. Ambers and C. Cartwright of the Brit. Mus., London.

OxA-7622charcoal, NMS X.EQ 1003, δ13C = −24.8‰3410 ± 50

Comment (J. A. S): the cremated remains of a young adult female and a late-term foetus or neonate were associated with 24 or 25 faience beads, of segmented, quoit and star shape, the largest single faience find in Scotland. The date of 1880–1520 cal BC is in line with other dates for faience in Britain and Ireland, e.g. 3450 ± 50 BP (GU-3260, 1890–1620 cal BC) from Stoneyburn, Lanarkshire (Banks 1995) and 3360 ± 75 BP (OxA-3550, 1780–1450 cal BC) from Eagleston flat, Derbyshire (Barnatt 1994).

Seathwaite Valley

All three samples were collected from a section of peat (70 cm thick) overlain by colluvium and exposed by the River Derwent c. 1 km south of Seathwaite Farm in the upper reaches of Borrowdale, Cumbria (NGR NY 235111), submitted by D. E. Anderson, Eton College. Windsor, Berks.

OxA-7749seeds, SW1, δ13C = −25.8‰150 ± 40
OxA-7750seeds, SW2, δ13C = −27.7‰520 ± 40
OxA-7751seeds, SW3, δ13C = −27.1‰535 ± 45

Comment (D. E. A.): OxA-7749 was extracted from the top of the peat section (at 1–2 cm depth) close to the upper contact with the colluvial fan. The aim of the sample was to obtain a maximum age estimate for the landslide that buried the peat deposit, thereby helping to infer when slopes became unstable in the valley. This information contributes to determining when human activity became intense enough in the area to significantly increase rates of hill slope erosion. The date is younger than expected, and given the degree of soil formation on top of the landslide, it is likely that the actual date of the event is towards the older end of the 2σ confidence range (1663–1955 cal AD).

OxA-7750 was extracted from 33–34 cm depth in the peat section, marked an increase in Plantago lanceolata pollen and other ruderal pollen-types. The purpose of this sample was to determine the timing of a substantial phase of woodland clearance in the valley for pastoral agriculture. This phase was expected to represent the Norse period of settlement around 1000 AD, however the radiocarbon date places this event within Medieval times (c. 1326–1447 cal AD at 95% prob.).

OxA-7751 was extracted from 61–62 cm depth, near the base of the peat section, with the intention of obtaining an age estimate for the formation of the peat. The date is younger than expected, suggesting that the entire peat deposit dates from the Medieval period.

Subsequent to sampling for pollen analysis, a large quantity of worked wood was found embedded in the peat section, and English Heritage funded an excavation of the site. Analysis of the wood and radiocarbon dates on selected worked timbers (AA-27747: 325 ± 60 BP and AA-27748: 395 ± 40 BP) have revealed late Medieval fencing and managed hedge (Wild et al. in press).

Pollen data for the site are published in Parker et al. (1994). OxA-7749 to −7751 are presented and discussed in Wild et al. (in press).

Mortuary Roll

The aim of this project was to date a wooden roller attached to a fifteenth century mortuary roll (Latin MS 114) in the John Rylands Library in Manchester, to find out whether or not the roller was contemporary with the mortuary roll itself. Submitted by S. Boldrick, The John Rylands Library, 150 Deansgate, Manchester.

OxA-7749seeds, SW1, δ13C = −25.8‰150 ± 40
OxA-7750seeds, SW2, δ13C = −27.7‰520 ± 40
OxA-7751seeds, SW3, δ13C = −27.1‰535 ± 45

Comment (S. B.): OxA-7787 confirms specialists’ assumptions that the roll holder was produced at about the same time as the precisely-dated manuscript. A small portion of the roll holder was subjected to radiocarbon dating tests in 1998. According to P. Pettitt (ORAU), the roll holder most probably dates to between AD 1430 and 1520 at 95% probability.

Abingdon Vineyard

Sample from Abingdon Vineyard (NGR SU 49909724), Oxon, submitted by T. Allen, Oxford Archaeol. Unit.

OxA-4517seeds, outer ditch, ABVR 9637, δ13C = −26.7‰2140 ± 55
OxA-4518wooden twigs, inner ditch, ABVR 4612, δ13C = −25.8‰2080 ± 55

Comment (T. A.): OxA-4517 (cal 365–10 BC) and 4518 (cal 337–326 BC or 200 BC–AD 65) come from waterlogged deposits at the base of two large parallel ditches found during excavations at Abingdon, Oxfordshire by the Oxford Archaeological Unit in 1991 and 1993. Both ditches were 10 m or more wide and approximately 2.5 m deep, and are interpreted as the defensive diches surrounding a large settlement of Late Iron Age and Early Roman date found during excavations in 1989 and 1990. A third parallel ditch, slightly smaller, was also found between these two. Mid-late first century AD pottery came from the middle ditch fills, and the dating was undertaken to establish when the ditches were first cut.

OxA-4517 came from the outermost ditch, and OxA-4518 from the inner ditch. Two further samples from ditches, one a cattle bone, the other a fragment of antler, were submitted for conventional dating to the Cambridge laboratory, and the following determinations were obtained: context 9638, Q-2354: 1935 ± 30 BP (5–120 cal AD); context 4612, Q-2353: 2015 ± 35 BP (cal 105 BC–AD 5). All of the four dates centre between the last two centuries BC and the first century AD, and are thus within the expected range.

Saddler's Cottage

Sample of bone from Saddler's Cottage, Medbourne (NGR SP 480294), England, submitted by J. Lucas, Assistant Keeper, Jewry Wall Museum, St. Nicholas Circle, Leicester.

OxA-7144Equus asinus bone, A134,1994, δ13C = −21.6‰180 ± 35

Comment (J. L.): the donkey bone was submitted as it was found in a Roman deposit, which our bone expert told us was very unusual. As the excavation team concerned was a training unit doing a salvage excavation in building trenches, and the context was in very close proximity to some deep cutting later intrusions, we were not confident of identifying a rare sighting of a Roman donkey and so had it carbon dated. The result indicated that we were right not to be confident.

Bronze Age Metalwork Programme (DOB Series)

Forty-six samples were dated in this programme, which has been comprehensively published in Needham et al. (1997); full contextual details and discussion of the significance of the results is to be found there. A further discussion paper has appeared in Hedges et al. (1999). The overall aim was to confirm or revise current perceptions on the absolute chronologies of Middle and Late Bronze Age metalwork assemblages. Samples chosen came from organic remains in direct physical contact with diagnostic types of bronze implement. In the great majority of cases the organics were seen to be functional parts of the original implement—hafts, shafts, handles, linings. Species of these organic materials were identified and growth stages assessed. Submitted by S. Needham, Dept. Prehistoric and Romano-British Antiquities, The British Museum, London.

OxA-4502Kingston-on-Thames, Pomoideae wood, WC 1673, δ13C = −25.8‰2925 ± 50
OxA-4503Battersea/Chelsea, Acer/Corylus wood, 1854 11–16 1, δ13C = −26.1‰2930 ± 55
OxA-4504Taplow, Thames, Fraxinus sp. wood. 1903 6–23 1, δ13C = −24.5‰2965 ± 65
OxA-4505Woodbastwick, Pomoideae charcoal, SMR 29362, δ13C = −25.1‰2865 ± 45
OxA-4506Near Windsor, Pomoideae wood, 1896 1–21 1, δ13C = −23.6‰2825 ± 50
OxA-4651Thetford, Fraxinus sp. wood, DOB 2, δ13C = −24.9‰3220 ± 80
OxA-4652Bodwrog, Wales, Salix wood, DOB 3, δ13C = −26.7‰2720 ± 45
OxA-4653Chertsey, Quercus sp. wood, DOB 4, δ13C = −27.3‰2910 ± 55
OxA-4654Nore Hill, Hampshire, Quercus sp. wood, DOB 5, δ13C = −23.9‰2765 ± 45
OxA-4655Shepperton, Pomoideae wood, DOB 7, δ13C = −27.3‰2760 ± 50
OxA-4656Thames, Fraxinus sp. wood, DOB 11, δ13C = −24.7‰3005 ± 75
OxA-4657Little Bentley, Alnus sp. wood, DOB 15, δ13C = −24.2‰525 ± 50
OxA-4658Kew Deer Pk, Thames, Salix/Populus wood, DOB 17, δ13C = −26.1‰2545 ± 55
OxA-4714Methwold, Whiteplot, Taxus baccata wood, DOB 1, δ13C = −27.6‰3995 ± 75
OxA-4715Thirsk, Yorkshire, Junus sp. wood, DOB 6, δ13C = −21.6‰455 ± 45
OxA-4716Thetford, Melford Common, Fraxinus sp. wood, DOB 16, δ13C = −26.1‰2780 ± 50
OxA-5183Amerden, Thames, Fraxinus sp. wood, DOB 22, δ13C = −26.2‰2930 ± 40
OxA-5184Blackmoor, Hants, Fraxinus sp. wood, DOB 23, δ13C = −25.7‰2830 ± 65
OxA-5185Blackmoor, Hants, Quercus sp. wood, DOB 24, δ13C = −26.1‰2770 ± 50
OxA-5186Blackmoor, Hants, Fraxinus sp. wood. DOB 25, δ13C = −25.4‰2840 ± 40
OxA-5187Meadow Lane, St. Ives, Fraxinus sp. wood, DOB 26, δ13C = −27.l‰3045 ± 55
OxA-5188Blackburgh Pit, Fraxinus sp. wood, DOB 27, δ13C = −23.4‰2780 ± 35
OxA-5196Datchet, Thames, Fraxinus sp. wood, DOB 19, δ13C = −25.0‰3035 ± 40
OxA-5197Thames, Fraxinus sp. wood, DOB 20, δ13C = −25.l‰2910 ± 50
OxA-5198Nr, Teddington, Thames, Fraxinus sp. wood, DOB 21, δ13C = −25.4‰2820 ± 70
OxA-5948Mortlake, Thames, wood, A.11811, δ13C = −25.0‰3225 ± 65
OxA-5949Strand nr Kew, Thames, wood, 0.1456, δ13C = −26.2‰3110 ± 50
OxA-5950Strand nr Kew, Thames, wood, 0.1218, δ13C = −26.5‰2910 ± 45
OxA-5951Barnes, Thames, wood, A. 17108, δ13C = −25.7‰2980 ± 45
OxA-5952Mortlake, Thames, wood. 0.1442, δ13C = −28.7‰2965 ± 45
OxA-5953Isleworth, Thames, wood, 0.1440, δ13C = −24.6‰3015 ± 45
OxA-5954Hammersmith, Thames, wood, 0.1413, δ13C = −24.8‰3025 ± 55
OxA-5955Nr. Hampton Court, Thames, wood, A.27215, δ13C = −26.6‰2900 ± 45
OxA-5956Thames, Staines, wood, A.6503, δ13C = −29.3‰2850 ± 50
OxA-5957Staines, Thames, wood. A. 10957, δ13C = −23.3‰2810 ± 45
OxA-5958Ruskington, wood, δ13C = −25.1‰840 ± 40
OxA-5959Fengate Power Station, wood, 24, δ13C = −25.2‰2965 ± 45
OxA-5960Fengate Power Station, wood. 33, δ13C = −27.0‰3230 ± 45
OxA-5962Birchington, wood, δ13C = −26.7‰2685 ± 60
OxA-5976flag Fen, wood, 93, δ13C = −25.4‰2740 ± 45
OxA-5977Feltwell, Whiteplot, wood, NCM 5169, δ13C = −21.3‰2620 ± 45
OxA-6177Thames at London, wood, DOB 47, (0.1447), δ13C = −25.4‰3055 ± 50
OxA-6216Buscot axehead, wood, δ13C = −25.l‰2480 ± 50

Comment (S. P. N.): the results were largely very coherent, but have indicated the need to shift the absolute chronologies of metalwork assemblages by varying amounts (Zürn and Schiek 1969; Needham et al. 1997; Hedges et al. 1999). A small number of determinations were unsatisfactory, but in virtually every case the organics concerned, although physically associated with the bronze, were not in direct functional association.

Comment: This series of determinations has been further analysed using Bayesian statistics (implemented in OxCal), see Needham et al. (1997).

Cefn Drum

A sample of charcoal found in a pit in front of a newly-discovered passage grave on Cefn Drum, near Pontlliw, Swansea (NGR SN 613045), submitted by J. Kissock and R. A. S. Johnston, Res. Centre for the Study of Culture, Archaeol., Religions & Biogeog., Univ. Wales College, Newport.

OxA-10056charcoal, LH1/F24, δ13C = −24.7‰859 ± 34

Comment (J. A. K. and R. A. S. J.): excavation continued on Cefn Drum in 2001 and. in addition to a sheepcote of post-medieval date, a passage grave was excavated. Prior to excavation this was thought to have been a sunken shelter for the storage of commodities produced at the sheepcote. Samples of charcoal attached to bone were recovered from a pit lying immediately in front of the passage and between the horns. The pit was, however, stratigraphically later than the grave itself. The discovery of possible human bone was reported to the coroner's officer in the usual manner. At this point police visited the site and required that excavation cease as the discovery of partial remains of the skeleton of a murder victim had been made in the vicinity in 1996. Radiocarbon dating of the bone was requested in order to determine whether or not this was more of the body. The date range of the middle eleventh to early twelfth century demonstrates that this is clearly not so.

This date creates a further problem. The 14C determination is inconsistent with the date which might be expected for a passage grave. However, the date is broadly contemporary with medieval house platforms which exist elsewhere on Cefn Drum. It should also be noted that another prehistoric burial site at Pentre Farm, around 3 kms south-west of the Cefn Drum excavations, has produced material which, upon radiocarbon dating, gave a medieval result.

Fishers Road East

The aim of the radiocarbon programme was to directly charred cereals found in the 1995 excavations at the late pre-Roman and Roman Iron Age settlement at Wshers Road East, Port Seton, East Lothian, Scotland (NGR NT 40927540), thus contributing to current debates concerning the timing of farming innovations north of the Tyne. The co-occurrence of spelt and emmer wheat in the same contexts implies that the settlement was occupied at the period when spelt had started to replace emmer on lowland sites, while the apparent association of naked and hulled barley at a period some centuries after the latter supposedly replaced the former is equally notable. Submitted by C. C. Haselgrove and J. P. Huntley, Dept. Archaeol., Univ. Durham, supported by the NERC-funded ORADS facility.

OxA-7399charred seeds, hulled barley, 3/89, δ13C= −24.4‰2210 ± 40
OxA-7400charred seeds, naked barley, 4/89, δ13C = −24.8‰2110 ± 40
OxA-7401charred seeds, spelt glume, 5/103, δ13C = −24.2‰2235 ± 35
OxA-7402charred seeds, emmer glume. 6/103, δ13C = −23.8‰2230 ± 35
OxA-7518charred seeds, hulled barley, 1/101, δ13C = −23.4‰2030 ± 50
OxA-7519charred seeds, naked barley, 2/101, δ13C = −22.9‰2075 ± 50
OxA-7520charred seeds, spelt glume, 7/105, δ13C = −25.6‰2045 ± 55
OxA-7521charred seeds, spelt glume, 7/105, δ13C = −25.2‰2075 ± 55
OxA-7522charred seeds, emmer glume, 8/105, δ13C = −24.9‰2030 ± 50
OxA-7523charred seeds, hulled barley, 9/67, δ13C = −23.2‰1970 ± 50
OxA-7524charred seeds, naked barley, 10/67, δ13C = −23.0‰1910 ± 65
OxA-7525charred seeds, spelt glume, 11/815, δ13C = −23.4‰1955 ± 50
OxA-7526charred seeds, emmer glume, 12/815, δ13C = −25.4‰1955 ± 50
OxA-7527charred seeds, spelt glume, 13/822, δ13C = −25.3‰2055 ± 55
OxA-7528charred seeds, emmer glume, 14/822, δ13C = −25.5‰1945 ± 55
OxA-7529charred seeds, bread wheat, 15/413, δ13C = −23.3‰1130 ± 50

Comment (C. H. and J. H.): the radiocarbon determinations from spelt and emmer samples show that both wheats were in use together over the period c. 400 cal BC–200 cal AD. Spelt wheat, however, increased significantly in importance within this period; the earliest context dated (103, OxA-7401 to −7402). which predates the main occupation, had the highest proportion of emmer of any carbonised assemblage from the site, whereas spelt outnumbers emmer in most of the latest deposits (e.g OxA-7525 to −7526). The radiocarbon determinations similarly indicate that both hulled and naked barley were grown throughout the lifetime of the site, although in this case it is clear that hulled barley was the main crop throughout.

A single grain of bread wheat, apparently from an undisturbed Iron Age context, was also dated (OxA-7529) in order to test the hypothesis that the inhabitants of Fishers Road East may have been experimenting with bread wheat at an early date, as has been observed at other Iron Age sites in south-west Scotland and north-east England. The seed, however, yielded a date in the early historic period, indicating that it was in fact intrusive from later agricultural activity on the site, present in the form of cultivation ridges like the one which sealed the deposit in which the bread wheat was found.


Samples of charcoal from Broom, Warwickshire (NGR SP 08465377), submitted by S. Palmer, Field Archaeology Office, Warwickshire Mus., Warwick.

OxA-6282Quercus charcoal, SA93C141 (821/1/1), δ13C = −23.6‰2570 ± 55
OxA-6283Quercus charcoal, SA93C142 (821/1/1), δ13C = −23.4‰2475 ± 55
OxA-6284hazelnut seeds, SA93C143 (822/2/1), δ13C = −24.1‰4195 ± 55
OxA-6285hazelnut seeds, SA93C144 (833/1/1), δ13C = −26.2‰4215 ± 55

Comment: (S. P.): these dates are all satisfactory and are now published in Palmer (2000).

Ling Hall

Samples of charcoal from Ling Hall, Warwickshire (NGR SP 45007800), and from Ling Hall Quarry, England (NGR SP 44957330), submitted by S. Palmer, Field Archaeology Office, Warwickshire Mus., Warwick.

OxA-6393Ling Hall, birch/hazel charcoal, LH94C141 (339/1/1), δ13C = −26.0‰2590 ± 60
OxA-6394Ling Hall, alder/hazel charcoal, LH94C142 (324/2/1), δ13C = −27.2‰2505 ± 60
OxA-8428Ling Hall Quarry, Betula charcoal LH97C143(1068/1/1), δ13C = −25.3‰2635 ± 50
OxA-8429Ling Hall Quarry, charcoal, LH97C144(1026/1/1). δ13C = −27.0‰1190 ± 40

Comment (S. P.): these two pairs are from the same site, albeit two independent posthole alignments. OxA-6393, −6394 and −8428 make sense as a suite but OxA-8429 must surely be anomalous. It is not yet certain how the dated samples relate to the actual posts in the alignments but it is (stratigraphically) certain that the posts were erected prior to a pit alignment that pre-dates a Middle Iron Age enclosure. The quarry site contains a complex of boundary and enclosure features that have been excavated gradually since 1990. Work will shortly be undertaken in an area where many of the alignments meet and it is hoped that stratigraphic proof will be forthcoming. Samples have already have obtained from some of the alignments but these will not be submitted until the stratigraphic situation is resolved.

Oakley Causeway Timbers

Four samples of oak were selected from an undated causeway through peat, which lead towards the river Waveney, along the Norfolk/Suffolk border at Scole (NGR TM 14737850). Submitted by A. Tester, Suffolk Archaeol, Unit, Bury St. Edmunds, Suffolk.

OxA-5064oak wood, OKY005 82086, δ13C = −28.2‰1495 ± 65
OxA-5978oak wood, OKY005, 80343, δ13C = −26.4‰1625 ± 40
OxA-5979oak wood, OKY005, 80499, δ13C = −25.4‰1435 ± 35
OxA-5980oak wood, OKY005, 80519, δ13C = −27.0‰1965 ± 50

Comment (A. T.): the site is c. 100 m from the river crossing of an important Roman road which ran between Colchester and Caistor St. Edmund, and a Roman ‘small town’ which was centred to the north of the river.

OxA-5980 was slightly separate and stratigraphically the earliest of the four timbers. It is argued that this was part of a Late Iron Age causeway which led to a natural fording point of the river. This causeway probably fell out of use during the Roman period following the construction of a bridge.

The three later results suggest that following the abandonment of the Roman settlement, and perhaps the collapse of the bridge at the end of the 4th century, a ford was re-established along the earlier alignment. The site probably had natural advantages and it is unlikely that there was continuity of use.

Crankley Lane, Easingwold

Samples from Crankley Lane, Easingwold (NGR SE 52156825), Yorks, submitted by A. R. Hall, Envir. Archaeol, Unit, Univ. York, Heslington, York. Comments by M. Whyman, York Archaeol, Trust for Excavation and Res. Ltd., York.

OxA-5379wood, 110A, 1354, δ13C = −25.8‰855 ± 45
OxA-5380wood, 110B, 1354, δ13C = −26.6‰870 ± 45
OxA-5381plant remains, 151A, 1411, δ13C = −30.1‰170 ± 40
OxA-5382plant remains, 151B, 1411, δ13C = −28.7‰160 ± 40
OxA-5383Sambucus nigra seeds, 157A, 1443, δ13C = −25.2‰1960 ± 45
OxA-5384Sumbucus nigra seeds, 157B, 1443, δ13C = −25.9‰1905 ± 45
OxA-5385wood, ?oak, 164A, 1230, δ13C = −25.2‰1895 ± 45
OxA-5386wood, ?oak, 164B, 1230, δ13C = −25.2‰1890 ± 45

Comment (M. W.): all the samples were taken from the excavation of a settlement site which produced quantities of pottery, and a small number of other artefacts, apparently indicative of a late pre-Roman Iron Age date. The site is located on low-lying sands and gravels in the Vale of York, some 10 miles north of the city of York itself. OxA-5383, −5384, −5385 and −5386 were taken from the backfills of two natural-cut linear features, probably palisade slots, and confirm the occupation of the site in the first centuries BC/AD. The first pair is of particular significance in that the material was stratified in direct association with an assemblage of crucibles used in bronze-casting.

OxA-5379 and −5380 were taken in order to confirm a similar date for a natural-cut bowl furnace associated with a quantity of iron smelting slag. That it did not do so, giving a date of c. 1000 years later, has caused some problems in the interpretation of the site! The 14C determination is confirmed by an archaeomagnetic determination, obtained from the scorched base of the feature from which these samples were taken, which returned a date of AD 1160–1190 for the last firing.

OxA-5381 and −5382 were samples of plant material taken from a massive ditch, which seemed likely to be of medieval or even modern date. The radiocarbon determinations confirmed the latter.

Glastonbury Lake Village

Samples of bone and antler from Glastonbury Lake Village (NGR ST 492407), Somerset, submitted by S. Minnitt, Somerset County Museums Service, Taunton Castle, Taunton, under the NERC-funded ORADS facility.

OxA-5379wood, 110A, 1354, δ13C = −25.8‰855 ± 45
OxA-5380wood, 110B, 1354, δ13C = −26.6‰870 ± 45
OxA-5381plant remains, 151A, 1411, δ13C = −30.1‰170 ± 40
OxA-5382plant remains, 151B, 1411, δ13C = −28.7‰160 ± 40
OxA-5383Sambucus nigra seeds, 157A, 1443, δ13C = −25.2‰1960 ± 45
OxA-5384Sumbucus nigra seeds, 157B, 1443, δ13C = −25.9‰1905 ± 45
OxA-5385wood, ?oak, 164A, 1230, δ13C = −25.2‰1895 ± 45
OxA-5386wood, ?oak, 164B, 1230, δ13C = −25.2‰1890 ± 45

Comment (S. M.): Glastonbury Lake Village was excavated by A. Bulleid and H. St. George Gray in the period 1892–1907. In the 1990s a major re-evaluation of the structural evidence and the artefacts was carried out by J. Coles and S. Minnitt, with assistance from a number of specialists. The initial aim was to try and establish the absolute dates for the duration of the settlement which up until then had depended upon artefact typology. Secondly, analysis of the structural evidence resulted in a proposed four-phase sequence for the settlement. It was hoped that AMS determinations would test this hypothesis and perhaps provide absolute dates for the phases. In the event, the determinations did not refine the dating of the site, because the calibrated ranges were too wide. Artefact typology continued to provide greater precision. The results were published in Coles and Minnitt (1995).

Ogden Down

Samples from Ogden Down, Dorset (NGR ST 97101290), submitted by R. Bradley, Dept. Archaeol., Univ. Reading, through the NERC-funded ORADS facility.

OxA-4519red deer antler, 1/T.3, 4, S.22, δ13C = −23.8‰2820 ± 65
OxA-4520Bos skull bone, 2/T.6, 3/4, δ13C = −24.1‰4010 ± 55
OxA-4555red deer antler, 1/T.3, 4, S.22, δ13C = −21.6‰2810 ± 70
OxA-5125oak charcoal, OGD 3 PH 102, L.2, δ13C = −26.2‰2870 ± 50

Comment: (R. B.): the site is a complex ring ditch close to the Dorset Cursus. The ring ditch is approached by an avenue of paired posts and enclosed within a timber circle. OxA-4520 suggests that, like other sites close to the Cursus, the monument originated in the Late Neolithic period, when the Bos skull was deposited in its ditch. That ditch was substantially recut during the Late Bronze Age: an episode dated by OxA-4519 and −4555. The post settings were apparently constructed during the same phase, and this is indicated by OxA-5125. These results provide unusual evidence for the re-use of a Late Neolithic monument during the Late Bronze Age and a chronological context for the post settings, which have few parallels within the British Isles.

Bollihope Common

This sample of lead slag, containing charcoal, was taken from Bollihope Common (NGR NY 981353) in County Durham, submitted by M. Manchester, Vine House, Headley Hope, Bishop Auckland, C. Durham. The radiocarbon dating was funded by the Tony Clark Memorial Fund for Archaeological Science, through the Royal Archaeological Institute.

OxA-10094charcoal, lead slag, 1, δ13C = −24.0‰1055 ± 50

Comment (M. M.): this date is particularly interesting because it predates the earliest historical records of lead extraction in the county, and Bollihope Common is now the earliest known lead-working site in the region. Not much is known about early lead-working in the Northern Pennines. There are different types of furnaces and slags to be found but little work had been done on dating them. It is hoped that this determination will assist further research into when and how lead was worked.

Yarnton Saxon: Cresswell Field settlement

Samples of bone and plant remains from the site Yarnton Cresswell Field. (NGR SP 474109), Yarnton. Oxon, submitted by G. Hey, Oxford Archaeol. Unit, Oxford, through English Heritage.

OxA-6448bone, YCF95 7972(same as 7326)(b), δ13C = −22.0‰1300 ± 50
OxA-6449plant remains, cereal, YCF95 8564(a), δ13C = −24.7‰1140 ± 50
OxA-6453plant remains, cereal, YCF95 8576(a), δ13C = −25.8‰3125 ± 55
OxA-6456plant remains, cereal, YCF95 8586(a), δ13C = −22.9‰1255 ± 50
OxA-6556Pomiodeae charcoal, YCF95 8586(b), δ13C = −25.7‰1230 ± 50
OxA-6825cattle bone, YCF95 7396(a), δ13C = −21.0‰1520 ± 55
OxA-6826cattle bone, YCF95 7396(b), δ13C = −21.1‰1425 ± 55
OxA-6827bone, YCF95 7326 (a), δ13C = −20.9‰1435 ± 55
OxA-6828bone, YCF95 7129, δ13C = −21.2‰1555 ± 55
OxA-6829Prunus sp. charcoal, YCF95 8564(b), δ13C = −24.7‰2850 ± 55
OxA-6830Prunus sp. charcoal. YCF95 8576(b), δ13C = −24.8‰3410 ± 60
OxA-7372animal bone, YCF95 7129, δ13C = −21.5‰1305 ± 50

Comment (G. H.): the radiocarbon results are in statistical agreement with the hypothesis that occupation on this site was uninterrupted. It seems likely that the features dated are not all contemporary, and that one of the sunken-featured buildings was earlier than the other structures. This structure appears to be contemporary with the early Saxon phase of occupation on the Yarnton Saxon site. Other features have a similar date range to the earliest phase of middle Saxon settlement at Yarnton.

Yarnton Saxon: floodplain channels

Samples of seeds and plant remains from the site Yarnton Saxon floodplain channels, Section B (NGR SP 47361084), Yarnton, Oxon, submitted by G. Hey, Oxford Archaeol. Unit, Oxford, through English Heritage.

OxA-4816seeds, OSL2 957c, δ13C = −27.6‰1835 ± 55
OxA-7361plant remains, 6a, δ13C = −25.6‰1640 ± 50
OxA-7362plant remains, 6b, δ13C = −27.0‰1730 ± 50
OxA-7363plant remains, 7, δ13C = −26.5‰1390 ± 50
OxA-7364plant remains, 8, δ13C = −26.5‰1155 ± 50

Comment (G. H.): the introduction of large quantities of mineral sediment in levels 5 and 6 appear to date to the Roman period. The cessation of this alluviation probably reflects the ending of intensive arable cultivation at the end of the Roman period. The organic deposits, which built up above these sediments are early and middle Saxon in date. The deposition of inorganic alluvium resumed in the later Saxon period, perhaps indicating renewed emphasis on arable agriculture.

Over this period of time the evidence from pollen, invertebrate and waterlogged plant remains suggests that the floodplain was used for pasture throughout. There is no suggestion of woodland regeneration, not even the presence of scrub.

Roman settlement at Yarnton has been excavated immediately to the west of the Saxon site and fields on the floodplain would have been farmed from here. The deposition of organic material correlates with Phases 1 and 2 of the Yarnton Saxon settlement. Although the evidence for early Saxon occupation is slight, it is apparent that there was continued grazing on the floodplain. The onset of alluviation in the later Saxon period is more difficult to relate to particular phases of occupation on the gravel terrace because the dating evidence is too imprecise.

The results of the OSL dating programme are in excellent agreement with the independent sequence provided by the stratigraphy and the absolute dating evidence provided by radiocarbon. This method of dating appears to be highly suitable for slowly deposited floodplain sediments, and provides a method for directly dating the deposition of sediments (especially where organic preservation is poor and few samples are available for radiocarbon), but care must be taken for layers, which may have been deposited very rapidly.

Yarnton Saxon: Oxey Mead channel

Samples of plant remains from the site, Yarnton Saxon channels (NGR SP 47901095), Yarnton, Oxon, submitted by G. Hey, Oxford Archaeol. Unit, through English Heritage.

OxA-7359plant remains, 3/5, δ13C = −25.6‰1300 ±65
OxA-7360plant remains, 3/7, δ13C = −24.5‰1675 ± 55

Comment (G. H.): the objectives of the dating programme were met in that it provided a date for the introduction of the hay meadow at cal AD 650–850. This is most likely to be contemporary with the middle Saxon settlement at Yarnton.

Yarnton Saxon: Worton settlement

Samples of charred seeds from the site Cassington Worton Rectory Farm (NGR SP 45951116), Yarnton, Oxon, submitted by G. Hey, Oxford Archaeol. Unit, Oxford, through English Heritage.

OxA-7092charred barley seeds, CWRF96 50, δ13C = −22.1‰145 ± 35
OxA-7140charred cereal seeds, CWRF96 71, δ13C = −22.8‰1300 ± 55

Comment (G. H.): one sample appears to be intrusive, the other suggests that the Worton structure was in use in the middle Saxon period, probably during the earlier middle Saxon phase at Yarnton.

Yarnton Saxon: Worton Rectory Farm

Samples of charred seeds and charcoal from the site Yarnton Worton Rectory Farm (NGR SP 475113), settlement and halls, Yarnton, Oxon, submitted by G. Hey, Oxford Archaeological Unit, Oxford, through English Heritage.

OxA-3914charcoal, YWRF CA/6, δ13C = −22.9‰1270 ± 75
OxA-4679charred barley seeds, YWRF 3317/A, δ13C = −24.1‰1215 ± 60
OxA-4680charred wheat seeds, YWRF 3317/B, δ13C = −21.9‰1135 ± 60
OxA-4681oak charcoal, YWRF 3332, δ13C = −25.6‰835 ± 60
OxA-4682oak charcoal, YWRF 3309/A, δ13C = −23.5‰1165 ± 60
OxA-4683charred barley seeds, YWRF 3693, δ13C = −22.7‰1255 ± 65
OxA-4684oak charcoal, YWRF 3442, δ13C = −24.5‰1335 ± 60
OxA-4685Prunus of spmosa charcoal, YWRF 3422, δ13C = −25.6‰1185 ± 65
OxA-4686hazel/alder charcoal, YWRF 3466, δ13C = −23.4‰1155 ± 65
OxA-4687oak sapwood charcoal, YWRF 3198, δ13C = −24.4‰1235 ± 60
OxA-4688charred barley seeds, YWRF 3192, δ13C = −22.7‰1285 ± 60
OxA-4689charcoal, YWRF 3903, δ13C = −26.7‰1205 ± 70
OxA-4737sheep/goat bone, YWRF 3871, δ13C = −20.1‰1360 ± 55
OxA-5468cereal seeds, YWRF 3738, δ13C = −23.l‰1210 ± 50
OxA-5469cereal seeds, YWRF 3436, δ13C = −26.l‰820 ± 50

Comment (G. H.): the estimated date range for the use of the hall buildings was in Phases 2 and 3, the middle Saxon phases of the site. The beginning of Phase 2/3 is cal AD 660–780 (at 95% prob.), and for its end is cal AD 840–940 (at 95% cprob.). The estimated date range for the end of Phase 4 of the Yarnton Saxon settlement is cal AD 910–1160 (at 95% prob.). The Phase 2/3 settlement appears to have been occupied for between 90–250 years (at 95% prob.).

Twelve of the radiocarbon determinations obtained are regarded as being useful for dating the occupation of the post-built structures. The results suggest that it is unlikely that all the buildings were in use at any one time. Although there are too few measurements to suggest reliably the order in which the buildings were constructed or demolished, general trends can be seen in the estimated dates. For example, it is likely that building B3348 was demolished after buildings B3959 and B3619 (89% prob.), and that its use overlapped with that of building 3620 (over 95% prob.).

Yarnton Saxon: settlement and other features

Samples of charred seeds and charcoal from the site Yarnton Worton Rectory Farm (NGR SP 475113), Yarnton, Oxon, submitted by G. Hey, Oxford Archaeol. Unit, Oxford, through English Heritage.

OxA-3915charcoal, YWRF CA/7, δ13C = −23.1‰1040 ± 65
OxA-5467cereal seeds, YWRF 3666, δ13C = −21.3‰1055 ± 50
OxA-5470charred cereal seeds, YWRF 3043, δ13C = −23.4‰1200 ± 45
OxA-5471charred cereal seeds, YWRF 3043, δ13C = −24.2‰1205 ± 50
OxA-7365wheat seeds, 2616b, δ13C = −20.2‰1225 ± 50
OxA-7371wheat seeds, 2616a, δ13C = −21.3‰1190 ± 35

Comment (G. H.): the waterholes were situated in the south-east of the settlement. A wooden object waterlogged in the bottom of 3029 (GU-5138:1390 ± 50 BP) was made of oak and it seems likely that the result has a significant inbuilt age, and so is older than the date of manufacture of the object. For this reason the result is not useful for dating the waterhole. Two samples of charred cereals from the adjacent waterhole 3043 produced an estimated date range for this feature of cal AD 770–900. Two samples, one of cereal and one of unidentified charcoal from hearth 3666 of smithy 3926 suggested an estimated date range of cal AD 910–1160 for the use of the smithy.

Yarnton Saxon: Yarnton settlement, sunken-featured buildings

Samples of seeds from the site Yarnton Worton Rectory Farm (NGR SP 475113), settlement and sunken-featured buildings (SFB), Yarnton, Oxon, submitted by G. Hey, Oxford Archaeol. Unit. Oxford through English Heritage.

OxA-3915charcoal, YWRF CA/7, δ13C = −23.1‰1040 ± 65
OxA-5467cereal seeds, YWRF 3666, δ13C = −21.3‰1055 ± 50
OxA-5470charred cereal seeds, YWRF 3043, δ13C = −23.4‰1200 ± 45
OxA-5471charred cereal seeds, YWRF 3043, δ13C = −24.2‰1205 ± 50
OxA-7365wheat seeds, 2616b, δ13C = −20.2‰1225 ± 50
OxA-7371wheat seeds, 2616a, δ13C = −21.3‰1190 ± 35

Comment (G. H.): seven of the determinations from the ten samples submitted are regarded as providing useful information for the purposes of dating these buildings. Buildings 2652 and 2556 each produced two radiocarbon dates, and one came from each of 2689, 888, and 2577. Strictly these represent termini post quern for the disuse of these SFBs.

The estimated date range for the beginning of Phase 1 of the Yarnton Saxon settlement is AD 380–550 (at 95% prob.).

Although the pattern was more complex than originally perceived, the radiocarbon results tend to support a broad shift in settlement from west to east through the Saxon period.

The 14C results show that these structures were probably not all in contemporary use. Those from buildings 2689, 2652 and 888 are significantly earlier than 2556 and 2577. The preliminary phasing, which placed all of these buildings in Phase 1 and was supported by the finds data then available, had to be abandoned. Firstly, it became apparent that not all the SFBs were of the same date, and secondly that the period of use of the later ones appeared to overlap with the use of the halls.

Conderton Camp

Samples of charcoal and bone from the site Conderton Camp, Evesham, Hereford and Worcester (NGR SO 972 384), submitted by N. Thomas, 11 Highbury Villas, Kingsdown, Bristol BS2 8BX, through English Heritage.

OxA-8516Rosaceue charcoal, O(a), δ13C = −26.0‰2115 ± 40
OxA-8517Rosaceue charcoal, O(b), δ13C = −25.9‰2225 ± 35
OxA-8568Quercus sp. charcoal, Ala), δ13C = −24.7‰2510 ± 40
OxA-8569Quercus sp. charcoal, A(b), δ13C = −26.1‰2385 ± 45
OxA-8612pig bone, D, δ13C = −22.1‰2415 ± 30
OxA-8614animal bone, Li(a), δ13C = −21.2‰2260 ± 40
OxA-8615animal bone, Li(b), δ13C = −22.0‰2355 ± 40
OxA-8639pig bone, F, δ13C = −21.2‰2285 ± 25
OxA-8640cattle bone, Period 2(I), δ13C = −21.8‰2265 ± 25
OxA-8641horse bone, δ13C = −21.6‰2235 ± 25
OxA-8642cattle bone, δ13C = −21.1‰2295 ± 25
OxA-8643sheep/goat bone, δ13C = −21.8‰2155 ± 25
OxA-8644sheep/cattle bone, Liv(a), δ13C = −21.2‰2315 ± 35
OxA-8645sheep/cattle bone, Liv(b), δ13C = −21.5‰2310 ± 35
OxA-8646sheep/cattle one, Lv(a), δ13C = −21.3‰2270 ± 35
OxA-8647sheep/cattle bone, Lv(b), δ13C = −22.3‰2275 ± 35
OxA-8648bone, sheep, N, δ13C = −21.4‰2280 ± 25
OxA-8649cattle bone, δ13C = −21.2‰2265 ± 25
OxA-8727animal bone, Lii(b), δ13C = −20.8‰2085 ± 55
OxA-8785animal bone, Lii(a), δ13C = −20.8‰2275 ± 40

Comment (N. T.): Period 1, the start of the hillfort, is satisfactorily early and samples A(a) and A(b) are important for the beginning of the middle Iron Age hereabouts. The cross-rampart (OxA-8640) is early enough to contain all the excavated structures within the constricted circuit of ramparts. But it is also made clear by radiocarbon that the development of the community began at once; and resistivity and excavation showed that dense occupation had just begun before construction of the cross-rampart. The order of house building appears to be (1)H4, (2)H6/H2, (3)H1/H3. This is acceptable, although ceramically H6 should have been as late as H3. Radiocarbon dating does not support the stratigraphic order of Pit BB (earlier) and Pit CC (later). The early date of Pit A goes against the late ceramic phase for the large urn PO81 found there. All in all, radiocarbon dating gives no more than qualified support to the ceramic phasing.

Fennings Wharf

Samples of charcoal from the site of Fennings Wharf, London (NGR TQ 32848036), England, submitted by B. Watson, Museum of London through English Heritage.

OxA-8763Quercus sp. charcoal, FW84 435a, δ13C = −22.7‰3360 ± 40
OxA-8764Quercus sp. charcoal, FW84 435b, δ13C = −22.6‰3400 ± 45
OxA-8765Quercus sp. charcoal, FW84 546a, δ13C = −22.6‰3345 ± 45
OxA-8766Quercus sp. charcoal, FW84 546b, δ13C = −22.9‰3425 ± 40
OxA-8767Quercus sp. charcoal, FW84 547a, δ13C = −22.7‰3420 ± 40
OxA-8768Quercus sp. charcoal, FW84 547b, δ13C = −22.5‰3490 ± 40
OxA-8769Quercus sp. charcoal, FW84 573a, δ13C = −22.7‰3430 ± 45
OxA-8770Quercus sp. charcoal, FW84 573b, δ13C = −22.8‰3545 ± 40

Comment (B. W.): the ring ditch has three phases of funerary activity and the radiocarbon determinations confirm that it was used for a relatively short time (c. 1900–1600 cal BC). This duration is interesting as it means that the monument may only have been in use for a few generations. Without the radiocarbon analysis this important monument would not be properly dated and its duration of use would have been impossible to estimate. Nor would we have realised that the central feature was a secondary addition to the monument, as the ceramics from it would have been used to date the entire monument in the absence of other reliable dating evidence.

Thames Foreshore

Samples of antler from the site of Thames Foreshore at Richmond, (TQ 17697450), submitted by R. Cowie, Richmond Archaeol. Soc., through English Heritage.

OxA-7897red deer antler beam mattock, δ13C = −23.7‰121.9 ± 0.7% mod

Comment (R. C): the modern radiocarbon determination for the antler object from Richmond Bridge is surprising, since typologically the object is very similar to prehistoric antler beam mattocks. Moreover the condition of the object suggested that it had been in the river for a considerable length of time. If the object is modern it brings into question the identification of similar (but as yet undated) river finds as antler beam mattocks. Nevertheless, on balance I think the object is probably a genuine artefact that has somehow been contaminated. Could its poor condition and riverine environment have contributed to this?

Camber Castle

Samples of bone from Camber Castle, Rye, E. Sussex (NGR TQ 92181845) England, submitted by S. Davis. English Heritage.

OxA-9065rabbit bone, 834692/4/PR153, δ13C = −21.4‰342 ± 36
OxA-9066rabbit bone, 834714/4/PR280, δ13C = −20.6‰424 ± 39
OxA-9067rabbit bone, 834692/4/PR86, δ13C = −21.3‰365 ± 50
OxA-9068rabbit bone, 834692/5/PR147, δ13C = −22.5‰301 ± 36

Comment (S. D.): with five out of six results (see comment on OxA-7533 and −7534 in Arch. List 27) dating roughly to the period of occupation of the castle, it looks most probable that the majority of the huge (>1000) accumulation of rabbit bones at Camber is indeed old and not derived from perhaps eighteenth or nineteenth century intrusions. This does, of course, solve one important problem. Rabbits are burrowing animals, and their bones, when found in archaeological deposits, tend to be dismissed as “intrusive”. Clearly zoo-archaeologists need to be careful on this point! So, at the very least, dating the six Camber rabbit bones had not been a useless exercise. But the main zoo-archaeological question of how the rabbit bones got there in the first place still remains to be answered. With only one rabbit bone showing signs of butchery while the small number of similar-sized bird bones have many cut marks, still makes me think that the soldiers at Camber were not responsible for the majority of the rabbits. This means that they probably got there as a result of ‘natural’ means. Perhaps some died in infancy in their burrows while others were brought into the castle by predators such as large birds of prey, or dogs and cats. We noted two age groups of rabbits, the very young ones may have been taken by the former category of prey and the older rabbits perhaps by the latter category of prey.

Arbury Camp

Sample of leather from Arbury Camp (NGR TL 44366142), Cambridge, submitted by R. Jackson, Dept. Prehist. & Romano-Brit. Antiq., Brit. Mus., London. Comments by J. D. Hill, Dept. Prehist. & Early Europe, Brit. Mus., London.

OxA-6582leather, 1/2989, δ13C = −22.7‰2160 ± 50

Comment (J. D. H.): the determination is significant as it is from leather, or more exactly treated animal hide. Almost no leather or hide survives from Britian or northern Europe before the Roman period. This is probably because hides were not tanned in the way Roman and later leather was/is prepared.

The date is from a group of leather/hide scraps and offcuts found in the ditch terminal of a circular lowlying hillfort or ringwork on the northern outskirts of Cambridge. Excavated by the Cambridge University Archaeological Unit, the ringwork was empty of archaeological features and appears to have been a temporary refuge, large animal kraal, etc. There were a tiny number of handmade Middle/Later Iron Age sherds of pottery from the ditch around the ringwork. These span the period c. 350–1 BC.

The presence of the leather/hide was unexpected. The ditch terminal was waterlogged, and also contained oak chippings from constructing the gateway of the site(?). This may have helped with the preservation. The study of the leather/hide is ongoing. Future work aims to investigate the species of the animal, how it was originally treated and why it, unusually, survived.

Comet Lode

Sample from Comet Lode prehistoric opencast mine, Copa Hill, Cwnystwyth Mines, Dyfed (NGR SN 811751), Wales, submitted by S. Timberlake, SAFS, Univ. N. Wales, Bangor, Gwynedd, through the NERC-funded ORADS facility.

OxA-6684red deer antler, CH95/D8/077 (find 41), δ13C = −21.4‰3405 ± 70

Comment (S. T.): the context for the find of this antler tool within the basal mining deposits of the Early Bronze Age opencast copper/lead mine on Copa Hill was really quite significant for the site. During excavations of the front part of this infilled 10 m plus deep opencast trench mine, between 1993 and 1999, an 8 m long (2 m deep) entrance cutting was revealed, cut along the cleavage direction through an unmineralized area of the slaty country rocks, most probably for the purposes of drainage. Within the base of this cutting had been found (in situ) an intact 5 m long hollowed out wooden tree trunk of alder, which had been placed here, and subsequently re-positioned, at some date after the cutting of the original trench. A 14C determination for the alder of 3690 ± 90 BP had already been obtained, but beneath this lay some 20–30 cm of compact line-grained waterlogged mining sediments with hammerstones, broken withy handles (hafts), basketwork, withy rope and other cut wood. It was possible therefore that such tools would date a slightly earlier period of mining. In particular, the antler pick had been heavily used and worn, and showed subsequent evidence for re-use as a hammer, most probably in the working of the slaty country rock rather than the hard quartz veins.

The determination obtained does not suggest any significantly earlier working, and appears to conform with most of the other charcoal dates obtained from the opencast, suggesting that the duration of mining spanned no more than 300–400 yr at the most between c. 1900 and 1600 BC.

Crosby Airport

Samples of charcoal from Crosby Airport (NCR NY 47556065), Carlisle, Cumbria, submitted by P. Flynn, Carlisle Archaeology Unit, Dept. Leisure and Community Development, Carlisle. Comments by M. R. McCarthy, Carlisle Archaeol. Ltd., Carlisle.

OxA-6180charcoal, CBA S 3, δ13C = −25.5‰4645 ± 60

Comment (M. R. M.): OxA-6180 came from a site that yielded features and pottery indicative of prehistoric activity, and the sample was submitted to enhance the case for further archaeological work. This was successfully achieved and a range of new features were identified, although as it happens very little else capable of producing dating was recovered. Nevertheless, a potential Bronze Age ritual site and other features typical of prehistoric activity were recovered.

We entirely accept the difficulty in drawing conclusions from single determinations, but OxA-6180 was useful in the context that was needed at the time, and has enabled us to focus thinking a little more clearly on archaeological remains and ancient land-use in this area.

Knowefield and Tarraby Lane, Carlisle

Sample of charcoal from a site close to Hadrian's Wall at Knowefield and Tarraby Lane (NGR NY 40355745), Carlisle, Cumbria, where features thought to antedate the building of the Wall were recovered, although further excavation yielded no additional cultural material. The sample was submitted by P. Flynn. Carlisle Archaeology Unit, Dept. Leisure and Community Development. Carlisle to enhance the case for more archaeological work. Comments by M. R. McCarthy, Carlisle Archaeol. Ltd., Carlisle.

OxA-6181charcoal, KNF B Q 32, δ13C = −26.8‰2465 ± 50

Comment (M. R. M.): We entirely accept the difficulty in drawing conclusions from single dates, but OxA-6181 was useful in the context that was needed at the time, and has enabled us to focus thinking a little more clearly on archaeological remains and ancient land-use in this area.

Pwll Fanog slate ship

Sample of wood from Pwll Fanog slate ship (NGR SH 535707), Wales, submitted by D. C. Jones, Dept. Ocean Sci., Univ. Wales, Bangor, Gwynedd.

OxA-7969wood, 0972, δ13C = −24.5‰395 ± 35

Comment (D. C. O.): this date fits perfectly with the theoretical calculations and designates this as the oldest collection of Welsh roofing slates yet found.

Hatton/Hilton/Foston Bypass

Samples from Hatton/Hilton/Foston (NGR SK 204313), Derbyshire, submitted by R. Roseff, Field Archaeology Unit, Univ. Birmingham, Edgbaston.

OxA-4401Quercus sp. charcoal, HHF/93, 5, δ13C = −24.9‰3480 ± 65
OxA-4402Quercus sp. charcoal, HHF/93, 12, δ13C = −25.1‰1130 ± 60
OxA-4403Quercus sp. charcoal, HHF/93, 17, δ13C = −26.3‰1185 ± 60
OxA-4404Quercus/Corylus charcoal, HHF/93, 32, δ13C = −24.7‰3370 ± 65

Comment (R. R.): in March 1993, a ring-ditch was excavated near Foston in southern Derbyshire, prior to the construction of the A564 Foston-Hilton-Hatton bypass. Two phases were recorded. Radiocarbon determinations suggesting an Early Bronze Age date were obtained from the fill of the Phase 1 ditch and from a nearby small circular pit, interpreted as a satellite cremation. Radiocarbon determinations suggesting a Middle to Late Saxon date were obtained from the fill of the Phase 2 ditch. The only artefacts recovered were a single pottery sherd, probably dating to the Early Bronze Age, and three flint flakes.

National Museums of Scotland dating programme: 1994–98

Samples submitted by J. A. Sheridan, with comments by her and T. G. Cowie and F. J. Hunter, National Museums of Scotland, Edinburgh.

The new Museum of Scotland (NMS) opened on 30 November 1998, and has attracted 1.4 million visitors to date (Dec 2000). Preparations for its Early People gallery included the commissioning of a series of dates for organic objects and human bone. The specimens reported on here were submitted by J. A. Sheridan of NMS between 1993–7; see Datelists 16, 20 and 29 for others from the same programme. Many of the specimens had been in the national collection for many years. The programme paid particular attention to objects found in peat. With wooden objects, special care was taken to sample the outermost rings, and in most cases the dated sample lies within 20 years of the tree death date. Three specimens were found to be of recent date; two (from human bone) produced anomalous dates; and the others were either as old as had been expected, or significantly older than had previously been assumed. The results vindicate the strategy of targeting peat finds. All determinations have been calibrated using OxCal v3.5, and all calibrated dates are cited at 95%.


A birch bark fragment from an oak trunk coffin found in 1878 in a peat bog at Dalrigh, Nr. Oban (NGR NM c. 860307) Argyll and Bute (Abercromby 1905; Mapleton 1879; Mowat 1996, 102–3). Submitted 1996.

OxA-6813birch bark, NMS X.EQ 192, δ13C= −27.1‰3555 ± 60

Comment (J. A. S): drilled holes indicate that this had come from a larger, sewn, object, either a cover for the (completely decayed) body or a grave good. The coffin had been laid on boggy ground, pinned into place by logs, covered by moss, branches of birch and probably hazel, and hazelnuts, then enveloped within a cairn of stone and peaty earth. Its date of 2040–1690 cal BC (2040–1730 cal BC at 94.4% prob., 1710–1690 cal BC at 1.0% prob.) confirms this as an Early Bronze Age tree-trunk coffin, very rare in Scotland (Mowat 1996, 137–41).


Femur and ulna from female cist burial under a low mound in the grounds of Mountstuart House, Bute, Argyll and Bute (NGR NS 105605). Excavated in 1887 (Bryce 1904; Callander 1916, 239; Clarke et al. 1985, 289, Wgs 4.90, 5.35). Submitted 1995 and 1996.

OxA-6130human ulna, NMS X.EQ 102, δ13C= −21.1‰2645 ± 50
OxA-6579human femur. NMS X.EQ 102, δ13C=-21.8‰2955 ± 55

Comment (J. A. S): the bones are from a young adult female, buried in a crouched position on her right side, facing east. A hole in her skull, traditionally interpreted as a trepanation, has been confirmed as resulting from disease (M. Kaufman, Edinburgh Univ., pers. comm.). She was wearing a spacer plate necklace of jet and cannel coal and was accompanied by a food vessel, a decayed bronze object (probably an awl), some burnt stuff and ‘what seemed like the remains of pins or skewers’. The intention of the dating was to add to the few dates currently available for spacer plate necklaces and for Scottish food vessels. The results of cal 920–660 BC (920–760 BC at 94.2%, 690–660 BC at 1.2% prob.) and cal 1380–1000 BC (1380–1330 BC at 3.1%, 1320–1000 BC at 92.3% prob.) are both several centuries later than anticipated and are inconsistent with each other. No obvious source of contamination is known.

Kiloran Bay

Neck bone of horse from a Viking boat burial in Kiloran Bay, Colonsay, Argyll and Bute (NGR NR c. 400976), excavated 1883 (Anderson 1907, Graham-Campbell and Paterson forthcoming). Submitted in 1996.

OxA-6604horse neck bone, NMS Z. 1990.113, δ13C= −22.7‰1110 ± 45

Comment (J. A. S): dating was carried out to establish whether the horse, whose bones are in much better condition than the human (adult male) skeleton, was indeed part of the Viking boat burial. The result of cal AD 780–1020 confirms that it is consistent with the expected date range for pagan Norse burials in Scotland (AD 850–950). It appears that the horse had been sacrificed to accompany the man, with his possessions, in his boat grave.

Kerrow Farm

Fragment of haft from a socketed axehead found in a washed-out bank of the River Glass at Kerrow Farm, Kilmorack parish, Highland in 1994 (NGR NH 323303; Hanley 1994; Sheridan 1999, 54). Submitted, on behalf of Inverness Museum and Art Gallery, in 1995.

OxA-6040haft fragment, hardwood (probably ash), INVMG 1994.038, δ13C= −26.3‰,2960 ± 60

Comment (T. G. C): the axehead falls within Schmidt and Burgess' category of Gillespie type, characterised by faceted broad baggy bodies, and may be classified more particularly as an example of the Culloden variant on the basis of the slight moulding or collar around the mouth (Schmidt and Burgess 1981, 191–7). The calibrated date of 1380–1000 cal BC (1380–1330 cal BC at 5.2%, 1320–1000 cal BC 90.2% prob.) is consistent with the view that axeheads of this general form had developed by the Wilburton/Ewart Park transition, which can be placed around the early tenth century BC in the light of the recent Oxford dating programme (Needham et al. 1997, 90–3).

Arniclee Farm

Fragment of shaft from socketed spearhead found in 1987, on peaty hill slopes ploughed for afforestation, behind Arnicle Farm in Barr Glen, Kintyre, Argyll and Bute (NGR NR 72253675), submitted in 1995. The wood was too degraded to allow positive species identification.

OxA-6041shaft fragment, wood. NMS X.DG 115. δ13C= −24.4‰3115 ± 60

Comment (T. G. C): this is a side-looped spearhead of Coles’ type D, the most widespread of the Middle Bronze Age spearhead types found in Scotland (Coles 1964, 104, 143, fig 11). A calibrated date of 1690–1320 BC (OxA-5948: 3225 ± 65 BP) for a side-looped spearhead from the River Thames at Mortlake suggests that the type may have emerged early in the Middle Bronze Age (Needham et al. 1997, 86 and table 1). However, the longevity of the form remains uncertain. At South Lodge Camp. Dorset and Tormarton, Gloucestershire, fragmentary spearhead blades, possibly from side-looped spearheads, were found with burials with calibrated dates around the twelfth/eleventh centuries BC (ibid. 84–5). The date of 1520–1210 cal BC (1520–1250 at 92.8%, 1240–1210 at 2.6% prob.) for the Arnicle spearhead is thus consistent with the available dates, which may hint at a relatively long currency for the type. It also provides a useful addition to the relatively few radiocarbon determinations available for the Acton and Taunton metal assemblages generally (ibid. 84–6).


Fragment of shaft from socketed spearhead found at Premnay, Aberdeenshire (NGR NJ c. 6224), presented to the national collections by Captain H P Lumsden of Auchindoir in 1937; date with circumstances of find unknown. Submitted 1996.

OxA-6811shaft fragment, ash, NMS X.DG 100, δ13C= −25.1‰130 ± 45

Comment (J. A. S): basal-looped spearheads, Coles’ Type E (Coles 1964), are known to have been in use during the Taunton/Penard and the early Wilburton metalworking phases, with an estimated date range of c. 1400–1100 BC (Needham et al. 1997). The date of 1670–1960 cal AD indicates that what has been dated is a modern shaft fragment, suitably ‘antiqued’.

Fort William

Handle of peat spade found at a depth of nearly 2.5 m during building works at Fort William, Highlands (NGR NN 1275) (Anon 1939), and presented to the national collections in 1938. Submitted in 1995.

OxA-6053peat spade handle, pine, NMS W.MP 601/PF 13, δ13C= −26.3‰145 ± 40

Comment (J. A. S): the date of cal AD 1660–1960 confirms suspicions that this is a relatively recent peat spade, as its well-preserved leather binding and metal tacks had suggested.

Howe Farm

Leather thong from wooden box containing leatherworker's tools, found by peat-cutters in a bog near Howe Farm, between the parishes of Evie and Rendall and Birsay and Harray, Orkney (NGR HY c. 3523) in 1885 (Cursiter 1886; Earwood 1993a, 105, fig 70, 267; Kirkness 1953; Stevenson 1952). Submitted in 1996.

OxA-6814leather thong, NMS X.FC 262(4), δ13C= −21.3‰1300 ± 50

Comment (J. A. S): the ‘Evie box’ is a famous decorated box of alder, with a sliding lid; leather thongs had been used, as replacements for earlier metal bands, to secure the lid. Inside were found tool handles of wood, bone, antler and horn (but lacking their metal blades); a lump of pumice; a slitted leather strap fragment, possibly for retaining blades; a decorative stamp; a strip of antler and three worked antler tines. It was identified as a leatherworker's box by Kirkness in 1953. The date of 640–870 cal AD confirms its suspected Early Historic date (see, for example, Stevenson 1952).

Cheese Bay

Scoop or boat bailer, found 1.37 m deep in a peat bank at No. 5, Cheese Bay, North Uist, Western Isles, (NGR NF c. 959736), in 1974 or 1975 (Close-Brooks 1984; Earwood 1993a, 117, 119, 270). Submitted in 1997.

OxA-7476scoop or bailer, probably larch, NMS X.IP 3, δ13C= −24.6‰2090 ± 35

Comment (J. A. S): AMS dating was the only way of establishing a date for this simple-shaped object. Like many wooden items from the tree-poor Western and Northern Isles, it had been made from driftwood, either larch or spruce, most probably the former, and had been hollowed from a piece of cleft roundwood. Its size, shape and proximity to the sea indicates that it could have been for bailing boats, despite its flimsylooking handle. Even if the wood had been drifting for the best part of a century before use, the date of cal 210–1 BC establishes that this item is of considerable antiquity. Other dates for similar-looking Scottish bailers from NMS collections have produced a wide date spread: 1400–1020 cal BC (Beta-114706: 2990 ± 60BP) from Grimsay, N Uist, 810–410 cal BC (AA-29712: 2540 ± 50 BP) from near Creagan Nighean Domhnuill Mhic Iain, Lewis, and 410–620 cal AD (AA-29711: 1555 ± 50 BP) from Priesthoulland, Shetland. These highlight the long-term design stability of this simple utensil.


Ladle, found at a depth of almost 2m in a peat bank at Bragar, Lewis. Western Isles (NGR NB c. 2947); presented to the national collections in 1947/8 (Anon 1948). Submitted in 1997.

OxA-7474ladle, bird cherry, NMS W.SFE 9/ME 832, δ13C= −25.6‰140 ± 30

Comment (J. A. S): dating was undertaken to check whether, as stated in the donation entry (Anon 1948), this ladle was ‘relatively recent’. The result of 1670–1960 cal AD confirms that this object, like the peat spade from near Fort William (see above), relates to peat cutting in the relatively recent past. It had simply been abandoned or lost in a peat bank during cutting.


Roughout for wooden cup, found in a bog in Watten, Caithness, Highland (NGR ND 2354), and presented to the national collections in 1878 (Anon 1878). Submitted in 1997.

OxA-7553roughout for cup, probably alder, NMS X.AQ 43, δ13C= −24.6‰2055 ± 35

Comment (J. A. S): this partly-hollowed cup with a single vertical handle near its rim was carved from a branch. A general resemblance in size and shape (but not function) to some of the stone lamps of the Scottish Iron Age (eg Inverkeithny, Banffshire: MacGregor 1976, no 334) created the desire to check whether it might have been contemporary with them. Its date of cal 170 BC-AD 30 confirms that it was, although no special significance should be read into the similarity with the lamps. Another simple cup from the NMS collections, unhandled and made of driftwood, from Hill of Garth, Shetland, has produced a date of cal AD 340–610 (AA-29706: 1590 ± 55 BP).

Unfinished carved bowls

Unfinished carved bowls, all found during peat digging: at Airds, Ness, Lewis, Western Isles, (NGR NB c. 5364; Anon 1910; Earwood 1993a, 264) around 1909; at Dalness, Sutherland, Highland (NGR NC c. 6554; Anon 1937) around 1936; and at Kirkchrist, Penninghame, Wigtownshire, Dumfries and Galloway (NGR NX c. 3659; Anon 1889; Earwood 1993a, 278; Maxwell 1889, 230, fig 55), in 1884, at a depth of 1.22 m. Submitted in 1997.

OxA-7552Airds, larch or spruce, NMS W.SFA 13/ME 401, δ13C=-23.4‰2320 ± 35
OxA-7477Dalness, alder, NMS W.SFA 24/ME 747, δ13C= −26.0‰1875 ± 35
OxA-7551Kirkchrist, alder, NMS W.SFA 8/ME 219, δ13C= −26.6‰1850 ± 35

Comment (J. A. S): these dates contribute to Earwood's typochronology of early domestic wooden vessels in Britain and Ireland (Earwood 1990; 1993a,b; 1997, and see Arch. Lists 12 and 16). The fact that all are unfinished, like several others from watery contexts (e.g. Bracadale, Skye: Crone 1993), may well indicate that they had been placed there to soak, making them easier to work. All have been carved, not turned. The Airds bowl is roughly circular, flat-based, with gently splaying walls and an inturning rim around 240 mm in diameter; its date of 490–200 cal BC (490–350 BC at 75.9%; 320–200 BC at 19.5% prob.) confirms Earwood's prediction of an Iron Age date. The Dalness bowl is flat-based, oval in plan and has two vertical, squarish, (as-yet?) unperforated handles at either end; its overall length is around 340 mm. With some features reminiscent of the bowls from Bracadale and Talisker Moor, Skye (Crone 1993; Barber 1982; Mowat 1996, 106), it is perhaps not surprising that its date is also comparable, 60–240 cal AD(cf. cal 50 BC–AD 220 (UT-1698: 1930 ± 50 BP) and 20–400 cal AD (OxA-3542: 1830 ± 80 BP), respectively). The Kirkchrist bowl is also oval, with handles at either end, but these are horizontal, and the bowl is very large (length c. 610 mm). Earwood had estimated a date in the late first millennium BC; the result is 70–250 cal AD.


Piece of shaped calf-skin, found with a flint barbed and tanged arrowhead around 1901 at Corrydown, Gartley, Aberdeenshire (NGR NJ c. 4333), under c. 1.37 m of peat, at the bottom of the bog (Anon 1902). Submitted in 1996.

OxA-6812calf skin, NMS X.AD 1106, δ13C=-21.4‰2835 ± 60

Comment (J. A. S): the piece of calf skin, although shaped and directly associated with the arrowhead, does not seem to be a container for the arrowhead; whether it had been part of a garment or rug-like object is unknown. It retains its thick, wavy hair. The date of 1190–830 cal BC (1190–1170 BC at 1.7%, 1160–830 BC at 93.7% prob.) makes this the latest dated flint arrowhead in Scotland, along with a set of five barbed and tanged arrowheads from a cremation in a pit at Grandtully, Highland (1400–800 cal BC, GaK-1396: 2880 ± 100 BP, Simpson and Coles 1990).


Sword of Late Bronze Age type made of yew, found during peat cutting in 1957, at least 1.83 m below the original surface of the peat, near Grotsetter, Tankerness, Orkney (aka Groatster, NCR HY 493063; Stevenson 1958). Submitted in 1996.

OxA-6779sword, yew, NMS X.DL 62, δ13C= −23.3‰2710 ± 50

Comment (J. A. S): this is a skeuomorph of a leaf-shaped sword, complete with its (damaged) hilt. It is broadly comparable with swords of the Ewart Park metalworking phase (Colquhoun and Burgess 1988) whose date range appears to be around 1000–750 BC (Needham et al. 1997, 80). The date of cal 980–790 BC (980–950 BC at 3.6%, 940–790 BC at 91.8% prob.) is consistent with this, and in line with Stevenson's prediction. The fact that yew is not native to Orkney means that the sword must have been imported; the nearest locations where yew would have grown at the time are north-west England and Ireland (Tipping pers. coram.). That the sword may have come from Ireland is suggested by the fact that another yew sword was found there, in 1958 (Raftery 1960, 24, fig 12; cf. a reference to “swords, made of hard wood”, dug from a bog on Jura before 1855: Kilkenny Archaeol. Soc. J., 3 (1854–5), 339–40). The item cannot have been used as a mould for making blades; perhaps it was a prestigious and exotic display object and then a votive deposit. Perhaps the yew had symbolic significance; and Stevenson pointed out that its colour would have resembled bronze. That it had seen some use is attested by the absence of the pommel, broken off in antiquity; by polish on the hilt; and by ancient damage to the tip, which had been re-worked.

Ravenstone Moss

Rudder or (less likely) leeboard or paddle, found before 1866 with four or five other ‘paddles’ at a depth of c. 2 m in Ravenstone Moss, Wigtownshire, Dumfries and Galloway (NGR NX c. 4042), probably during drainage operations (Mowat 1996, 104, fig 41; Stuart 1866, 122–3,150, pl XII.4 (incorrect reconstruction; also incorrectly locates Ravenstone Moss); Stuart 1870, 19–20). Submitted in 1996.

OxA-6815rudder, oak, NMS X.HU 14, δ13C= −24.5‰2470 ± 55

Comment (J. A. S): these items were found “close to a mass of timber”—almost certainly a crannog. The date of cal 770–400 BC places the object within the Late Bronze Age; if one assumes contemporaneity with the adjacent crannog, this makes it of comparable antiquity to the crannog at Oakbank, Perth and Kinross (Mowat 1996, 101–2), but earlier than most dated Scottish crannogs, including those in the nearby Dowalton Loch (see below and Hunter 1994). Logboats and their accessories are frequently associated with crannogs (ibid.).


Probably a hand flax scutch (beater), made of alder, found before 1892 “in a moss in Wigtownshire”, Dumfries and Galloway (NGR NX; NMAS 1892). Submitted in 1997.

OxA-7554hand scutch, alder, NMS H.MP 119, δ13C= −25.1‰2435 ± 35

Comment (J. A. S): this implement is long and flat with a centrally-placed integral handle at one end. It had been made from a radially-split alder trunk around 18 cm in diameter. While superficially similar to oars and paddles, its shape and proportions find a better match in implements used to break flax fibres by hand. Its date of 770–400 cal BC (770–610 BC at 34.1%, 600–400 BC at 61.3% prob.) makes it the earliest dated piece of flax-working equipment in Britain and Ireland (although not the first evidence for flax itself, which appears to have been a Neolithic introduction: Earwood 1993, 125–31; Greig 1991, 300).

Near Gutcher

Paddle from a horizontal mill-wheel, found at the bottom of a peat bank around 2.13 m below the surface, Probably in the early 1950s, near Gutcher in Yell, Shetland (NGR HU 547989; Maxwell 1956; Mowat 1996, 89). Made of driftwood, most probably larch; submitted in 1997.

OxA-7475mill wheel paddle, probably larch, NMS W.PD 13/MP 710, δ13C=-23.4‰1900 ± 35

Comment (J. A. S): initial identification of this object as a mill wheel paddle was tentative, as it was not found near a burn, nor did it resemble traditional Shetland mill wheel paddles of the more recent past (Maxwell 1956). However, its shape and size are consistent with such a function (and Mowat 1996 concurs with this identification, it does not seem to be from a boat, for example). Whether the fact that it was found within 100 m of the seashore indicates its use in a tide-mill is unclear. It is V-shaped in cross-section, and its broad end curves backwards. The date range of 20–220 cal AD is exceptionally early for a horizontal mill wheel paddle, however, even if one allows for a period of up to 70 years for drifting of the wood. (There is no danger of a ‘marine effect’ bias, as the tree was dead when it entered the sea.) The earliest horizontal mill in Europe, at Little Island, County Cork, Ireland, dates to around AD 630 (Rynne 1989). At present, no obvious explanation for the early Shetland date suggests itself, unless the paddle had been made by recycling precious driftwood from an old structure. A second horizontal mill wheel paddle, closer in design to the early Irish examples, from Bankhead Farm, Dalswinton, Dumfriesshire produced a date range of 640–880 cal AD (AA-29705: 1310 ± 55 BP, oak; Maxwell 1956, Mowat 1996, 82). This date is more in line with the traditional theory that watermills were introduced to Scotland from Ireland during the Early Historic period.


Fragment of woven woollen cloth in which a hoard of iron age bronze artefacts had been wrapped, found around 1861 during drainage work at Balmaclellan, Stewartry of Kirkcudbright, Dumfries and Galloway (NGR NX 6579), at a depth of around 0.8 m (Anon 1862, 293–5; Bender Jørgensen 1992, 198; Crowfoot 1948; Henshall 1950; MacGregor 1976, nos 273, 342–8). Submitted in 1996.

OxA-6742woollen cloth, NMS X.FA 14, δ13C= −24.2‰1835 ± 55

Comment (F. J. H): the cloth is a twill weave, and it had been used to wrap four parcels of bronze objects, many of which had been folded. The hoard comprised a mirror of Late Iron Age type, together with decorative sheet bronze mounts which may have come from a wooden chest or some other item of furniture; a decorated rotary quern upper stone was found close by. Assuming the cloth was not an heirloom, the date of 60–340 cal AD gives a terminus ante quern for the deposition of the hoard. This is rather later than might have been suggested for some of the pieces on stylistic grounds, and they may have been old when deposited. However, it is consistent with Roman stylistic influences on other pieces, and confirms that much Iron Age metalwork in Scotland is Roman period in date.

Moss of Auquharney

A deer trap (of tread trap type), found around 3 m deep during peat cutting on the Moss of Auquharney, Aberdeenshire (NGR NK 0139) in 1921 (Graham-Smith 1923). The object is in Marischal Museum, Aberdeen. Submitted in 1995.

OxA-6052deer trap, alder, ABDUA 17613, δ13C=-29.l‰1440 ± 45

Comment (J. A. S): the workings of, function, and parallels for this trap are described in Graham-Smith (1923). Of the Irish examples discussed, one, from Larkhill bog, County Fermanagh (Allingham 1896), was recently dated for the National Museum of Ireland, and its date of 530–820 cal AD (GrA-900: 1370 ± 70 BP, Ó. Floinn pers. comm.) is closely comparable with the result of 530–680 cal AD for the Auquharney trap. This kind of trap is shown on an eighth or ninth century cross-shaft at Clonmacnoise, County Offaly.

Dowalton Loch

Fragment of the upper of a decorated leather shoe found during investigations of the crannogs in Dowalton Loch (crannog no. 2), Wigtownshire, Dumfries and Galloway (NGR NX 40614681) in 1863 following drainage (Mowat 1996, 24; Munro 1882; Stuart 1866). Submitted in 1996.

OxA-6052deer trap, alder, ABDUA 17613, δ13C=-29.l‰1440 ± 45

Comment (J. A. S): the shoe fragment, decorated with an intricate stamped design, was found on crannog no. 2 in the Loch, the one with a logboat found under its superstructure. Other finds from this crannog included bones of ox, pig, sheep and goat, most split for marrow extraction; iron slag and iron hammers; whetstones; and ‘a small circular brooch of bronze’. The date of 540–670 cal AD indicates that the crannog was in use during the Early Historic period; other finds dating from the Late Bronze Age onwards (discussed in Hunter 1994) demonstrate that the loch was used for many centuries and for votive as well as domestic use. The shoe joins the small but growing body of evidence relating to Early Historic footwear in Scotland, namely: a complete stamped leather shoe, similar to contemporary Irish examples, from the seventh century occupation at Dundurn citadel, Perthshire (Alcock et al. 1989, 217, illus 16); seventh century fragmentary plain and decorated shoes from the Columban monastery on Iona (Groenman-van Waateringe 1981, discussed in Barber 1981, 365–6); and a shoe-last from Buiston crannog, Ayrshire, dating to between AD 585–630 (Crone 2000).

White Moss

Probably a hobble for a sheep, found in 1984 in peat on the White Moss, Peeblesshire, Borders (NCR NT 1449). Submitted in 1996.

OxA-6603hobble, alder or ash, NMS X.IP 11, δ13C= −26.1‰1395 ± 45

Comment (J. A. S): this object consists of two pieces of carved wood, originally with two fixing pegs, with a socket small enough to enclose the leg of a sheep. The date of 550–770 cal AD (550–720 at 93.8%, 750–770 at 1.6% prob.) indicates that it is of Early Historic date, and makes it comparable in age with the birch withy cattle hobble from the crannog at Buiston, Ayrshire, dating to between AD 585–630 (Crone 2000).

Charred food remains on Hebridean pottery

This set of AMS determinations was obtained to test the possibility of 14C dating of charred food residues on Hebridean Iron Age pottery. There are problems associated with scientific dating of most other materials available on archaeological sites in this environment, and consequent controversy over the dating of pottery styles (for summary, see Lane 1990). A full programme of chemical analysis was also carried out on the residues to identify the food sources. Two Hebridean sites with good quality excavations and a range of comparative scientific dates were chosen for the study: Sollas (NGR NF 801756) (Campbell 1991) and Eilean Olabhat (NGR NF 750753) (Armit et al. forthcoming), both on North Uist, Western Isles, Scotland. UK. Although dating of residues extracted from within pottery fabrics has been carried out, the procedure has not been attempted on charred food residues. Samples submitted in 1997 by E. N. Campbell and R. A. Housley, Dept. Archaeol., Univ. Glasgow, through the NERC-funded ORADS facility.

Eilean Olabhat Phases 1 & 2

OxA-6972charred food remains, 1, δ13C = −24.8‰2325 ± 35
OxA-6973charred food remains, 2, δ13C = −25.2‰2310 ± 35
OxA-6948charred food remains, 6, δ13C = −26.0‰2500 ± 50
OxA-6949charred food remains, 8, δ13C = −24.9‰2205 ± 50
OxA-6950charred food remains, 9, δ13C = −25.6‰2170 ± 50

Comment (E. N. C. and R. A. H.): these five determinations are consistent with an Early Iron Age date. and show both phases were nearly contemporary. OxA-6972, −6973, −6949 and −6950 centre around the fourth century cal BC. OxA-6948 is slightly earlier, but came from a vessel sunk in the floor that may have been curated, or from a residual vessel used as a foundation deposit. The dates are important, firstly in dating a pre-wheelhouse Atlantic roundhouse, suggesting that wheelhouses developed in this area at some Point between the fourth and first centuries cal BC. Secondly, the dates show that vessels decorated with thick cordons, finger impressions and occasional incised chevrons were being produced as early as the fourth century cal BC, marking the beginnings of the highly decorated Middle Iron Age pottery sequence. This phase also produced some unusual vessels whose bases are impressed by woven mats.

Eilean Olabhat Phase 3

OxA-6946charred food remains, 4, δ13C = −25.6‰1575 ± 45
OxA-6947charred food remains, 5, δ13C = −25.0‰1590 ± 50
OxA-6970charred food remains, 3, δ13C = −25.5‰1440 ± 35
OxA-6971charred food remains, 11, δ13C = −25.6‰1970 ± 35

Comment (E. N. C and R. A. H.) three determinations, OxA-6970, −6946, and −6947 were obtained from different areas of a single vessel to check for reliability of the technique. OxA-6946 and −6947 are virtually identical, but OxA-6970 is slightly younger at a statistically significant level. This raises the possibility of some inaccuracy in the technique (see also below), but OxA-6970 may only be a statistical outlier. Seven conventional dates on charcoal twigs from this phase had given a spread of calibrated dates from the nineth century cal BC to the sixth century cal AD (GU-2326, −2327, −3230 and −3234). The AMS determinations confirm a sixth/seventh century AD date for Phase 3, and show, as had been suspected on taphonomic and typological grounds, that the charcoal was of mixed residual origin, collected for metalworking purposes. The determinations provide the first reliable dating of flaring rimmed pottery of the Late Iron Age, and leads to a re-assessment of the dating of the key pottery sequence at the Udal, North Uist. The final result, OxA-6971, is the only problematic one from the site, being considerably older than the expected age. This sample came from a vessel which had been glued together with PVA. Although the area of residue sampled appeared to have no glue, in retrospect it seems possible that it may have been contaminated with dead carbon from the petroleum-derived chemical constituents.

Sollas Site A

OxA-6945charred food remains, S14, δ13C = −25.9‰1895 ± 50
OxA-6966charred food remains, S12, δ13C = −25.1‰1575 ± 35
OxA-6967wood charcoal, S19, δ13C = −26.0‰1845 ± 35
OxA-6968charred food remains, S13, δ13C = −25.3‰1960 ± 35
OxA-6969charred food remains, S18, δ13C = −25.7‰2045 ± 35

Comment (E. N. C. and R. A. H.): Site A, the earlier of two adjacent wheelhouses, had no existing radiometric dates. One charcoal sample, OxA-6967 was dated for control purposes. OxA-6968, −6969 and −6945 fit with the charcoal date and the expected archaeological age of first century cal BC to first/second century cal AD, but OxA-6966 is three or four centuries too young (see below for discussion). The lack of everted rim pottery from this site shows that this key form was not developed until after the first/second centuries AD (cf. Campbell 1991, 168; Parker Pearson and Sharpies 1999, 210).

Sollas Site B

OxA-6944charred food remains, Sll, δ13C = −26.0‰1755 ± 50
OxA-6961charred food remains, S2, δ13C = −25.1‰1790 ± 50
OxA-6962charred food remains, S3, δ13C = −25.1‰1650 ± 45
OxA-6963charred food remains, S6, δ13C = −24.6‰1575 ± 50
OxA-6964charred food remains, S7, δ13C = −24.8‰1475 ± 50
OxA-6965charred food remains, S8, δ13C = −25.1‰1685 ± 50

Comment (E. N. C. and R. A. H.): site B had five conventional dates from articulated animal burials, one date from charcoal and an archaeomagnetic date, all but one centring on the first/second centuries cal AD. All the new AMS dates are younger than this. OxA-6944 and −6961 could just fit with a second century AD occupation, but OxA-6963 and −6964 are again three to four centuries too young compared to the conventional dates and archaeological expectation.

General Comment (E. N. C. and R. A. H.): charred food residues can be used as a dating tool, and in general terms seem to give dates comparing tolerably well with controls from more conventional dating material. However, a proportion of the determinations appear to be too young. Chemical analysis showed that the charred food remains consisted of varying proportions of elemental carbon and surviving fatty acids (lipids and alkanes). We suspect that some of the fatty acid component is causing the problem here, perhaps because of differential cross-linking with humic acids, but further work will be necessary to confirm this. For further discussion see Campbell et al. (forthcoming).



Sample of charcoal from Knowth, Co. Meath, (53:42N 06:30W) Ireland, submitted by E.G. McCormac, Radiocarbon Dating Laboratory, Queen's Univ., Belfast, with comments by G. Eogan, Knowth Excavations, Univ. College, Dublin.

OxA-7786charcoal, UB-4090, δ13C = −25.3‰4890 ± 40

Comment (G. E.): the sample was collected from the basal structural layer of Site 1 (the large mound). This layer mainly consists of humus, which was stripped from the surrounding area. It was generally formed naturally but on the whole it is considered as representing material that was either contemporary with, or a little earlier, than the date at which the large mound was built. Taking the 14C determinations from other passage tombs (except Carrowmore), OxA-7786 is a little earlier then “normal”. However, it must be emphasised that we only have a limited number of 14C results from passage tombs. In summary, the sample came from material that would have been accumulating in the period immediately preceding the construction of Knowth, Site 1.


Sample of dog bone from Knowth, Co. Meath (53:42N 06:30W), Ireland, submitted by G. Eogan, Knowth Excavations, Univ. College, Dublin.

OxA-7670dog bone, K95 AB 32, δ13C = −19.9‰1175 ± 35

Comment (G. E.): this dog ‘burial’ was found in the basal fill of a ditch that surrounded a habitation site. The main layer of occupation in the ditch dated to about the tenth century. The base of the ditch was on average 3 m below this layer. The bulk of the fill consisted of “softish” material, generally darkish in colour.

The archaeological evidence (i.e. finds) indicates that the ditch was dug in about the seventh century and as the bones were close to the base that ‘should’ also be the date for them. The only explanation for the date is that a pit, close to 4 m deep, was dug downward from the ‘Norman’ level (late twelfth century on historic dating) but why such a deep pit was dug to insert the remains of a dog is puzzling. Nevertheless, that appears to be the only explanation. No evidence for a pit came to light but the soft dark nature of the fill would have made detection very difficult, if not impossible.


Sample of bone from Ballymacaward, Ballyshannon, Co. Donegal (54:31N 08:15W), Ireland, submitted by E. O'Brien, 7 Ailesbury Lawn, Dundrum, Dublin 16.

OxA-8430bovid tibia, 97E0154, δ13C= −20.8‰1495 ± 40

Comment (E. O.): the sample was recovered from the upper levels of an area which also yielded spreads of charcoal and very fragmented cremated bone. A radiocarbon determination obtained from the charcoal (UB-4425) indicates that it was deposited possibly in the second or third century AD (O'Brien 1999). However the result obtained from the animal bone indicates that this was deposited at a somewhat later period.

In view of the absence of any evidence for habitation at or around the burial site, it has been concluded that this bone probably formed part of a funerary offering or funerary feast associated with the early medieval inhumation burials. This result is interesting in that in Ireland, although we have early literary references to the practice of funerary feasts in the early Medieval period, this is apparently the first dated archaeological sample on record.

Toormore Wedge Tomb

Sample from Toormore Wedge Tomb, (51:32N 09:38W) Ireland, submitted by William O'Brien, Dept. Archaeology, Univ. College, Galway, Ireland.

OxA-3996seeds, Triticum type, Context 13, 06, δ13C = −23.1‰390 ± 65

Comment (W. O'B.): the sample was submitted to date a late deposition in this megalithic tomb chamber. It relates to a deposit of mixed cereal grain at the site in later Medieval times. The circumstances of this act remain unclear, but the result does provide evidence of cereal cultivation in this part of south-west Ireland in late Medieval times.

Camp Varouf, Guernsey

Samples of charcoal from Camp Varouf (49:27:26N 02:39:20W), Guernsey, submitted by B.W. Cunliffe, Inst. Archaeol., Univ. Oxford.

OxA-7977charcoal, CV98, δ13C = −23.9‰6240 ± 50
OxA-8328charcoal, CV98, δ13C = −25.2‰5185 ±45

Comment (B. W. C.): the two determinations are for samples taken from the lowest level of a stratified Neolithic site exposed by coastal erosion on Guernsey. The results obtained are consistent with other evidence suggesting that the Neolithic colonization of Guernsey paralleled the development of the earliest Neolithic cultures in adjacent parts of France.


Courtavant, Aube

Samples were taken from early collections in the British Museum in the hope of dating two important Continental grave groups with diagnostic artefacts, thus to lend further support to the Bronze Age chronological revisions of recent years. Submitted by S. Needham, Dept. Prehist. & Romano-Brit. Antiq., Brit. Mus., London.

The samples were taken from Courtavant, Aube (48:30N 03:30E), France and Harthausen, Baden-Wurttemburg, Germany (see this Arch. List). A boar's tusk and tooth found on the chest of an extended inhumation uncovered beneath a tumulus in 1871 (BM registration ML1180, Morel collection). Other grave goods include a Rixheim sword, ribbed pin and Urnfield single-edged knife, an association datable to Reinecke Bronze D.

OxA-6221boar tooth, ML 1180, δ13C = −21.1‰895 ± 50
OxA-6721boar bone, ML1180, δ13C = −19.9‰1035 ± 40

Comment: (S. P. N.): the original date and the double-check are consistent with one another but considerably later than the date of the grave (c. 1350–1200 BC). The burial rite as described is of a format known in the region and there was no indication of later intrusion. Although it was seen that the tusk had received some conservation treatment, including some reconstruction, the area sampled was thought to be clear of consolidants. It must now, however, be considered most likely that contamination from modern (post-bomb) consolidants are the cause of the young dates.

Le Yaudet

Samples of charred seeds from Le Yaudet, Cotes d'Amor (48:80N 03:30W), France, submitted by B.W. Cunliffe, Inst. Archaeol., Univ. Oxford.

OxA-6716charred seeds, LY/F114(8), δ13C = −19.2‰1410 ± 45
OxA-6717charred seeds, LY/F195(3), δ13C = −21.3‰1325 ± 55

Comment (B. W. C.): the two samples from Le Yaudet were taken from corn-drying ovens associated with a phase of very early Medieval agriculture. The dates are entirely consistent with the stratigraphy of the site in that the ovens post-date occupation of the fourth and early fifth centuries AD and pre-date a phase of Medieval settlement beginning in the tenth century AD.


Roc d'Enclar

The site of the Settlement of Roc d'Enclar is located in the south of Andorra on the top of a granitic-mountain at an altitude of 1200 m (42:30N 01:30E). Excavated by the Andorran government, Patrimoni Cultural d'Andorra, between 1979 and 1992 the site gave a very wide chronology, between the final Bronze Age and the thirteenth century AD, with a hiatus of unknown occupation between the final Bronze Age and the first century AD. From previous 14C dating we can corroborate an intensive concentration of occupation between the fifth and the eighth centuries AD. The problem of dating the construction of the church remained, so it was decided to collect another sample. Among twelve samples collected to study the site sample N became one of the most important to date the main building on the site. It was decided to take charcoal embedded in a limestone pavement inside the church. Samples submitted by X. Lovera Massana, Patrimoni Cultural carretera de Bixessari s/n, Aixovall, St. Julià de Lòria, Andorra. Comments by M. A. Ruf Riba, Patrimoni Cultural carretera de Bixessari s/n, Andorra.

OxA-5921vegetal charcoal, Sample N, δ13C = −23.9‰1300 ± 50

Comment (M. A. R. R.): the results gave a great deal of information and advanced research into the evolution of the site. Combined with the results on samples taken from graves next to the church, OxA-5921 indicated when the church was constructed: between the end of the seventh and the beginning of the eighth century AD. The publication of the research on the Roc d'Enclar (Lovera Massana et al. 1997) contributed to the study of the landscape and communities for a period previously unknown (fifth to ninth centuries AD), especially in the Pyrenees.


Cova de l'Or

Samples from Cova de l'Or, Beniarrés, Alicante (38:50N 00:22W), Spain, submitted by J. Zilhão, Instituto Português de Arqueologia, Lisbon, Portugal with comments by J. Zilhão and B. Martí, Servei d'Investigació Prehistòrica, Valencia, Spain, through the NERC-funded ORADS facility.

OxA-10191Triticum aestivum-durum seeds, level 14, square J4, δ13C = −22.9‰6275 ± 70
OxA-10192Triticum aestivum-durum seeds, level 17, square J4, δ13C = −23.8‰6310 ± 70

Comment (B. M. and J. Z.): previously available dates for the Cova de l'Or's Cardial levels were conventional results obtained first in the 1960s, from cereal seeds collected in the 1955–58 excavations and second in the 1980s, from bulk charcoal samples collected in the 1975–76 excavation of Sector J (Martí et al. 1980). The former were 6510 ± 160 BP (KN-51), for the basal Cardial levels, and 6265 ± 75 BP (H-1754/1208) for their upper part. The latter were 6720 ± 380 BP (GANOP-C13) and 6630 ± 290 BP (GANOP-C12) for levels 16–17 and 14–15 of square J4, respectively, which also corresponded to the lowermost and uppermost sections of the Cardial deposit.

The contemporaneity between the Early Neolithic context from the Portuguese cave site of Galeria da Cisterna (Almonda karstic system) (Zilhão et al. 1991) and the earliest Cardial levels from Cova de l'Or was suggested in the framework of the pioneer colonization model for the spread of agriculture across west Mediterranean Europe (Zilhão 1993; 1997; 2000). In order to test this hypothesis, it would have been ideal to direct date Or samples consisting of the same kinds of Cardial ornaments from Cisterna that were submitted to ORAU (see comment in this Arch. List), which are characteristic of those levels (Pascual 1998). For curatorial reasons, this was not possible. Therefore, it was decided to submit samples each composed of two charred wheat seeds collected in levels that had supplied those kinds of ornaments. The seeds were classified by G. Pérez-Jordá as Triticum aestivum-durum and were collected among the material previously studied by P. López (cf. Martí et al. 1980).

OxA-10191 (5459–5002 cal BC, 95% prob.) and −10192 (5469–5067 cal BC, 95% prob.) are statistically identical They are also more precise than those previously obtained by conventional methods and indicate that the Cardial deposit accumulated at the site during the time interval between c. 5500 and c. 5000 cal BC. This deposit is characterized by the overwhelming predominance of Cardial decorated ceramics, although other kinds of impressed, incised and corded decorations are also represented. The flint lithic assemblage is made of blade/bladelet blanks and comprises sickle elements, borers, trapeze and crescents. Bone tools include awls, spoons, rings and pendants, and ornaments made of marine shells are very common. Abundant remains of cereals and domestic animals (sheep/goat) document a fully developed agro-pastoral economy.

The dates for Cisterna fall within the lower end of the Cova de l'Or range. Consequently, it can be concluded that the archaeological hypothesis according to which the dated Cisterna ornaments and associated Cardial ceramics indicated both contemporaneity and close cultural links with the Valencia region is independently verified by radiocarbon. In conjunction with other results obtained in Iberia through the direct AMS dating of diagnostic samples (cereals, animal domesticates, human remains, bone artefacts), the Or and Cisterna determinations suggest a very rapid spread of the Cardial, with settlements extending from Catalunya to central Portugal developing within the few decades centered around 5400 cal BC.

Arc de Berà

The Arc de Berà is a Roman monument at Roda de Berà (41:10N 01:29E), Tarragona, Spain built at the end of the first century before Christ. In recent centuries it has been restored on various occasions.

The question was posed of whether the ashlars with the inscription on the north-east face of the monument were in their original position, i.e. from the Roman epoch, or had been altered during the 1840 restoration. These ashlars had a covering of lichens associated with oxalate deposits. Seven samples were dated in order to try to solve the position or positioning of certain ashlars in the 1840 restoration. The samples were submitted by J. Girbal and R. Rocabayera, Saxum, Universitat Autònoma de Barcelona, Unitat de Botànica, Bellaterra 08193, Barcelona, Spain.

OxA-6531oxalate, 062093, δ13C = −23.9‰119.6 ± 0.7 % mod
OxA-6532oxalate, 062094, δ13C = −25.9‰108.2 ± 0.9 % mod
OxA-6533oxalate, 062095, δ13C = −18.2‰3505 ± 70
OxA-6534oxalate, 062096, δ13C = −29.1‰112.3 ± 0.8 % mod
OxA-6535oxalate, 062097, δ13C = −23.1‰110.4 ± 0.7 % mod
OxA-6536oxalate, 062098, δ13C = −28.3‰365 ± 70
OxA-6537oxalate, 062099, δ13C = −28.6‰3880 ± 70

Comment (J. G. L.): the results from the dating of the biodeposits alone do not allow for the dating of the ashlars. However when considered in combination with the archaeological evidence it is possible for the ashlars that still remain in their original Roman position to be differentiated from those repositioned in successive restorations. However, there were variations in their chronology, with on occasion a sub-contemporary age and on other occasions an absolute age older than the age of the monument itself.

The results were presented at 17th International Radiocarbon Conference, and two different hypotheses about the possible ageing mechanisms are discussed. Either the carbon is absorbed by the lichenic metabolism from the burning of fossil fuels, or it is incorporated directly from the carbonated substrates on which the lichens grow.


Galeria da Cisterna

Samples from Galeria da Cisterna (39:30:23N 08:36:49W), Almonda karstic system, Portugal, submitted by J. Zilhão, Instituto Português de Arqueologia, Lisbon, Portugal with comments by J. Zilhão and B. Martí, Servei d'Investigació Prehistòrica, Valencia, Spain.

OxA-9287Cervus elaphus pierced canine tooth, AMD2-G21–2278, δ13C = −19.2‰6445 ± 45
OxA-9288bone bead imitating the shape of a red deer canine, AMD2-F19–52, d13C = −21.4‰6445 ± 45

Comment (J. Z. and B. M.): the thin Holocene deposits in area AMD2 of Galeria da Cisterna were excavated in 1988–89 (Zilhão et al. 1991). The archaeological assemblage recovered therein featured baroquely decorated Cardial pottery stylistically similar to that from the Cova de l'Or, as well as a similar range of ornament types. Among the latter, there were 10 pierced red deer canines and 5 bone beads imitating their shape. In Portugal, these two kinds of ornaments are completely unknown in Mesolithic and later Neolithic or Copper Age sites. This association suggested, first, that the ornaments had been deposited in the cave in the framework of its use as a burial ground of the earliest agro-pastoral settlers of littoral-central Portugal; and, second, pointed to a strong connection with the Valencian Cardial, where those settlers probably had their ultimate cultural background, as argued in the framework of the pioneer colonization model for the spread of agriculture across west Mediterranean Europe (Zilhão 1993; 1997; 2000).

The Cisterna AMD2 Cardial context, however, was part of a palimpsest accumulated over time which also included later Copper Age, Bronze Age and Iron Age pottery. Therefore, it could not be dated on the basis of samples of the associated faunal and human bone material. Recent publication of the ornaments from the Valencian early Neolithic (Pascual 1998) showed that pierced red deer canines and bone beads imitating their shape were exclusively represented in the initial part of the sequence. This strengthened the hypothesis that the identical artefacts from the Galeria da Cisterna were indeed associated with the stylistically early Cardial pottery found therein. This hypothesis was tested by AMS direct dating of one specimen of each type of pendant, which provided the exact same result 5477–5321 cal BC (95% prob.).

The results are staliscally identical to those previously obtained on conventional cereal samples from Cova de l'Or (Martí et al. 1980) and to the new OxA results obtained for new samples from the same levels (see corresponding comment in this Arch. List).

Casinha Derribada, Monument 3

A sample of charcoal from Casinha Derribada 3 (40.43N 08:58W). Portugal, submitted by D. J. da Cruz, Instituto de Arqueologia, Faculdade de Letras de Coimbra. Coimbra, Portugal.

OxA-5291charcoal 14, δ13C = −25.1‰2985 ± 60

Comment (D. J. C): monument 3 of Casinha Derribada is part of a complex of five tumuli of small proportions, quite low in height, constructed with local materials, mainly quartz. The utilisation of this material contrasts with the local geographic context. Each monument differs from the others in its internal structures. The tumulus of monument 3 covered a pit opened from the substratum, which was capped with an engraved granite flagstone. Four ceramic vases were found inside the pit, identifiable to the Late Bronze Age (c. 1400–800 BC). The tumular complex is interpreted as being of funerary and ritual character (Cruz et al. 1998). Two radiocarbon determinationss were obtained previously (Arch. List 19). OxA-5291, as well as the previous one (OxA-4910), was taken from charcoal collected from the top of the closing flagstone and is attributable to a ritual fire. The results are statistically similar and the chronology of the site can be placed at the beginning of the Late Bronze Age. This chronology is confirmed by another result (GrN-21303: 3120 ± 110 BP) from the same context and using the same type of material.

Orca das Castonairas

Samples of charcoal from Orca das Castonairas, V. N. de Paiva, Portugal (40:00:16N 07:54:42W), submitted by D. J. da Cruz, Instituto de Arqueologia, Faculdade de Letras de Coimbra, Coimbra, Portugal.

OxA-7432vegetal charcoal, 18, δ13C = −24.9‰4510 ± 45
OxA-7433vegetal charcoal, 30, δ13C = −23.8‰4590 ± 55
OxA-7434vegetal charcoal, 33, δ13C = −25.6‰4470 ± 45
OxA-7435vegetal charcoal, 33, δ13C = −25.6‰4365 ± 50
OxA-7436vegetal charcoal, 56, δ13C = −24.5‰3365 ±45

Comment (D. J. C): Orca das Castonairas is a megalithic monument of great dimensions; it stands about 300 m from the left bank of the Paiva river. It is a dolmen with a passage-grave embraced by an earth tumulus superficially covered with stones. The megalithic structure of the monument was excavated in 1964 by V. Leisner and L. Ribeiro (Leisner, 1968, 41–51). In the region, this type of monument has been placed at the beginning of the fourth millennium BC (4000–3700 cal BC). The earlier determination (GrN-4924: 5060±50 BP), obtained from carbonized cork taken from the base of the funerary chamber, seems to confirm this chronology. In 1996, archaeological fieldwork was undertaken at the entrance of the monument (atrium and passage inside the tumulus). Here, structures were identified in this intact area connected with its access and to the rituals being performed outside the burial area. The fieldwork revealed lithic and ceramic materials, with the latter especially common. The monument was ritually closed up by a layer of stones that sealed the entrance.

The set of determinations is related to the ritual activities that took place within the atrium and in the intratumular passage; fire was important in this context. The final utilization of the monument, in general terms, may be placed in the second half of the fourth millennium BC, no doubt being related to the previous date, which was also from carbonized cork taken from inside the chamber (GrN-4925: 4610 ± 50 BP). This chronology is also confirmed by another set of results (GrA-9307: 4540 ± 50 BP, GrA-9308: 4440 ± 50 BP, GrA-9312: 4520 ± 50 BP and GrA-9313: 4380 ± 50 BP), obtained from the same type of material and from the same context. OxA-7436, as well as GrA-9314: 3250±50 BP was on material taken outside the atrium, perhaps through later disturbances of the soil.



Samples of charcoal from the site of Alleur, near Liège, Belgium, submitted by M Otte, Préhistoire, Univ. Liège, Liège, Belgium.

OxA-10221charcoal, 001–1 A2-(0–40), δ13C= −23.4‰6190 ± 60
OxA-10287charcoal, 002–26C-(20–30), δ13C = −24.0‰6145 ± 75

Comment (M. O.): OxA-10221 and −10287 both fit very well with what we expected, late sixth millennium BC, as often given for the Early Neolithic in Western Europe. When calibrated, they are slightly younger than expected (about one century). This means that these populations stayed a little longer in the region then previously thought.


Two samples of wood from the water-logged site of Melsele-Hof (51:15N 04.17E), Belgium. Submitted by L. Keeley, Dept. Anthropol., Univ. Illinois, Chicago, Illinois, USA.

OxA-3087wood, 1, δ13C = −26.0‰5130 ± 80
OxA-3092wood, 2, δ13C = −26.0‰4950 ± 80

Comment (L. K.): this site has yielded Late Mesolithic stone artefacts and an idiosyncratic, poorly-fired pottery. Four artefacts attributable by typology and raw material to the Early Middle Neolithic were also recovered. The radiocarbon samples were from two separate pieces of preserved bark found in the bottom of a bark-lined pit (Feature 2). The fill of this pit contained Mesolithic lithics and some very small fragments of the Melsele ceramic. The date of the pit would imply a minimum age for the undated ceramics which are suspected of being Mesolithic in age and association. The results obtained here indicate that the pit dates to the brief Early Middle Neolithic occupation rather than to the Mesolithic period.

Via Mansuerisca

A sample of wood from Via Mansuerisca (50:32N 06:04E), Belgium, submitted by M. Otte, Préhistoire, Univ. Liège, Belgium to date a Merovingian road made of wood and stone in the Fagne region of Belgium.

OxA-4948wood, δ13C = −24.9‰1165 ± 55

Comment (M. O.): the result corresponds well to the Merovingian phase that was estimated for the road (Otte and Streel 1994).



A sample of charcoal from Bernstorf (48:25N 11:37E), near Freising/Munich, Germany, submitted by R. Gebhard, Prahistorische Staatssammlung, Postfach 220028, D-80535 Munchen, Germany.

OxA-8361?oak charcoal, Bernstorf E-Nr98,39a, δ13C = −25.6‰2995 ± 40

Comment (R. G.): in 1998 a spectacular gold hoard was found in a Bronze Age hillfort at Bernstorf. The gold hoard contains the oldest ‘crown’ north of the Alps. One piece of twisted goldfoil contained the piece of charcoal dated (OxA-8361). This date is essential to approximate the age of the gold hoard (Gebhard 1999).

Slovak Republic

Cejkov I

Samples of charcoal collected during excavations at Cejkov 1 (48:30N 21:50E), Slovak Republic in 1960 by L. Banesz. The sample was taken at a depth of 120–135 cm in a feature known as ‘Objekt 3’, provisionally interpreted as a dwelling. The sample was made available by J. Hromada, Inst. Archaeol., Slovak Academy of Sciences, Nitra, Slovak Republic and submitted by A. Verpoorte, Faculty of Archaeology, Univ. Leiden, the Netherlands.

OxA-10267charcoal, Cejkov 1–2, δ13C = −25.1‰1804 ± 36

Comment (A.V.): the site of Cejkov I is located in the vicinity of the obsidian outcrops in Eastern Slovakia. The feature ‘Objekt 3’ contained obsidian artefacts that are typical for the Gravettian. However, the dated charcoal is clearly not associated with these Middle Upper Palaeolithic artefacts. It may indicate that the feature is not of a Palaeolithic age either.


Casale di Valleranello

Sample of charcoal from the pebble pavement located near the tenth km of the via Laurentina in the southeast part of the territory of Rome, the open-air site of Casale di Valleranello (Tenuta di Vallerano) lies on the left bank of the floodplain of the Rio Petroso river at about 57 m a.s.l. Submitted by A. P. Anzidei and G. Carboni, Soprintendenza Archeologica di Roma and Dipartimento di Scienze Storiche, Archeologiche e Antopologiche dell’ Antichita, Universita di Roma ‘La Sapienza’.

OxA-8916charcoal, 2 (VL98-Hv373 III), δ13C = −25.0‰5280 ± 45

Comment (A. P. A. and G. C): at this site two large portions of a palaeosurface with pits, pot-holes, a pebble pavement and a trench have been brought to light. OxA-4328 (4320–3960 cal BC) dates charcoal from the inside of a pit (Arch. List 19). OxA-8916 (4230–3980 cal BC) on charcoal from the pebble pavement confirms that the structures belong to a single stratigraphic horizon belonging to a late phase of the Ripoli culture (Anzidei and Carboni 1998)


Samples of charcoal from tombs 3 and 4 in Eneolithic cemetery of Lunghezzina, located at Tenuta della Lunghezza, at the seventh km of the motorway A24 Roma l'Aquila. Two groups of hypogaeous burial structures and an isolated single tomb have so far been excavated.

The ceramic set can be referred to an initial phase of the Eneolithic of central Italy. Submitted 1998–1999 by A. P. Anzidei and G. Carboni, Soprintendenza Archeologica di Roma and Dipartimento di Scienze Storiche, Archeologiche e Antopologiche dell’ Antichita, Universita di Roma “La Sapienza”.

OxA-8078Acer sp. charcoal, 1 (tomb 3), δ13C = −26.1‰4740 ± 45
OxA-9004Ulmus sp. charcoal, 2 (tomb 4), δ13C = −16.1‰381 ± 39

Comment (A. P. A. and G. C): OxA-8078 (3640–3370 cal BC) confirms the chronological attribution of the funeral sites to the middle of the fourth millennium, in agreement with the cultural complexes of the Fontenoce di Recenati settlement and the sixth stratum at Attiggio di Fabriano (dated to c. 3700–3100 cal BC). OxA-9004 (cal AD 1440–1640) from the upper part of the infill of the isolated tomb 4 is in contrast to the funeral set, attributable to an initial phase of the Eneolithic. Probably the presence of the charcoal is due to a post-depositional infiltration during historical periods (Anzidei and Carboni, in press: Anzidei et al. in press).

Cava di Selce

Sample of charcoal from the Cava di Selce, Rome, located in the south-east part of the territory of Rome, near the eleventh km of the via Laurentina, submitted in 1998 by A. P. Anzidei and G. Carboni, Soprintendenza Archeologica di Roma and Dipartimento di Scienze Storiche, Archeologiche e Antopologiche dell’ Antichita, Universita di Roma ‘La Sapienza’.

OxA-8064Cornus sp. charcoal, 1(CS98-Clt84 SE-V-tgl), δ13C = −23.5‰4345 ± 45

Comment (A. P. A. and G. C): OxA-8064 (cal 3090–2880 BC), places the Cava di Selce settlement between the end of the fourth and the beginning of the third millennium BC according to the cultural aspects here documented. The pottery sherds can be referred to the end of the Connelle facies (the site dates to 4310 ± 70 BP, cal 3350–2650 BC), and to the beginning of the Laterza culture (Castel Baronia–AV, 4400 ± 80 BP, cal 3340–2880 BC), when the “scale” treatment of the pottery surface increases (Anzidei and Carboni, in press).


Samples of charcoal from the infill of a pit at Torrino-Mezzocammino 1, located in the south-east part of the territory of Rome, at the twelfth km of the Cristoforo Colombo road, and from the ditch, level 6, V tgl. at Torrino-Mezzocammino 2, located at a distance of about 500 m from the Torrino Mezzocammino 1. Submitted 1999 by A. P. Anzidei and G. Carboni, Soprintendenza Archeologica di Roma and Dipartimento di Scienze Storiche, Archeologiche e Antopologiche dell’ Antichita, Universita di Roma ‘La Sapienza’.

OxA-8934Crataegus sp. charcoal, 1-TM1-Dt439SE/V tgl Pit 5, δ13C = −25.0‰.4445 ± 50
OxA-8991Phillirea sp. charcoal, 1-TM2-L V tgl, δ13C = −25.8‰4310 ± 70

Comment (A. P. A. and G. C): at Torrino-Mezzocammino 1 a portion of a palaeo-surface, with pits of various size, was brought to light. The typological characteristics of the pottery testify to a long period of usage of the site, from early Eneolithic to more recent aspects comparable to the Connelle/Rinaldone facies. OxA-8934 (3330–2920 cal BC) has been obtained on charcoal from pit 5, where a jug and a ‘Rinaldone type’ bowl were found. This dating can be correlated with the C phase dating at the Conelle ditch (Cazzella and Moscoloni 1999).

Numerous pottery sherds referring to the Conelle/Rinaldone facies and lithic instruments have been found at Torrino-Mezzocammino 2, a natural ditch filled during the Eneolithic period. OxA-8991 (3350–2650 cal BC) can be related to the D phase dating of the Conelle ditch (Cazzella and Moscoloni 1999). This site is therefore more recent than the Torrino Mezzocammino 1 site.

Casali di Porta Medaglia

Samples of charcoal from the palaeosurface at Casali di Porta Medaglia, Rome. Submitted 1999 by A. P. Anzidei and G. Carboni, Soprintendenza Archeologica di Roma and Dipartimento di Scienze Storiche, Archeologiche e Antopologiche dell’ Antichita, Universita di Roma “La Sapienza”.

OxA-7706charcoal, 1:PM97-M28SW(III)H, δ13C = −25.0‰5270 ± 50
OxA-7707charcoal, 2:PM97-L28SE(IV)F, δ13C = −25.9‰5300 ± 50

Comment (A. P. A. and G. C): the Neolithic settlement is situated in the SE part of the territory of Rome, between the via Ardeatina and the via Laurentina along the Rio Petroso river. Only a portion of the palaeo-surface, with a pit and a few potholes, has been excavated. The pottery is characterized by vessel forms and typological elements such as tubular handles, some with “extremities expanded as lobes”, which recall a late phase of the Ripoli culture (Ripoli III-Fossacesia).

Quadrato di Torre Spaccata

Sample of seeds collected from sector M10W of the Neolithic settlement at Quadrato di Torre Spaccata, located in the SE territory of Rome, on the slope of Alban Hills, where a portion of a palaeo-surface with some structures such as levels of baked clay, sub-rectangular pits, have been excavated. Submitted in 1995 by G. Carboni and A. P. Anzidei, Soprintendenza Archeologica di Roma and Dipartimento di Scienze Storiche, Archeologiche e Antopologiche dell’ Antichita, Universita di Roma “La Sapienza”.

OxA-6049Triticum aestivum/durum charred seeds, QTS M10 W, XV 10.10.85, d13C = −23.7‰5270 ± 50
OxA-6050Triticum sp. charred seeds, QTS M10, 3.4.85, δ13C = −24.0‰5280 ± 50

Comment (A. P. A. and G. C): the vast amount of pottery sherds show an articulated vessel typology, which places the site between the earliest manifestations of the Chassey-Lagozza culture in central Italy, according to the OxA-6049 and OxA-6050 results obtained from the palaeo-surface. The lithic industry, both on flint and obsidian, is characterized by a strong incidence of laminar products and by the presence of arrowheads (Anzidei and Carboni 1998).

S. Croce Cave

Samples of charcoal and charred seeds from Grotta S. Croce Cave (41:10:37N 16:27:59E), Italy, submitted by P. Gambassini, Departmento Archeologia, Sezione di Prehistoria, Siena University, Siena, Italy.

OxA-7594charcoal, SCR 69, δ13C = −24.5‰6345 ± 45
OxA-7595charcoal, SCR 72, δ13C = −24.0‰6375 ± 50
OxA-7596charred seeds, SCR 78, δ13C = −22.3‰6555 ± 50
OxA-8077charcoal, SCR US83, δ13C = −25.1‰135 ± 35

Comment (P. G.): the excavation of S. Croce Cave was stopped in 1958 and begun anew three years ago. These are the first dates of a Neolithic layer of about 80 cm thickness. Three dates are consistent with a late phase of ‘impressa’ ware and in good agreement with the same cultural period in south Italy. OxA-7596 is just above an important straw artefact. Dating a silt level under the Neolithic layer, OxA-8077 was expected to be older than OxA-7596.

San Lorenzo

Samples of human bone and wood from the excavations conducted by M. De Min of the Soprintendenza per I Veni Ambientali e Architettonici at the site of the church of San Lorenzo di Castello in the Castello district of Venice (45:12N 12:21E). According to historical records, a church was first build on the site at the beginning of the ninth century. The samples come from the excavations made on the west side of the site in 1995 and 1996, which examined the early levels associated with the church and the underlying stratigraphic units related to land reclamation at the site. The samples were submitted by A. J. Ammerman, Dept. Classics, Colgate Univ., New York.

OxA-11356human bone, No. 11, δ13C = −18.5‰1371 ± 34
OxA-6661human bone, No. 13, δ13C = −18.2‰1135 ± 45
OxA-6747human bone, No. 12, δ13C = −18.4‰1210 ± 50
OxA-6768wood, No.9, δ13C = −25.8‰1295 ± 55
OxA-6769wood, No.10, δ13C = −25.4‰1220 ± 50
OxA-7011Wood, No.8, δ13C = −30.2‰1020 ± 70
OxA-7012wood, 14, 438cm, US 1764, δ13C = −26.0‰1270 ± 90
OxA-8198wood, No.8, δ13C = −27.9‰1075 ± 35
OxA-8199wood, 14, 438cm, US 1764, δ13C = −25.8‰1285 ± 35

Comment (A. J. A.): for the first seven samples (nos. 1–7) from the site dated at Oxford before, see Arch. List 15, 349; Ammerman et al. (1992); Ammerman et al. (1999). All of the seven new samples (8–14) give essentially the ages that we expected. Samples 8 and 14 were re-measured (OxA-8198 and −8199) in order to reduce the measurement error associated with OxA-7011 and −7012. At the 95.4% confidence level, all of the samples are older than AD 1000 and yet more recent in age than AD 650. In stratigraphic terms, the youngest sample (8) comes from a wooden structure supporting an early crypt of the church. A Heidelberg date done on the same outer rings of the same timber (Hd-17754: 1079 ± 26 BP), is in good agreement with OxA-7011 and −8198; together the three dates yield an age falling in the tenth century. The three bone samples (11–13) come from early burials at the west end of the church; they give calibrated ages in the seventh to tenth centuries. Note that the Basilica on the island of Torcello has produced a series of burials some of which are older in age (see Arch. List 30, 472–3). OxA-6768 dates a wooden pole (10 cm in diameter) that, according to the excavator, belongs to the first church on the site. OxA-6769 dates a small wooden stake driven into a canal bank on the west side of the site. Sample 14 comes from a dense layer of twigs and canes that is found at an elevation of −1.93 m with reference to the 1987 standard for mean sea level in Venice. It represents a phase of land reclamation on the site's west side at a time prior to the construction of the first church. Land reclamation appears to have begun somewhat earlier on the east side of the site, as shown by OxA-3334 to −3339 (see Arch. List 15). Note that the elevations cited previously for these samples, 2–7, refer to a local site datum and not to the 1897 tide-gauge standard commonly used in Venice.

Basilica of S. Salvatore

Sample of bone, found within the cavity of the stone architrave of the portal of the Basilica of S. Salvatore in Spoleto (42 45N 12:45E), Italy. The church has three aisles with a presbytery divided in three parts. The plan is typical of the basilican churches of the Constantine period, but the presence of some particulars, such as some characteristics of Christian buildings of the oriental provinces, refer to later buildings at Constantinopolis and Ravenna. These special characteristics are the use of materials from older Roman buildings, in the external and internal decorations (pillars, capitals, architrave and decoration elements in the façade).

In the course of a recent restoration (1996), made by the Soprintendenza ai Beni Ambientali, Architettonici, Artistici e Storici dell’ Umbria, a fragmentary ceramic cup containing animal bones was found in the filling materials of the stone architrave of the major portal. A bone sample was submitted by B.G. Brunetti, Dept. Chem., Univ. Perugia, Italy

OxA-10461?animal bone, δ13C = −19.5‰1664 ± 35
OxA-10462?animal bone, δ13C = −19.5‰1639 ± 32

Comment (B. G. B): in addition to addressing questions of archaeological and historical interest (the site is at about 2 km from the historical Roman center), the dates here reported help in documenting the building of the major portal of the ancient Basilica, one of the first Early Christian churches in Italy. In fact, the complexity of the monument, the numerous manumissions and restorations, the loss of the internal stucco-work and the absence of a systematic archaeological study make the dating of the church very problematic. The most accredited and shared hypothesis is that the first step of the construction of San Salvatore goes back to the age of Theodosius (end of fourth century) (Biganti 1994; Salmi 1951). However, recent studies by Jäggy (1998) and previously other work by Deichmann (1943) place the construction of the monument in the period between the beginning of the seventh century and the middle of the eighteenth century. According to the results, after calibration, the most probable age of the bones is between the middle of the fourth and the beginning of sixth century. Previous dating of another bone sample from the same recovery, carried out at the Radiocarbon Analysis Laboratory of the University of Utrecht, gave a calibrated date in substantial agreement with those of OxA-10461 and −10462.


A sample of charcoal from Montebelluna (45:77N 12:04E), Italy, submitted by L. Stievano, Dipartimento di Chimica fisica, Univ. Venice, but now at Laboratoire de Réactivité de Surface, 4 Place Jussieu, 75252 Paris, France.

OxA-10904charcoal, DM-03, δ13C = −24.1‰610 ± 40

Comment (L. S.): the analysed sample comes from a vase shaped black body having a diameter of 14 cm and an height of 10 cm, which was used as a building stone in an ancient farm house dating back to the sixteenth century. This material was discovered recently (in March 2001) during the restoration of the house, being part of a restored wall. The owner thought in the first place that the body was a meteorite, and asked the research group at the University of Venice to analyse and identify it.

Several samples were collected from it, and analysed by X-ray diffraction, Moessbauer spectroscopy and thermoluminescence. During the collection of the samples, several inclusions of charcoal were identified in the interior of the material (which were later sent to ORAU for radiocarbon dating). The weight of the material itself was rather light, contrary to that of chondrites.

The XRD analysis shows the presence of metallic iron, iron oxides, and other minor components (olivine among them). Moessbauer spectroscopy indicates the presence of metallic iron (not kamacite, the iron-rich iron-nickel alloy usually present in chondrites), iron carbides, olivine and two iron oxide/oxihydroxides (haematite and goethite). finally, the low sensitivity to the beta-irradiation in the laboratory suggests that the content of ‘phosphors’ is rather low. In addition the measurement of the original sample has a rather low thermoluminescence effect, and confident calculations with such low signals are not possible. However, the low content in phosphors and the low thermoluminescence effect are in disagreement with the typical behaviour of a meteorite.

To conclude, I think that we have shown by these methods that the material is not a meteorite, in agreement with the very recent age found by radiocarbon dating. On the contrary, its shape and the results of the different analyses point to the identification of this material as an old iron slag, probably coming from a former workshop existing not far away from the place where the sample was found. The age found by 14C dating, however, is in agreement with the supposed age of the material, where it was probably mounted on the original wall of the farmhouse.

Conelle di Arcevia

Sample of bone from Conelle di Arcevia, Arcevia, Marche (43:31:30N 12:57:40W), Italy, submitted by A. Cazzella, Universita ‘La Sapienza’, Rome, Italy.

OxA-6837horse bone, Co AE S, δ13C = −19.9‰3085 ± 50

Comment (A. C): the horse bone comes from surface levels of the Copper Age ditch at Conelle. The site was inhabited again during the Late Bronze Age, but modern agricultural works destroyed the structures of that period. The date is fully consistent with Late Bronze Age (Calderoni and Cazzella 1999, 178).



The site of Markiani on Amorgos (36:49N 25:54E), Greece, is of significance as a settlement of the Early Cycladic period. The excavations were conducted from 1986–89 by the Inter-University Research Project for Amorgos and Keros, under the direction of L. Marangou, Univ. loannina, Greece, C. Renfrew, Univ. Cambridge and C. Doumas, Univ. Athens. Samples submitted by C. Renfrew under the NERC-funded ORADS facility.

The excavations at the site yielded materials of four stratigraphic phases: Markiani I, ceramic material related to the Grotta-Perlos culture of the Cyclades (generally assigned to the Early Cycladic I period); Markiani II, ceramic material related to the Kampos Group (transitional between the Grotta-Pelos and Keros-Syros cultures: ECI/II transitional); Markiani III, ceramic material contemporary with the Keros-Syros culture (generally assigned to the Early Cycladic II period); Markiani IV, ceramic material closely related to the Kastri group (variously assigned to later Early Cycladic II or Early Cycladic III). Of the samples submitted OxA-3297, −4003 and −4004 are from phase II contexts, and OxA-3291, −4006 and −4007 are from phase IV contexts. The remaining samples are from contexts lying chronologically between those of the previous two groups, with OxA-3295 and −3296 assigned to Markiani phases II or III, and the remainder to phases III or early IV. These are here collectively designated Markiani III.

OxA-3291animal bone, T1, 1, 27, δ13C = −21.0‰3810 ± 80
OxA-3292animal bone, T1, 1, 37, δ13C = −21.0‰3920 ± 80
OxA-3293animal bone, T3, 2, 6, δ13C = −21.0‰4090 ± 90
OxA-3294animal bone, T3, 2, 8, δ13C = −17.6‰4060 ± 75
OxA-3295animal bone, T8, 1, 8, δ13C = −21.0‰4105 ± 80
OxA-3296animal bone, T8, 2, 7, δ13C = −21.0‰4080 ± 75
OxA-3297animal bone, T1, 1, 6, δ13C = −21.9‰4380 ± 100
OxA-4003animal bone, Ma89, T7, 8, δ13C = −19.9‰4390 ± 65
OxA-4004animal bone, Ma89, T7, 10, δ13C = −20.3‰4160 ± 65
OxA-4005animal bone, Ma89, T1, 1, 44, δ13C = −18.9‰3730 ± 65
OxA-4006animal bone, Ma90, T1, 4, 29, δ13C = −18.7‰3990 ± 65
OxA-4007animal bone, Ma90, T1, 4, 31, δ13C = −18.8‰3860 ± 60

Comment (C. R.): the radiocarbon determinations present no striking stratigraphic problems. OxA-4005 gives a date somewhat out of sequence with the others (i.e. later) and, at the suggestion of R. A. Housley (then at ORAU) it was not included in the averaging process. It may be intrusive stratigraphically or it might be an outlier statistically.

The dates were examined statistically by S. Manning of the University of Reading, using his Calmaker calibration programme. He produced the following weighted average calibrations for the three date groups above, excluding OxA-4005 from the averaging process:

Markiani II (three samples) 2904 ± 69 BC (cf. Kampos Group)

Markiani III (five samples) 2571 ± 57 BC (cf. Keros-Syros culture)

Markiani IV (three samples) 2377 ± 59 BC (cf. Kastri Group)

These dates are a valuable addition to the few dates available and relevant to the Early Cycladic period. It should be noted that Markiani phase I, with material of the Grotta-Pelos culture comparable to the material seen at Phylakopi in Melos in the earliest pre-City levels, must precede Markiani phase II.

Details of the stratigraphic context of each sample and of the pottery found in each phase will be published in the excavation report, currently in preparation.


Petromoutti slag heap

Sample of charcoal from Petromoutti slag heap (34:40N 33:05E), Cyprus, submitted by S. Stos-Gale, Isotrace Laboratory, RLAHA, University of Oxford.

OxA-6908charcoal, PTMS 32, δ13C = −21.9‰2280 ± 50

Comment (S. S.-G.): the island of Cyprus is remarkable not only from the point of view of the wealth of its mineral resources, but also because they were exploited through the last three millennia. One of the most interesting archaeometallurgical remains of these activities are numerous (perhaps highest in number per square kilometre in Europe) copper slag heaps. The extensive mining operations in the first half of the twentieth century followed earlier copper production on Cyprus for the previous three or four millennia. The large slag heaps are gradually disappearing following the intensified agriculture of the southern part of Cyprus after 1974. Within the framework of the lead isotope mapping of Cypriot copper deposits we have undertaken in collaboration with the Hellenic Mining Company (HMC) a project of recording and lead isotope fingerprinting of these slags.

Out of 85 known slagheaps in the records of the HMC about 50 samples have been collected. The lead isotope data and description of slag heaps is published in Stos-Gale et al. (1998). There is relatively little evidence for the chronology of the above mentioned sites. Most remarks in the geological publications describe them as ‘ancient’ or Roman without any scientific basis. The former division into Phoenician or Roman, based merely on red or black colour is without foundation; these colours are related to the degree of weathering. There are twenty-nine radiocarbon determinations for the Cypriot slag heaps and mines in the literature and they all fall into the 500 BC–AD 500 range (Stos-Gale et al. 1998, p. 246). The slag pieces from the Petromoutti-Yerasa slagheap located in the forest north of Limassol, have visual characteristics of a more primitive, i.e. earlier, copper slag. It was very difficult to find amongst them typical ‘spongy’ slag that contains the inclusions of charcoal.

OxA-6908 was obtained from a charcoal trapped inside the gas holes in one of the pieces of slag collected on the slag heap located on the east side of the road. We had only one date allocated and there are no other dates. So far this is the earliest 14C determination on copper slag from Cypriot slag from a non-settlement context. On the basis of this one determination we can suggest that perhaps copper was smelted on this site in the first millennium BC, but without further evidence, either as 14C dates or dateable pottery, it is impossible to say if this was the only period when Petromoutti was used for copper extraction. Most of the slag heaps dated by 14C (Stos-Gale et al. 1998) yield dates from the Roman period, so at least we now know that this particular site was earlier. It remains to be seen whether further archaeological work will be undertaken at this site.


Samples from Kissonerga, Mylouthkia (34:50N 32:20E), Cyprus, submitted by E. Peltenburg, Dept. Archaeol., Univ. Edinburgh. This site is located on the west coast of Cyprus and had been the subject of the University of Edinburgh's Lemba Archaeological Project investigations since the late 1970s. Assays were obtained to help resolve dates of its first three periods. Period 1 possesses artefact-rich well fills lacking pottery that were common throughout the remainder of the site. Since the artefacts were unlike those that characterise aceramic Neolithic occupations elsewhere on the island, independent dates were highly desirable. Dates were required of ostensible Period 2 pits located over 100 m away from the main concentration of Period 2 occupation. Lastly, it was important to date a destroyed Period 3 building in order to assess whether a gap existed in the evidence between the Early and the Middle Chalcolithic of Cyprus.

OxA-7460charred barley seeds, C482, δ13C = −23.0‰9315 ± 60
OxA-7461charred barley seeds, C531, δ13C = −23.1‰8185 ± 55
OxA-7462charred Pistaca seeds, C529, δ13C = −22.1‰4650 ± 50
OxA-7463charred Pistaca seeds, C525, δ13C = −24.0‰4710 ± 50
OxA-7464charcoal, R415, δ13C = −23.6‰4885 ± 45

Comment (E. P.): OxA-7460 to −7461: these Period 1 determinations from two water wells indicate that the wells were filled with material belonging to the Early and Late Cypro-PPNB. Well 116, therefore, belongs to some of the earliest farmers to have settled in Cyprus. This evidence helps to establish the great antiquity of the Khirokitian culture (see Peltenburg et al. 2000). OxA-7462 to −7463: these Period 3 dates demonstrate that the Middle Chalcolithic of Cyprus, with the earliest evidence for stone architecture in the Chalcolithic, follows without interruption from the preceding phase. cf. BM-1473, −1475, −1476, −1539, −1540.

OxA-7464: it would appear that this part of the site is essentially of the same date as the area to the north, and that the slightly different ceramics may be accounted for in functional terms. cf. BM-1473, −1475, −1476, −1539, −1540.



Samples of charcoal from Parta, Sard, Romania (45:40N 21:06E), submitted by M. Otte, Préhistoire, Univ. Liège, Belgium to date this Early Neolithic site (and sanctuary) in Eastern Romania.

OxA-7453charcoal, PAR-58, δ13C = −24.8‰6050 ± 80

Comment (M. O.): the result corresponds well to what was expected on the basis of the archaeology and other methods.



Samples from Çatalhöyük, Konya Region, (32:49:41N 37:40:01E) Turkey, submitted by C. Cessford and I. Hodder, Dept. Archaeol., Downing Street, Cambridge, supported by the NERC-funded ORADS facility.

OxA-9771Triticum charred seeds, KOPAL 2, δ13C = −21.8‰7965 ± 55
OxA-9772Triticum charred seeds, KOPAL 3, δ13C = −21.0‰8025 ± 55
OxA-9774Scirpus charred seeds, South 1, δ13C = −23.2‰7935 ± 50
OxA-9775Triticum/ Scirpus charred seeds, South 3, δ13C = −20.5‰8090 ± 55
OxA-9776Scirpus charred seeds, South 6, δ13C = −23.5‰7985 ± 55
OxA-9777Lens charred seeds, South 8, δ13C = −23.9‰8160 ± 50
OxA-9778Triticum/Pisum charred seeds, South 9, δ13C = −21 4‰8240 ± 55
OxA-9892Lens charred seeds, South 7, δ13C = −23.7‰8150 ± 50
OxA-9893Triticum/Scirpus charred seeds, South 10, δ13C = −22.0‰8155 ± 50
OxA-9943Triticum charred seeds, KOPAL 2, δ13C = −22.1‰7910 ± 55
OxA-9944Triticum charred seeds, KOPAL 3, δ13C = −20.9‰7975 ± 50
OxA-9945Scirpus charred seeds, KOPAL 5, δ13C = −23.0‰7775 ± 50
OxA-9946Scirpus charred seeds, South 1, δ13C = −23.3‰7980 ± 55
OxA-9947Triticum/Hordeum charred seeds, South 2, δ13C = −22.4‰7985 ± 50
OxA-9948Triticum/ Scirpus charred seeds, South 3, δ13C = −20.7‰8090 ± 50
OxA-9949Pisum charred seeds, South 4, δ13C = −21.7‰8050 ± 50
OxA-9950Triticum/Pisum charred seeds, South 5, δ13C = −21.4‰8030 ± 50
OxA-9980Triticum charred seeds, KOPAL 1, δ13C = −19.4‰7955 ± 75
OxA-10092Triticum charred seeds, KOPAL 1, δ13C = −22.2‰7185 ± 65

Comment (C. C. and I. H.): these determinations cover the earliest excavated deposits at Çatalhöyük and derive from two stratigraphic sequences, one on the mound itself (termed South) and another off-site area immediately to the north of the mound where extraction and other activities occurred (termed KOPAL) The on-site determinations form an internally coherent sequence suggesting that the earliest surviving deposits in the area excavated date to the period c. 7400–7050 cal BC (OxA-9893, −9778, −9777, −9892), which is in broad agreement with AMS results obtained previously from coring. The determinations then form a good sequence with the uppermost samples dating to the period c. 7050–6650 cal BC (OxA-9948, −9775, −9947, −9946, −9774). These can be assigned to Level XI and XII as defined by J. Mellaart in the 1960s, and are broadly in agreement with the general pattern of radiocarbon dates from his excavations. The determinations suggest a time-span towards the shorter end of expectations for the stratigraphic sequence but one that is perfectly reasonable. The determinations from the off-mound area indicate that activity occurred here during the period c. 7100–6650 cal BC (OxA-9944, −9772, −9943, −9771). This demonstrates that the off-site activity is broadly contemporary with the on-site sequence, confirming the impression created by aspects of the material culture assemblage. Ox A-9945 gives cause for concern as it was anticipated that this would be the earliest determinations from the off-site area. The date suggests that the cut-feature from which it was obtained may be a later intrusion. OxA-9980 and −10092, obtained from the same sample, are not in good agreement with each other, which suggests that the layer in question is a mixed deposit containing material of different ages. It now seems likely that this is a composite deposit consisting of alluvial material and buried soils. If OxA-9980 relates to an early stage of these processes and OxA-10092 to a later stage these dates are compatible with the other dating evidence from this area. Overall these radiocarbon determinations place the earliest excavated phases of Çatalhöyük in the Aceramic or Early Neolithic and indicate that it is broadly contemporary with Canhasan III(as was suggested by aspects of the material culture assemblage), Suberde and Musular.



Samples from Dereivka (48:54N 33:50E), Ukraine submitted by D. W. Anthony, Anthropology Dept., Hartwick College, Oreonta, NY 13820, USA.

OxA-6577horse bone, Cult stallion, δ13C = −21.0‰1995 ± 60
OxA-7185horse tooth, δ13C = −20.3‰2295 ± 60

Comment (D. W. A.): the samples were submitted to examine the important date of when horseback riding began. My colleague D. Brown and I had detected bit wear on the premolars of a stallion deposited with parts of two dogs in the Eneolithic settlement stratum at Dereivka. This particular horse had been the subject of earlier metric studies that suggested it was domesticated. The settlement where it was found was dated by radiocarbon to about 4300–3600 cal BC. Our studies indicated that it had been bitted. It was announced and widely accepted as the earliest direct evidence for bitting, and therefore probably for horseback riding, in the world. But our critics voiced doubts about whether the bitted stallion was a later intrusion in the Eneolithic stratum. We obtained the bone sample identified as OxA-6577 from a Ukrainian colleague who assured us that it came from the cult stallion, and we paid Oxford to test it. The result was very far from the expected age of about 6000 BP.

We then learned that the bone we had been given was not from the cult stallion, but just a bone fragment from the same excavation unit and depth in the Dereivka collections. (It was not even certainly from a horse!) We then asked for and obtained the actual tooth with bit wear from the cult stallion. This time we checked it against the cast replica that we had made when we studied the original in 1989. It was the same tooth. So we obtained a further AMS analysis on material from this tooth (OxA-7185). By then our colleague had taken a real sliver of bone from the cult stallion's skull and obtained a result from the Kiev laboratory of 2490 ± 95.

Our critics were right, the Dereivka horse with bit wear was a later Iron Age intrusion dug into the Eneolithic layer. We retracted our claim for Eneolithic horseback riding at Dereivka in an article published in Antiquity (Anthony and Brown 2000). However, we continue to insist that we have identified bit wear on Eneolithic horse premolars at another site, Botai in northern Kazakhstan; and recently we have found bit wear on a horse premolar from a second Eneolithic site in Kazakhstan, Kozhai I. So the debate continues.


Gornyj, Kargaly

Samples of bone from Gornyj, Kargaly ancient mining and metallurgical centre, S. Ural (52:17N 54:47E), Russia, submitted by E. N. Chernykh, Inst. Archaeol., Dm. Ulyanova 19, 117036 Moscow, Russia.

OxA-5648bone, 10/Sq.5228, 180–200 cm, dwelling 22, phase A, δ13C = −19.9‰3290 ± 50
BM-2963bone, phase A3380 ± 35
BM-2962bone, phase A3340 ± 35
BM-3147bone, phase A3090 ± 35
OxA-5649bone, 12/Sq.5228, 170–190cm, complex 1, sacral gallery, phase B-1, δ13C = −19.1‰3460 ± 70
BM-2945charcoal, phase B-13270 ± 40
BM-3148bone, phase B-13120 ± 40
BM-3156charcoal, phase B-13180 ± 45
BM-3149bark, Betula, phase B-13360 ± 35
BM-3016bone, phase B-13120 ± 45
OxA-5650bone, 14/Sq,5228, 140–160cm, complex 1, pit 2. phase B-1, δ13C = −19.9‰3315 ± 50
OxA-5647bone, 7/Sq 5228, 120–140cm, floor of living area, phase B-1, δ13C = −19.0‰3455 ± 60
OxA-5646bone, 5/Sq 5228c, 80–100cm, phase B-2, δ13C = −19.8‰3245 ± 70
BM-2964bone, phase B-23170 ± 50
BM-2965bone, phase B-33220 ± 35
OxA-5645bone 2/Sq 5228a, 40–60cm, phase B-3, δ13C= −18.6‰3375 ± 60

Comment (E. N. C): six samples were used for dating the main phases A and B and sub-phases B-1, B-2 and B-3. Ten samples were dated conventionally at the British Museum (Gorny et al. 1999). The dates fit well into a three hundred year period (1690–1390 cal BC at 68.2% prob.). However there is not good agreement between the different phases. Some results display clear contradictions relative to the stratigraphical position. For instance, compare the earliest small pit-dwelling 39 (BM-3147) which is evidently later than the collapsed roof of the smelting yard (BM-3149), or OxA-5648 (phase A) and OxA-5645 (the highest layer at Gorny). It appears that the total chronological framework of 14C dates did not provide an accurate definition of the phases at Gorny.

The richest cultural level, sub-phase B-3 (a huge scrapheap of living and productive waste), was redeposited into the foundation pit of the living and smelting complexes in the final period of Gorny, before all habitation ceased. Similar redepositing processes were very characteristic of the mode of life of the Kargaly aborigines (miners and metallurgists). In all probability it could satisfactorily explain some discrepancies between the stratigraphic position of samples and their 14C dates. This series is very important for the historical conclusion.

Klin Yar

Klin Yar is an Iron Age to Early Medieval cemetery and settlement site, Nr. Kislovodsk in the Stavropol region, North Caucasus, Russia (43:55N 42:40E). Joint Anglo-Russian excavations were carried out in the cemetery areas in 1994–96 (see interim report, Härke and Belinsky 2000).

The first series of three dates (OxA-7683, −7669 and −7757) was commissioned by H. Hárke, Dept. Archaeol., Univ. Reading, to clarify the date of skeleton 385 (OxA-7683) found outside a recognizable grave context, on top of the fill of the dromos (access corridor) of the late Sarmatian/early Alanic chamber grave 386 (OxA-7757 and −7669). The determinations suggest that the bodies in the chamber and the one on top were deposited virtually contemporaneously, implying the possibility of human sacrifice outside the chamber after burial of the two bodies inside. The seven further determinations were obtained through the NERC-funded ORADS facility.

OxA-7683human tooth, KYIII/385, δ13C = −17.2‰1795 ± 35
OxA-7669human tooth, KYIII/386/2, δ13C = −18.0‰1680 ± 35
OxA-7757human tooth, KYIII/386/1, δ13C = −17.2‰1725 ± 55
OxA-9380human bone, KYIII/363A, δ13C = −15.7‰1960 ± 45
OxA-9386human tooth, KYIII/341/1, δ13C = −16.8‰1741 ± 39
OxA-9387human tooth, KYIII/341/3, δ13C = −17.6‰1568 ± 37
OxA-9388human bone, KYIII/363B/2, δ13C = −17.3‰1507 ± 38
OxA-9389human tooth, KYIII/378, δ13C = −13.1‰1879 ± 38
OxA-9390human tooth, KYIII/360/1, δ13C = −16.7‰1685 ± 37
OxA-9391human tooth, KYIII/360/2, δ13C = −16.5‰1556 ± 39

Comment (H. H.): because the absolute dates obtained in the first series were substantially earlier (by about a century) than was expected from the grave-goods in the chamber, a second series was commissioned, funded through the NERC-funded ORADS facility, in order to clarify this discrepancy. For this new series, material was selected from graves with East Roman or Byzantine imports (incl. coins) which can be dated closely, or from otherwise very rich graves with a wide range of datable finds. This series confirmed the systematic discrepancy of about 100 years between archaeological and radiocarbon dates, but also demonstrated that the problem must be the latter: skeleton 3 in grave 341 (OxA-9387) had a Byzantine coin of 582–602 AD, but the calibrated radiocarbon date was 410–600 cal AD (at 95.4% probability), respectively; and grave 363B contained a Byzantine coin of 634–641 cal AD, but the calibrated date for skeleton 2 (OxA-9388) from this grave gives 430–640 cal AD (at 95.4% probability), respectively.

Given that most samples were human teeth this systematic deviation may be the consequence of particular components of the diet (millett or marine food). Concerning marine food, however, the site is in the foothills of the central Caucasus Mountains, at a considerable distance from both Caspian and Black Seas, and there are no major bodies of water nearby. The problem is still being investigated.


Hezekiah's Tunnel

Samples of well preserved fine twigs and small pieces of worked wood were recovered from the oldest of a number of plaster applications in Hezekiah's Tunnel, Israel (31:46N 35:14E). Submitted by J. Rosenbaum, Dept. Earth Sciences, Univ. Leeds, through the NERC-funded ORADS facility.

OxA-8522wood, SR.53, δ13C = −26.0‰2620 ± 35
OxA-8523plant remains, SR.61, δ13C = −25.0‰2505 ± 35

Comment (J. R): the two samples give dates roughly compatible with the accepted biblical dates for the construction of the tunnel during the reign of King Hezekiah (715–687 BC), thereby supporting the historicity of the Old Testament at least until that time.

A complex system of subterranean waterworks underlies the City of David, considered the most ancient part of Jerusalem. Tradition holds that in the tenth century BC King David (2 Samuel 5:6–9 and 1 Chronicles 11:4–7) made this area the first capital of his kingdom, by having his general Joab capture the city from the Jebusites. This was accomplished by entering the city through a secret water system that connected the Gihon Spring, located outside the ancient city wall, to a tunnel inside it. A second tradition claims that King Hezekiah (late eight century BC) had the tunnel dug in order to bring water from the spring into the walled city in preparations prior to the siege of the city by the Assyrian monarch Sennacherib (2 Kings 20:20 and 2 Chronicles 32:30). Both traditions, although well recorded in the Old Testament, nonetheless require verification since (1) the latter cannot generally be viewed as a reliable historical document and (2) unequivocal scientific or archaeological evidence for the legend is still lacking.

A school of biblical historians (‘The Minimalists’, Biblical Archaeology Review July/August, 1997) claim that the Old Testament (at least until the Persian period in the fourth century BC) is not a legitimate historical source and that kings David and Solomon (of the United Monarchy) are but mythological figures. N. P. Lemche, one of the principles of this school of scholars, claims that the Old Testament “hardly predates the Greco-Roman period”, that is, the third century BC (J. Hebrew Scriptures, 3, 2000).

Timna Mines

Sample of charcoal from the Timna Mines, Site F2 (29:47:40N 34:55:20E), Israel, submitted by B. Rothenberg, Institute of Archaeology, University College London.

OxA-5912charcoal, F2/76, 1501, δ13C = −22.4‰925 ± 60

Comment (B. R.): Site F2 is one of two very small smelting sites in the actual mining region of Timna (30 km north of the Gulf of Eilat/Aqabah), on a slope next to prehistoric pit mining. At the time of its discovery, in 1975, it consisted of several stone mortars and hammers, of an early type, and a small quantity of very primitive slag around them, obviously a primitive smelting workshop, the like of which we had not seen anywhere in the region. Slagged lumps of soil indicated that the early smelting installation was a hole in the ground, but we did not find any such ‘furnace’. At Site F2 we found some flint tools and a quantity of rough pot sherds. In 1976, after we had removed all of the surface finds, we excavated several squares on the slope below the actual site, including a square around a small stone setting, about 50 m from workshop F2. Stratigraphically, this stone setting was clearly intrusive, cutting into the layer containing the primitive slag, which was also found around the stone setting (cf tratigraphic section drawing in Merkel and Rothenberg 1999, 153). On excavation, the stone setting turned out to be a small fireplace, containing charcoal, ashes and small lumps of slag. Paul Craddock, at the time a volunteer to our excavations in Timna, supervised this excavation of Site F2 and took samples of the charcoal from inside the furnace for 14C dating at the British Museum: BM-1368: 1080 ± 50 BP. We now understand this stone setting as a melting installation where Egyptian New Kingdom miners re-melted the early, very copper-rich slag in order to extract the copper. A fragment of a crucible was recorded by Craddock. Unfortunately, at the time it was taken by Craddock to the British Museum and could not be located and investigated.

The problems of dating Site F2: Based on petrographic studies of the pottery and ceramic comparisons, a Late Pottery Neolithic date was proposed. Since this is a very early date for copper smelting in the region, and it would be important to get support for our dating by 14C, we went again to the site and searched for charcoal in the early layer. There was charcoal only near the later stone setting, but we hoped that it might give an early date. Unfortunately, OxA-5912 indicates that charcoal had spilled from the New Kingdom melting installation into the layers around it. It confirms the New Kingdom date of the slag melting operation. Site F2 was published in Rothenberg and Merkel (1995); Segal et al. (1998); Merkel and Rothenberg (1999).

Timna Valley

Sample of charcoal from Timna, Site 39 (29:45;50N 32:59:30E), Israel, submitted by J. Merkel and B. Rothenberg, Inst. Archaeol., Univ. Coll. London.

OxA-7632ash(?) charcoal, sample 155485 ± 45

Comment (B. R.): Site 39 (a and b), located in the estuary of Wadi Nehushtan of the Timna Valley, is one of the many prehistoric copper smelting sites on the foothills along the southern Arabah Rift Valley. Many of these sites are of the same type; workshop and habitation at the foot of a hill and the actual smelting site on its top. Excavations showed that both the smelter on top and the workshop below are of the same date. Site 39a, at the foot of the hill, consisted of three habitation structures and a circular arrangement of workplaces of metallurgical character, with crushing tools, copper ore and crushed smelting slag. Evidently smelting charge was prepared here and lumps of copper-rich slag from the smelting site 39b on the adjacent hill were brought down to be crushed here in order to extract the copper prills. Metallurgically, the slag found on top of the hill, at Site 39b, and at the workshop below. Site 39a, were identical. We excavated one of the habitations in 1965 and found pot sherds and flint tools of the Chalcolithic period. Right next to the entrance to this building was a small cooking stove full of ashes. OxA-7632 originates from this stove.

From the habitation and workshop area a well-trodden ancient path leads up the adjacent hill. Site 39b, and here we excavated a smelting furnace, a simple, bowl-shaped hole-in-the-ground (25–40 cm in diameter, 40 cm deep) with a low superstructure of small rocks. Around the furnace we found slag lumps, slagged rocks and some pot sherds and flint tools of the same type and date as the pottery and flints found in the workshop below; the Chalcolithic period.

Ashkelon/Sha'ar Hagolan/Tel ‘Ali

The 20 determinations came from three Protohistoric sites in Israel, excavated by Y. Garfinkel on behalf of the Hebrew University of Jerusalem: Ashkelon (34:35, 31:41E), Sha'ar Hagolan (35:35N 32:41E) and Tel ‘Ali (35:33N 32:42E). Only two of the determinations are problematic (Tel ‘Ali 3–4), while the other 18 are acceptable based on their stratigraphy, association with material culture, or similar dates obtained from other sites. They help in dating three crucial aspects in the chronology of the Levant: the transition from PPNB to the PPNC, the dating of the Yarmukian culture, and the transition from Middle to Late Chalcolithic. Detailed discussions on these dates can be found in two publications (Garfinkel 1999a, 1999b; Garfinkel and Miller 2002).

OxA-7915ashes, 1, δ13C = −25.7‰7995 ± 50
OxA-7916ashes, 2, δ13C = −17.8‰7935 ± 50
OxA-7881ashes, 4, δ13C = −21.2‰7630 ± 65
OxA-7882ashes, 5, δ13C = −22.3‰8000 ± 110
OxA-7883ashes, 6, δ13C = −24.9‰7990 ± 90
Sha'ar Hagolan  
OxA-7884charcoal, 1, δ13C = −23.5‰6980 ± 100
OxA-7885charcoal, 7, δ13C = −25.0‰7270 ± 80
OxA-7917charcoal, 2, δ13C = −25.8‰7410 ± 50
OxA-7918charcoal, 3, δ13C = −23.8‰7465 ± 50
OxA-7919charcoal, 4, δ13C = −25.1‰7495 ± 50
OxA-7920Charcoal, 6, δ13C = −24.7‰7245 ± 50
Tel ‘Ali  
OxA-7800charcoal, 5, (Layer Ib), δ13C = −24.7‰5950 ± 45
OxA-7801charcoal, 6, (Layer Ib), δ13C = −24.7‰5815 ± 45
OxA-7802charcoal, 7, (Layer Ib), δ13C = −24.1‰5770 ± 45
OxA-7804charcoal, 8, (Layer Ib), δ13C = −24.5‰5930 ± 45
OxA-7805charcoal, 9, (Layer Ia), δ13C = −22.7‰5680 ± 45
OxA-7886charcoal, 2, (Layer II), δ13C = −26.1‰7975 ± 70
OxA-7921charcoal, 1, (Layer II), δ13C = −24.5‰7940 ± 50
OxA-7922charcoal, 3, (Layer II), δ13C = −25.6‰135 ± 35
OxA-7923charcoal, 4, (Layer II), δ13C = −24.3‰6030 ± 45

Comment (Y. G.): the dates came from four different cultural units, as follows:

Pre-Pottery Neolithic C (PPNC): the nine samples related to this unit were obtained from the single period site of Ashkelon, and from Layer II at Tel ‘Ali (Samples 1–4). Most of these dates fall between 8000–7940 BP, very early eigth millennium BP. One date from Ashkelon is a little later, 7630 BP, still earlier than the dates of the next cultural unit of Sha'ar Hagolan. Two other determinations from Tel Ali (Samples 3 and 4) are clearly off by thousands of years, and should be rejected. The seven acceptable determinations are the first coherent radiometric dating of the PPNC in Israel. They help solve various chronological problems, and also clarify that the Neolithic site of Ashkelon does not date to the Yarmukian culture, as mistakenly dated by Perrot and Gopher (1996).

The Pottery Neolithic (Yarmukian Culture): all the samples from Sha'ar Hagolan fall within the chronological frame of the Yarmukian culture (7495–6980 BP). Similar dates have been obtained from ‘Ain Rahub, Munhata, Byblos, Nahal Qanah and ‘Ain Ghassul (Garfinkel 1999a, 1999b).

Middle Chalcolithic: This phase represents cultural assemblages that are later then the Wadi Rabh culture (Early Chalcolithic) but earlier than the classical Ghassulian culture (Late Chalcolithic). The four determinations all came from Layer Ib at Tel ‘Ali (Samples 5–8), and they all fall in the early part of the sixth millennium BP (5950–5770). These dates clearly indicate that this phase continues into the first quarter of the sixth millennium BP.

Late Chalcolithic (Ghassulian Culture): One date (Sample 9) is related to this unit and it came from Layer Ia at Tel ‘Ali (5680 BP). It is clearly later than the dates of the Middle Chalcolithic unit.

Israel and Jordan

Levantine Middle Bronze Age

In the framework of the SCIEM2000 programme (Bietak 2000), funded by the Austrian Academy of Sciences, and a special grant from the Jubily fund of the City of Vienna, a research project was carried out by, and samples submitted by, E. Marcus, Recanati Institute of Maritime Studies, University of Haifa, Israel, to apply AMS radiocarbon determinations to issues of the southern Levantine Middle Bronze Age IIa (MBIIa). In the 1990s, a pilot project was carried out on samples from Tel Nami (OxA-4532 to −4537) and Tel Ifshar (OxA-5355 to −5359) (Arch, List 23, 256–7). These samples produced seemingly accurate and precise results, but were not sufficiently precise to answer crucial chronological questions. As a result, and following close examination of the calibration curve, a more focused approach was undertaken to explore answerable questions regarding the absolute dates of the early MBIIa phases. As conventional synchronization with Egypt has not provided an incipient date earlier than c. 1920 BC (Bietak 1993–1994; Bietak and Dorner 1998; Czerny 1998; Bagh 1998, 2000), a particular emphasis was made on dating the beginning of this culture, which has been previously suggested to be as early as c. 2000 BC (Ward and Dever 1994). In the course of this research, samples were run at both ORAU, the Viennese Environmental Research Accelerator and, in the case of two sufficiently large samples, for conventional radiocarbon dating at the Weizmann Institute. In some instances, where higher precision was desired, double runs were carried out at ORAU and are noted below.

The samples for this project were derived from the two previously studied sites and five new sites, a portion of which are reported herein. All were graciously provided by the excavators with the assistance of the project palaeobotanists. Samples from Tel Nami derive from MBIIa settlement levels that include evidence of contact with the Aegean, Cyprus, Egypt and the northern Levant (Artzy 1993; 1995; Marcus 1991; Artzy and Marcus 1992; Kislev et al. 1993; Marcus and Artzy 1995; Lev-Yadun et al. 1996). Systematic archaeobotanical sampling by M. Chernoff (1988; Chernoff and Paley 1998) during the excavations at Tel Ifshar produced voluminous and regular assemblages of single-year cultigens from well stratified layers of an MBIIa settlement that has demonstrable contacts with Egypt and the northern Levant (Paley and Porath 1993; 1997). Samples from Tell el-Hayyat derive from two MBIIa phases of a series of four temples (Falconer and Magness-Gardiner 1984, 1989a, 1989b, 1993; Magness-Gardiner and Falconer 1994; Falconer 1995). The samples from Tell Abu en-Niaj belong to an Intermediate Bronze Age/Early Bronze IV/Middle Bronze I settlement near Tell el-Hayyat (Falconer 1987). The last suite of samples was intended to investigate the nature of the IBA/MBIIa transition. In general, the measurements were consistent with those from the other laboratories; that and other issues will be discussed in future publications (eg Marcus forthcoming). The following comments are based solely on the ORAU results calibrated using the INTCAL98 calibration curve.

Tel Nami (32:00N 35:00E)  
OxA-10014Tel Nami L 356 Vitis vinifera seeds, (double run).δ13C = −21 9‰3514 ± 32
OxA-10134Tel Nami L421 Vicia faba seeds, δ13C = −22 7‰3575 ± 40
OxA-10135Tel Nami L 440 Vicia faba seeds, δ13C = −23 6‰3562 ± 36

Comments (E. M.): OxA-10135 (68% prob.: 1960–1870 [53.9%], 1850–1820 [8.3%], 1800–1780 [5.9%]; 95% prob.: 2030–1990 [5.6%]. 1980–1770 cal BC[89.8%]) and OxA-10134: (68% prob.: 2010–2000 [3.8%], 1980–1870 [60.3%], 1840–1820 [4.1%]; 95% prob.: 2040–1770cal BC) represent harvests, respectively, from sometime within the earliest phase of the site and immediately preceding its end. OxA-10014 is from immediately before the destruction of the second and final phase at the site (68% prob.: 1890–1770; 95% prob.: 1920–1740 cal BC). When OxA-10135 is combined with the previously run identical sample OxA-4536, the resultant average is 3535 ± 31 BP and the calibrated ranges are 68% prob.: 1920–1870 (28.3%), 1850–1810 (22.7%), 1800–1770 (17.2%); 95% prob.:1950–1740 cal BC. Similarly, when OxA-10134 is combined with the previously run identical sample OxA-4534, the resultant average is 3553 ± 34 and the calibrated ranges are 68% prob.: 1950–1870 (28.3%). 1850–1810 (22.7%). 1800–1770 (17.2%); 95% prob.: 2020–1990 (2.4%), 1950–1740 (93.0% cal BC).

Thus, while the individual calibrated ranges show dates earlier than 1950 cal BC. when determinations of L.440 (OxA-135 and −4536) are combined, the uppermost date range does not exceed 1950 BC. Individually and combined, L.421 does show date ranges into the twenty-first century BC. However, as this sample is stratigraphically later than L.440, its upper range should not exceed its predecessor. OxA-10114, which represents a local harvest from within the final phase of the Levantine MBIIa. is consistent with determinations from coeval OxA-4532, which represents a harvest of Aegean cultigens that was subsequently imported to the site; hence, these samples should not be averaged. Nevertheless, the date-ranges are consistent and place the final MBIIa phase at Tel Nami prior to 1740/1730 BC

Tel Ifshar (32:25N 35:35E)  
OxA-101 15Vicia faba seeds, L.626 Phase B, (double run), δ13C = −22.7‰3487 ± 35
OxA-10124Vicia faba/wheadt seeds, L.752 Phase B, (double run), δ13C = −23.4‰3546 ± 36
OxA-10136Vicia faba seeds, L.927, Phase B, δ13C = −23.3‰3504 ± 38
OxA-10137Vicia faba seeds, L.834 Phase B/C, δ13C = −22.5‰3505 ± 35
OxA-10153Vicia faba seeds, L.I 103 Phase G, (double run), δ13C = −20.4‰3480 ± 23
OxA-10154Vicia faba/wheat seeds, L.730/734 Phase B/C, (double run), δ13C = −23.6‰3483 ± 31
OxA-10284Vicia faba seeds, L.719 Phase B, δ13C = −23.1‰3490 ± 65

Comments (E. M.): Phase B, which is the incipient substantial building phase at the site, was a particular focus of attention. OxA-10284 was a small sample that dissolved during pre-treatment and resulted in a larger than usual error. OxA-10115 is from a floor fill just underneath an imported Egyptian vessel dated to the first half of the nineteenth century BC, which is consistent with the calibrated ranges of the sample: 68% prob.: 1880–1840 (24.9%), 1830–1790 (20.6%), 1780–1740 (22.7%); 95% prob.: 1890–1730 (89.6%), 1720–1680 (5.8%) cal BC. The calibrated ranges of OxA-10136 have an upper boundary of 1930 BC: 68% prob.: 1890–1740; 95% prob.: 1930–1730 (93.0%), 1710–1690 (2.4%) cal BC, which is also the case when averaged with the results of identical sample OxA-5355. The upper calibrated boundary of OxA-10137 is 1920 BC; when averaged with identical OxA-5356, the resultant average is 3499 ± 30, which produces the following calibrated ranges: 68% prob.: 1880–1870; 95% prob.: 1890–1730 (94.4%), 1710–1690 (1.1%) cal BC. These are the same 2σ T ranges produced by coeval OxA-10154: 68% prob.: 1880–1840 (24.6%), 1830–1790 (19.2%), 1780–1740 (24.4%); 95% prob.: 1890–1730 (90.4%), 1720–1690 (5.0%) cal BC. The only sample that produced ranges higher than 1920 BC is OxA-10124, which may derive from an early stage of Phase B or the ephemeral Phase A: 68% prob.: 1940–1870 (39.2%), 1850–1810 (17.1%), 1800–1770 (11.9%); 95% prob.: 2010–2000 (1.3%), 1980–1740 (94.1%) cal BC. OxA-10153 is from the penultimate MBIIa phase at the site. By itself and averaged with identical OxA-5359, it provides a lower bound of 1730 BC for the end of the MBIIa, which is consistent with samples from Tel Nami: 68% prob.: 1880–1840 (28.6%), 1820–1790 (11.8%), 1780–1740 (27.8%); 95% prob.: 1880–1730 (92.9%), 1710–1690 (2.5%) cal BC.

Tell el-Hayyat (32:35N 35:35E)  
OxA-10986Triticum aestivum seeds, MBA39, δ13C = −22.4‰3470 ± 36
OxA-10987Triticum aestivum seeds, MBA40, δ13C = −22.9‰3497 ± 37
OxA-10988olive stone, MBA41, δ13C = −21.3‰3502 ± 37
OxA-10989olive stone, MBA42, δ13C = −21.3‰3523 ± 39

Comments (E. M.): the four samples all produce calibrated ranges that are relatively indistinguishable in radiocarbon terms. The incipient settlement in Phase 5 does not appear to have begun before 1920 BC, viz. OxA-10987: 68% prob.: 1880–1740; 95% prob.: 1920–1730 (92.8%), 1710–1690 (2.6%) cal BC. A more probable date is in the early nineteenth century BC, which is consistent with its 1σ range (c. 1880) and the 2σ range of OxA-10986: 68% prob.: 1880–1840 (22.1%), 1820–1790 (16.4%), 1780–1730 (29.7%); 95% prob.: 1890–1680 (95.4%) cal BC.

Tell Abu en-Niaj (32:35N 35:35E)  
OxA-10990Hordeum vulgare seeds, MBA43, d13C = −23.3‰3932 ± 38
OxA-10991Hordeum vulgare seeds, MBA44, d13C = −22.4‰3877 ± 40
OxA-10992cereal seeds, MBA45, d13C = −22.8‰3886 ± 40

Comments (E. M.): the calibrated ranges of these three samples are somewhat older than anticipated, with even the lower boundary of the 2s ranges of the Wnal phase dating at 2200 or earlier, eg OxA-10991: 95% prob.: 2470–2270 (85.3%), 2260–2200 (10.1%); and OxA-10992: 95% prob.: 2470–2270 (88.5%), 2260–2200 (6.9%) cal BC. Early Bronze Age IV/IBA/MBI is generally dated to the period of 2300/2200–2000 BC. As these samples belong to the Wnal two late EBA IV phases of the site, there may be some need to reassess the relative date of the site or the absolute dates for this period. Clearly, there is at least a 200–300 year gap in settlement between Tell Abu en-Niaj and Tell el-Hayyat.


Tell esh-Shuna

Samples of charcoal and seeds from Tell esh-Shuna (Shuna Project) (35:36N 32:37E), Jordan, submitted by G. Philip, Dept, Archaeol., Univ. Durham, and funded by the SERC.

OxA-4633tamarix wood charcoal, 1, δ13C = −26.0‰4500 ± 120
OxA-4634Ulmaceae charcoal, 2, δ13C = −27.0‰4440 ± 80
OxA-4635tamarix wood charcoal, 3, δ13C = −27.3‰4520 ± 75
OxA-4636ash charcoal, 4, δ13C = −23.5‰4555 ± 70
OxA-4637olive wood charcoal, 5, δ13C = −22.8‰4920 ± 75
OxA-4638olive wood charcoal, 6, δ13C = −22.3‰4870 ± 150
OxA-4639olive wood charcoal, 7, δ13C = −24.4‰4975 ± 75
OxA-4640Quercus charcoal, 8, δ13C = −24.9‰4820 ± 80
OxA-4641charred olive seeds, 11, δ13C = −23.1‰5020 ± 75
OxA-4642charred olive seeds, 11, δ13C = −22.3‰5080 ± 75
OxA-4643charred hulled barley seeds, 12, δ13C = −23.8‰6055 ± 80
OxA-4644charred hulled barley seeds, 12, δ13C = −23.7‰6000 ± 75
OxA-5387tamarix wood charcoal, 9, Area D: 443, δ13C = −25.4‰6020 ± 65
OxA-5388Hordeum sativum seeds, 10, Area D 585, δ13C = −22. 1‰6040 ± 55
OxA-5389tamarix wood charcoal, 13, Ac 725, δ13C = −26.3‰4750 ± 55
OxA-5390lentil seeds, 14, Aa:813, δ13C = −22.5‰4665 ± 55
OxA-5391olive stone, 15, Aa:626, δ13C = −21.3‰4585 ± 50
OxA-5392emmer seeds, 16, Aa:696, δ13C = −20.9‰4590 ± 50
OxA-5393hulled barley seeds, 17, Ab:744, δ13C = −19.8‰4685 ± 75
OxA-5394emmer seeds, 18, Ab:753, δ13C = −23.2‰4735 ± 55
OxA-5395olive stone, 19, Ab:869, δ13C = −22.9‰6065 ± 55
OxA-5396emmer seeds, 20, Ab:841, δ13C = −22.9‰6070 ± 55

Comment (G. P.): the beginning of the Early Bronze Age (EBA) II period in Palestine and Jordan around 3100–3000 cal BC is generally taken as witnessing the appearance of complex societies in the region. However, these developments have their roots in the preceding, and poorly understood, EBA I. Recently excavated EBA I material has come from short-lived rural settlements offering little potential for the study of change within the period. Moreover, the few EBA I sequences identified elsewhere in the Levant show regionally differentiated material cultures, and lack either early EBA I deposits or a good set of radiocarbon dates. Tell esh-Shuna is unusual in offering a long EBA I occupational sequence. This comprises six stratigraphic phases, grouped into two distinct ceramic periods, which can be broadly equated with Stager's (1992: 39–40) EBA IA and EBA IB. The radiocarbon determinations from Tell esh-Shuna provide the first absolute chronology for an EBA I sequence, and will clarify both the duration of the period, and the possibility that there existed an overlap between late Chalcolithic in the south and the earliest EBA I in the north.

The Chalcolithic in the southern Levant witnesses early forms of social complexity that some would argue involve craft specialisation, significant levels of exchange of prestige goods, and stratified social and political organisation. However, these features have been identified predominantly in the later stages of the Chalcolithic and in the southern part of the region. The extent to which the latest Neolithic and Early Chalcolithic in these areas, and the northern Chalcolithic as a whole witnessed such developments is uncertain because of the problems of relative dating arising from highly regionalized ceramic assemblages.

The Chalcolithic assemblage from the lower deposits of the site is hard to fit into a relative chronological framework on ceramic grounds, as it represents a new regional variant of Late Neolithic to Chalcolithic in the southern Levant. The radiocarbon dates place this new material within an absolute chronological framework, circumventing tortuous ceramic arguments, and so address issues of early complexity. Finally, Gustavson-Gaube (1986) on the basis of her 1984–85 excavations suggested that there was continuity of occupation at Tell esh-Shuna from Chalcolithic through to EBA I. Radiocarbon determinations from the earliest part of the sequence will clarify the existence or otherwise of this continuity.

The excavated sequence: Excavation areas D and A were adjacent, but started at different absolute heights, being the bottom and top of a bulldozed terrace respectively. Phases are numbered from 1 (lowest) upwards. ID is the lowest phase, and immediately overlay natural gravels. It proved impossible to establish a direct stratigraphic connection between the D and A sequences because of intervening later disturbances, but the sequences can be very closely correlated. Six stratigraphic phases of EBA I occupation were preserved in Area A. These ran from 4A (earliest) to 9A (latest). Phases 4A–6A belong in ceramic terms to early EBA I, phases 8A and 9A to late EBA I.

Late EBA I: OxA-4633 to −46.36 and OxA-5391 to −5392 come from Phase 8A; all are short-life samples. OxA-4633 to −4636 were all taken from context 149, a midden deposit stratigraphically late within the phase. These form a fairly consistent group suggesting a date for the deposition of the midden within the period 3400–2900 cal BC. OxA-5391 and −5392 come from contexts stratigraphically earlier within the same phase, a point apparently confirmed by the calibrated dates which fall between 3500 and 3100 cal BC. Note that there was at least one subsequent stratigraphic phase (9A) which produced material which would be ‘late EBA I’ in ceramic terms.

Early EBA I. OxA-5393 and −5394 come from Phase 4A, stratigraphically the earliest part of the EBA I sequence. The calibrated dates fall between 3650 and 3350 cal BC. OxA-5389 and −390, which belong to the stratigraphically later Phase 6A, fall within the same limits when calibrated. All were taken on short-life samples collected in 1994. While early in comparison to the traditional dates for the beginning of EBA I (c. 3400–3300 BC), the dates from Shuna are only a little earlier than more recent estimates such as that of 3500 BC by Stager (1992: 40).

Calibrated results for OxA-4637to −4640 which come from Phases 5A and 6A fall between 3900 and 3350 cal BC and appear earlier than the dates above. However, three of these (OxA 4637 to −4639) were taken on charcoal from olive wood, a long-lived species which produces hard limber, potentially subject to re-use. The fact that the calibrated dates for OxA-4640, identified as sapwood of evergreen oak (3770–3375 cal BC), appear rather younger than the other three appears to confirm this suggestion. We believe that the short-life samples collected in 1994 offer the most reliable estimate for the duration of the early EBA I occupation at the site.

The calibrated dates from the Chalcolithic occupation, OxA-4643 and −4644, OxA-5387 and −5388, and OxA-5395 and −5396, all fall within the period 5100 and 4750 cal BC providing a reliable set of dates for the earliest excavated deposits at the site, and confirming the contemporaneity of the Chalcolithic occupational sequences in Areas A and D. They also confirm the existence of a substantial gap in occupation within the excavated area between the main part of the Chalcolithic sequence and EBA I deposits. That said, OxA-4641 and −4642, taken on short-life material, produced a calibrated date (4000–3650 cal BC) which is consistent with activity dating to a late stage in the Chalcolithic period. This sample was taken from a pit which clearly post-dates all the early Chalcolithic material in Area D, but the top of which had been truncated by twentieth century activity. Although no structures were associated with this pit, it provides evidence of activity on the site during the late fourth millennium cal BC, despite the lack of well stratified Late Chalcolithic deposits in the excavated area.


Islamic embroideries

Samples of textiles, submitted by R. Barnes, Eastern Art, Ashmolean Museum, Oxford through the NERC-funded ORADS facility. Comments by Marianne Ellis, Hillcroft, Pennys Lane, Farley, Salisbury, SP5 1AR.

OxA-6519flax/silk textile. 1993.210, δ13C = −23.9‰230 ± 75
OxA-6520flax/silk textile, 1993.112, δ13C = −24.1‰745 ± 75
OxA-6521flax/silk textile, 1984.267, δ13C = −24.0‰560 ± 75
OxA-6522flax/silk textile. 1993.236, δ13C = −24.3‰345 ± 75
OxA-6523flax/silk textile, 1984.491, δ13C = −24.0‰690 ± 75
OxA-6524flax/silk textile, 1988.48, δ13C = −24.8‰1240 ± 80
OxA-6525flax/silk textile, 1993.333, δ13C = −23.0‰490 ± 75
OxA-7009flax/silk textile, 1993.116, δ13C = −25.8‰695 ± 70

Comment (M. E.): a surprising result was that a tiraz fragment, OxA-6524, resulted in a radiocarbon determination of 1240 ± 80 BP but is dated by its inscription to the third decade of the tenth century. However it falls within the 95.4% confidence range ending with 960 AD. It was interesting to see that an embroidery with a flowing design worked in freestyle and counted filling stitches, OxA-6520, was contemporary with the better known band patterns stitched in darned counted needlework. A result of 690 ± 75 BP for a sampler, OxA-6523, demonstrates that by the thirteenth century a different pattern darning technique had been introduced. Instead of resembling woven decoration as seen on a sampler dated a century earlier, the darning has been modified to produce designs in reserve with added outlines.

The result of 560 ± 75 BP for OxA-6521 was helpful because the embroidery concerned is from a puzzling category of fragments unlike others in the collection. One of them had been tentatively attributed to the fourteenth century on stylistic grounds. The results contain two warnings that the dates may extend out of range confirming the impression that these pieces are from the Ottoman period. Part of a tunic front, OxA-6525, is dated 490 ± 75 BP; this is significant for two reasons. The collar is embroidered in pattern darning demonstrating the continuous use of this type of embroidery from pre-Mamluk times. In addition. it is helpful to know a particular openwork embroidery technique was practised in Egypt as early as the fifteenth century.

Mons Porphyrites

Samples of carbonized wheat chaff from the Roman quarry settlement at Mons Porphyrites, located in the Eastern Desert of Egypt (27:15N 33:20E). Excavated by D. Peacock. Univ. Southampton, and V. Maxfield, Univ. Exeter, between 1996 and 1998. Submitted 2000 by M. van der Veen, School of Archaeology & Ancient History, Univ. Leicester.

OxA-9974Triticum durum rachis charred seeds, MP98 FES 521, δ13C = −25.5‰1770 ± 50
OxA-9991Triticum durum rachis charred seeds, MP98 FES 521, δ13C = −21.2‰1771 ± 35

Comment (M. v. d. V.): the samples originate from an ash deposit found close to the kitchen area at the fortified settlement of the main complex. The chaff within the deposit represents the fuel used in the bread ovens of the kitchen. The deposit contained an archive of ostraca, most of which represent so-called ‘bread-receipts’. These carry no date, but the style of handwriting is thought to be suggestive of the second century (Van Rengen 1997 and pers. comm.). The ceramic assemblage consisted of predominantly late second century AD sherds, but included some late fourth/fifth century AD vessels (Tomber 1997). The samples were submitted for dating to identify at what time during the operations at this complex, organised bread-making was in operation. The 95% prob. calibration range for both samples is 130–390 cal AD and the starting date for the late fourth century vessels is usually put at c. AD 390 (Tomber. pers. comm.). Thus. the radiocarbon determinations indicate that the chaff, and consequently the bread making, date to the late second rather than the late fourth/fifth century AD.



Samples of charred seeds from Saar (26:11N 50:29E), Bahrain, submitted by J. Moon, Upper House, Stoke St. Milborough, Ludlow, Shropshire.

OxA-8275Phoenix dactylifera charred seeds, 4130.06, δ13C = −25.3‰3665 ± 30
OxA-8276Phoenix dactylifera charred seeds, 3041.27, δ13C = −21.7‰3670 ± 50
OxA-8277Phoenix dactylifera charred seeds, 5111.05, δ13C = −23.3‰3595 ± 45
OxA-8278Phoenix dactylifera charred seeds, 5510.103, δ13C = −23.0‰3355 ± 35

Comment (J. M.): all four determinations come from Bronze Age houses in the settlement at Saar, Bahrain, dated by traditional methods (mainly pottery chronology) to c. 2000–1800 BC. The particular samples were chosen to provide confirmation of the date of the main phase of occupation at the settlement.

OxA-8275, from a floor in Building 208, belongs to the earlier part of the sequence. The date is expected to be at the more recent end of the 95.4% confidence range. OxA-8276, collapse from Building 301. is towards the end of the sequence, and a date was expected of between 19(H)-1800 BC. The 68.2% confidence ranges give a slightly older date, but one still within the time range of the settlement as a whole. The 68.2% confidence range for OxA-8277 from the floor of Building 220 (mid-sequence) is within the established chronological range of the settlement. OxA-8278 (from the floor in Building 224, which is next to Building 220) should be the same or slightly earlier than OxA-8277. The given determination, therefore, is loo young by c. 200 years.

For further information about excavations at Saar, see Crawford et al. (1997).



The five samples of charcoal all come from the site of Thiéhel, situated in the village of Guédé in the western part of the Middle Valley of the river Senegal, in the north of the country (16:30:16N 14:44:9W). The site is composed of several mounds which were excavated by Ndehye Sokhna Guèye in 1996. Samples were submitted by Pierre de Maret, Université Libre de Bruxelles, and Ndèye Sokhna Guèye, CODESRIA, Dakar, Senegal, in 1998.

OxA-7972charcoal, C14 (1) L2/F1/Ph5, δ13C = −24.4‰370 ± 35
OxA-7973charcoal, C14 (2) L4/Ph5, δ13C = −26.0‰1165 ± 35
OxA-7974charcoal, C14 (3) L9/Ph4, δ13C = −26.1‰895 ± 35
OxA-7975charcoal, C14 (4) L11–12/Ph4, δ13C = −25.4‰2475 ± 45
OxA-7976charcoal, C14 (5) L12/Ph4, δ13C = −25.5‰2460 ± 40

Comments (N.S.G.): the excavations were aimed at reassessing the history of occupation of the northern part of Senegal. Previous excavations by B. Chavane had dated the occupation at Thiéhel between 1330 and 1950 AD (Chavane 1985). The new determinations (OxA-7972 to −7974) certainly confirm the recent phase of occupation at Thiéhel. Four out of five determinations are in good stratigraphic order, but OxA-7974 should have been older than OxA-7973. The inversion may be explained by bioturbation of which traces were observed at the surface. Within the series of determinations, there are two very distinct groups. The first is composed of three dates indicating the beginning of the recent occupation phase around the eight century AD which continued into the sixteenth century. The second group of two almost identical results (OxA-7975 and −7976) reveals an occupation between the fifth and eighth century BC at the site. These dates are unexpected and are the oldest dates obtained so far for the occupation of the Middle Valley of Senegal. They invite reconsideration of the issue of the population of the northern part of Senegal.


Lopé Faunal Reserve

All charcoal samples come from sites excavated in the Faunal Reserve of Lopé, situated in central Gabon (Assoko Ndong et al. 1999; Angoué et al. 2000; Assoko Ndong 2002 and in press): Okanda 6 (Au-Porte de l'Okanda), situated at the right bank of de Ogooué river (0:06:35N 11:34:45E); Lindili-1 (0:09:29S 11:30:09E); Lopé 2 (0:13:05S 11:35:15E); Okanda 5 (0:07:16S 11:33:11E). These sites were excavated by A. Assoko Ndong as part of his doctoral thesis. The samples were submitted for dating by P. de Maret, Préhistoire, Musée Royal de l'Afrique Centrale, Université Libre de Bruxelles, 3080 Tervuren, Belgium and comments by A. Assoko Ndong, Musée Royal de l'Afrique Centrale, Université Libre de Bruxelles, Tervuren, Belgium, in 1995 and 1996.


OxA-7604charcoal, AU/96 F2 (–100cm), δ13C = −24.4‰2160 ± 35

Comment (A. A. N.): the sample was taken at 100 cm below surface in one of the 4 pits on the site. The charcoal was found associated with pottery of Epona tradition, palm nuts, pitted stones, etc. The newly obtained date falls well in the range of the Epona pottery which is one of the two pottery traditions recognised in the area dating from the transition from Stone to Iron Age, which has been dated between 2450 and 1800 BP.


OxA-7602charcoal, LIN/95 F.11/8(–60cm), δ13C = −25.9‰1905 ± 35
OxA-7753charcoal, LIN/95 F. 17/14(–40cm), δ13C = −28.6‰1830 ± 35

Comments (A. A. N): the samples of charcoal were collected respectively at a depth of 40 and 60 cm below surface and come from two different pit structures (11/8 and 17/14). Analysis of pottery manufacturing, shape and decoration show that the pottery found in the Lindili-1 pits belong to two different traditions: Okandan and Lindilian. These two traditions belong to the Early Iron Age (c. 2000–1300 BP) and were found in pits with charcoal, Elaeïs kernels, tuyere fragments, ironslag, pitted stones, etc. The newly obtained dates firmly place these traditions within the time span of the Early Iron Age in the Faunal Reserve of the Lopé.

Lopé 2  
OxA-7601charcoal, LOP.2/96 F1 (–50cm), δ13C = −26.6‰2370 ± 35
OxA-7704charcoal, LOP.2/96 F1 (–50cm), δ13C = −27.1‰2535 ± 40
OxA-7705charcoal, LOP.2/96 F1 (–50cm), δ13C = −27.4‰2560 ± 45

Comments (A. A. N.): The samples of charcoal were collected at a depth of 50 cm in the only pit containing pottery at the site. The pottery belongs to the Yindo tradition, which has a similar age as that on other sites in the Faunal Reserve of the Lope’, and the age of OxA-7601 falls well within the range of the transition from Stone to Metal Age in the area which is dated to between 2450 and 1800 BP. Within this period there is an older Epona tradition and a more recent Yindo tradition.

However, the two remaining dates (OxA-7704 and −7705) associated with Yindo pottery which we did not receive at the time of the preparation of the disssertation, would question this sequence. Yindo would be older than Epona, and the transition from Stone to Metal would be older by at least two centuries and may well be contemporaneous with the same event at the coast in Gabon.

Okanda 5

OxA-7603charcoal, OKA.5/96 SONDAGE A1(–25cm), δ13C = −26.7‰1980 ± 35

Comments (A. A. N.): The charcoal sample was collected at a depth of 25cm in square A1 of the excavation, which explored an occupation level buried at 10 cm below the present surface at the site. Okanda 5 is a large site with pit structures situated on a hilltop in the savanna. Apart from the pits, the Early Iron Age site also revealed an occupation surface which yielded pottery from a different tradition than that found in the pits. Dating the occupation level has enabled us to understand that its pottery and that found in the pits are absolutely contemporaneous.



The sample comes from the site of Akouen which is situated 50 km to the south-east of Fumban, Dept. Mbam of the Republic of Cameroon (5:23N 11:19E). During ethno-archaeological research conducted in the area by Olivier Gosselain, a brief archaeological survey was carried out by P. Lavachery in 1991. Along a newly built road in the slope he noticed an occupation layer composed of sherds and charcoal. The pottery is very similar to the Tikar pottery still produced in the village of Akouen. The sample was submitted to date this pottery by P. de Maret, Université Libre de Bruxelles, in 1995.

OxA-6058charcoal, AKU 2 (–40), δ13C = −26.0‰140 ± 40

Comments (P. de M. and O.G.): the result is more recent than expected; however, it still carries the Tikar occupation of the area beyond the limits of oral tradition.


Mulundu is situated in the far east of the Republic of Cameroon (2:00N 15:10E). Sherds and associated charcoal were found in 1987 in a pit structure visible in a freshly cut slope. The pottery was the first to be excavated in this area, and is unlike any pottery known from neighbouring regions. The sample was submitted to date this pottery in 1995 by P. de Maret and O. Gosselain, Université Libre de Bruxelles.

OxA-6059charcoal. MUL.87/-110, δ13C = −26.2‰340 ± 40

Comments (P. de M. and O. G.): the date is recent, and remains difficult to comment on on since no other dating evidence is available from the area.

Shum Laka

Shum Laka rock shelter is situated in the Grassfields of the Northwestern Province of Cameroon (5:51:31N 10:04:40E), and has been excavated in 1980, 1984, 1991–2 and 1993–4 (Asombang 1988, 1994; Asombang and de Maret 1992; Cornelissen 1996; de Maret 1996; de Maret et al. 1987, 1993, 1995; Lavachery 1996, 1998, in press; Lavachery et al. 1996; Lavachery and Cornelissen 2000; Moeyersons 1996a, b, 1997; Moeyersons et al. 1996; Orban et al. 1996). These samples were submitted for dating by P. de Maret, Université Libre de Bruxelles, and E. Cornelissen, Musée Royal d'Afrique Centrale, Tervuren, Brussels. In the comments, dates have been arranged in chronological order.

OxA-4945charcoal, LAK93-C12/475–470/C14no359, δ13C = −24.4‰31 700 ± 750
OxA-4944charcoal, LAK93-B12/460–455/C14no364, δ13C = −25.2‰30 300 ± 1600
OxA-7875charcoal, LAK93–1 14/740–735/C14252, δ13C = −26.3‰20 260 ± 180
OxA-7912charcoal, LAK93–1 14/760–755/C14213, δ13C = −25.9‰18 040 ± 100
OxA-5200Charcoal, LAK93-B12/630–625/C14No288, δ13C = −26.2‰12 800 ± 110
OxA-5635charcoal, LAK93-C12/680–685/C14No196, δ13C = −24.5‰9880 ± 100
OxA-5636charcoal, LAK93-C14/755–750/C14No245, δ13C = −24.5‰9095 ± 70
OxA-5202charcoal, LAK93-G11/149/C14No3 14, δ13C = −26.7‰8540 ± 90
OxA-5203human bone, LAK93-SII/C13/187, 188, δ13C = −19.7‰7150 ± 70
OxA-5204human bone, LAK93-SIII/E18 & E17, δ13C = −19.8‰3300 ± 90
OxA-5205human bone, LAK91-H/E16/318’ 385 371, δ13C = −20.3‰3045 ± 60
OxA-5206human bone, LAK91-F/G14/123 128, δ13C = −20.5‰3025 ± 60
OxA-5207human bone, LAK91-A/G14/198 203, δ13C = −20.4‰2940 ± 60
OxA-5201animal bone, Svncerus caffer nanus, LAK91-H 10–38, δ13C = −9.2‰1310 ± 65

Comments (P. de M. and E. C): OxA-4944 and −4945 were selected at a depth of 3.30 m to date the base of the S-deposits, or the lower part of the sequence at the entrance of the rock shelter. OxA-4945 was collected in the square next to OxA-4944, 5 cm higher in stratigraphy. The difference in age is explained by the sloping of the sediments in this area of the site. The results of 30 000 BP were beyond expectations because the oldest 14C dates obtained during previous research in the area (Asombang 1988) were around 18 000 BP. The associated Late Stone Age microlithic industry on quartz, thought to have been confined to the Holocene in Western Central Africa, turns out to have Later Pleistocene origins as has been demonstrated for similar quartz industries in eastern Central Africa.

OxA-7912 and −7875 were selected in order to refine the chronology of the lower deposits in the rock shelter. OxA-7875 was intended to date the base of the S-deposits and the top of the supposedly older P-deposits which were only excavated in the central part of the rock shelter. The P-deposits could, however, also represent a local fades of the S-deposits. The date of 20 260 BP favours the latter explanation and also reveals continued occupation of the shelter at the onset of the Last Glacial Maximum. This is reinforced by the date of OxA-7912 of 18 040 BP. This charcoal sample was selected again at the entrance of the shelter in order to refine the internal chronology of the S-deposits and associated quartz industries. The result also confirms that there are no stratigraphic gaps in the accumulation of the S-deposits and that the shelter was used during the Last Glacial Maximum.

OxA-5200 was selected to date the top of the S-deposits which attain a thickness of approximately 1.55 m at the entrance. The date of 12 800 ± 110 BP dates the top of the S-deposits to the early Holocene. Apparently the S-deposits span the entire part of the Later Pleistocene including the Last Glacial Maximum. OxA-5635 comes from 50 cm higher in the stratigraphy and was selected to date the overlaying T-deposits and to give a terminus ante quern for the overlaying stone-line on top of which human burials were found, which were dated to ± 7000 BP. The date of 9880 ± 100 BP falls within the expected range.

As a general comment on the Later Pleistocene part of the sequence, it must be noted that all charcoal samples identified so far are mono-specific: Protea madiensis, or a savanna shrub. This may indicate that the environment around the rock shelter never experienced any drastic change. All dates obtained from the later Pleistocene and early Holocene deposits are in correct stratigraphic order.

OxA-5636 and −5202 were selected to date the base of the ochre ash layer. The charcoals were associated with a rich Late Stone Age industry on quartz and abundant bones of forest game such as buffaloes and hogs at the bottom of the layer. OxA-5636 is a Celastraceae (Maytinus acuminatus) which is a mountain forest tree and OxA-5202 (Hypericum sp.) is a grass indicating a rather open landscape. The early Holocene hunters who visited the cave were exploiting a mixed environment.

OxA-5203 (a fragment of a rib) dates the burial of two children. The tomb is contemporaneous with the Upper Horizon of the ochre ash layer, which consisted of a Ceramic Late Stone Age industry on quartz similar to the preceding occupation, associated with a few potsherds decorated with comb impressions and rare large bifacial tools on basalt. As during the preceding occupation, the visitors to the shelter hunted only forest game (mostly buffalo and hog). Forest trees and open environment plants were identified, suggesting that, as before, a mixed landscape was exploited. A few specimens of the edible fruits of Canarium schweinfurthii, a tree that naturally occurs in savanna or open forest, were also collected.

OxA-5204 to −5207 date three probably contemporaneous burials found in the upper part of the Holocene fill of the shelter, the grey ash layer. OxA-5204 was taken from a double burial where the two adults were found in anatomic connection, lying back to back in the fætal position. OxA-5205 dates the skeleton of a child found. OxA-5206 and −5207 come from a pit were a least eight children were buried together, the bones in disarray. These tombs can be linked with the end of the occupation dated by OxA-4538 (Arch. List 21), around 4000 BP. The industry consists mostly of large basalt tools such as scrapers, bifacial waisted axes and retouched blades but quartz and chert microliths are still produced. A much more abundant pottery decorated with sticks and combs is present. Forest game is common but fruits of trees from open environment such as Canarium schweinfurthii are also eaten.

OxA-5201, on an animal bone indicative of a forested environment, was selected for dating in order to refine the chronology within the Grey Ash Layer. The date sets the upper limit of the middle horizon within the Grey Ash, or Iron Age phase I.


Talaky and Bevala

Samples from Talaky (45:43:15S 25:27:21E) and Bevala (45:13:39S 25:35:49E). Southern Madagascar. submitted by Dr. M. Parker-Pearson, Dept. Archaeol. & Prehistory, Univ. Sheffield through the NERC-funded ORADS facility. Comments by M. Parker-Pearson, K. Godden, 9 High St, Souldrop, Beds, Ramilisonina, Lot 35 Mahajoarivo, Antananarivo, Madagascar, Retsihisatse, Analamahery, Andalatanosy, Madagascar, J.-L. Schwenninger, 25a Apsley Road, Oxford and H. Smith. Dept. Archaeol., Univ. Bournemouth.

OxA-8118charcoal, AND 16–97, δ13C = −25.1‰1080 ± 35
OxA-81 19charcoal, AND 22–97, δ13C = −23.4‰1025 ± 35
OxA-8121charcoal, AND 20–97, δ13C = −25.8‰900 ± 30
OxA-8135charcoal, AND 7–97, δ13C = −25.0‰560 ± 35
OxA-8136charcoal, AND 12–97, δ13C = −23.2‰530 ± 35
OxA-8147charcoal, AND 17–97, δ13C = −24.8‰915 ± 40
OxA-8148charcoal, AND 8–97, δ13C = −24.1‰1035 ± 40
OxA-8149charcoal, AND 10–97, δ13C = −25.4‰1010 ± 40
OxA-8269Aepyornis maximus, carbonate, AND 5–97, δ13C = −15.7‰1735 ± 35
OxA-8270Aepyorms maximus, carbonate, AND 6–97, δ13C = −15.1‰1240 ± 35
OxA-8271Mulleroms sp., carbonate, AND 33–97, δ13C = −13.8‰1825 ± 30
OxA-8272Aepyornis maximus, carbonate, AND 4–97, δ13C = −14.6‰1270 ± 40
OxA-8273Mulleronis sp., carbonate, AND 2–97, δ13C = −14.9‰1780 ± 35
OxA-8274Aepvornis maximus, carbonate, AND 1–97, δ13C = −14.4‰1835 ± 30
OxA-8279Aepyornis maximus, carbonate, MAD-97/16, δ13C = −13.0‰5480 ± 35
OxA-8280Aepyornis maximus, carbonate, MAD-97/19, δ13C = −10.7‰4030 ± 40
OxA-8281Aepyornis maximus, carbonate, MAD-97/45, δ13C = −1 3.3‰2285 ± 35

Comment: (M. P-P.): Talaky is a group of large coastal shell middens which were excavated in 1997. Since 1963 this has been recognised as one of Madagascar's earliest sites, based on a single 14C determination. The 1997 excavations uncovered two main styles of ceramics, Andranosoa (eleventh/thirteenth century AD) and Rezoky (fourteenth/fifteenth century AD) phases, and a major aim was to establish a reliable basis for their chronology. The dates for charcoal in association with the two styles confirmed this chronological framework. The other major aim was to establish whether the eggshells of Aepyornis and Mullerornis, giant extinct flightless ratites known as Elephant Birds, which were found in the midden were collected from nest sites contemporary with the human occupation. The dates for the eggshells are all older than the dates of charcoal from the same layers within the midden, suggesting that the eggshells were already ancient when their fragments became incorporated into the midden. We are currently investigating the possibility that the greater age of the eggshell fragments may partially result from the marine reservoir effect.


OxA-8120charcoal, AND 30–97, δ13C = −14.9‰535 ± 30
OxA-8268Aepyornis maximus. carbonate, AND 29–97, δ13C = −13.0‰1360 ± 35

Comment: (M. P-P.): Bevala is an aceramic midden which we had initially thought to be ‘Pre-ceramic’. However, the determination for the charcoal places it in the Rezoky phase.


United States of America

Florence St. Site

Samples from Florence St. Site (11-S-458, 38:33N 90.8W), submitted by T. Emerson, ITARP, Dept. Anthropology, Univ. Illinois, Urbana, Illinois, USA.

OxA-6678collagen, IC 24,Co2, δ13C = −21.0‰680 ± 55
OxA-6679collagen, IC 25,Co2,δ13C = −21.0‰60 ± 55
OxA-6680collagen, IC 27,Co2, δ13C = −21.0‰275 ± 70

Comment (T. E.): all three samples should date Cahokian Middle Mississippian burial from the Moorehead/Sand Prairie Phase transition period c. cal AD 1250–1300. OxA-6678 is in alignment with seventeen dates from similar contexts. OxA-6679 is six to seven centuries too late for a burial context verified by seventeen dates from similar contexts. OxA-6680 is four centuries too late for the same reason.

Range Site

Samples from Range Site (11-S-47, 38:30N 90:12W), Illinois, submitted by T. Emerson, ITARP, Dept. Anthropol., Univ. Illinois, Urbana, Illinois, USA.

OxA-6677collagen. IC 22, C02, δ13C = −21.0‰840 ± 70

Comment (T. E.): OxA-6677 should date Cahokian Middle Mississippian burial from the Moorehead/Sand Prairie Phase transition period c. cal AD 1250–1300. The date is a century early for burial context but within the generally acceptable range.

Mund Site

Samples from Mund Site (11-S-435, c. 38.30N, 90:12W), submitted by T. Emerson, ITARP, Dept. Anthropol., Univ. Illinois, Urbana, Illinois, USA.

OxA-6291charcoal, ICh 5034, Co2, δ13C = −26.0‰600 ± 60

Comment (T. E.): OxA-629 should date maize from Early Late Woodland Mund site c. AD 400–800. The date is nine to six centuries too late suggesting the specimen is contamination from a later Mississippian presence in the area.



Sample of bone from the site Xkipché, N. Yucatan, (20:18N 89:49W), submitted by H. J. Prem, Seminar für Völkerkunde, Univ. Bonn, Germany. Comments by M. Vallo, Institut für Altamerikanistik und Ethnologic Univ. Bonn, Germany.

OxA-7655human bone, 1515, building B18a, grave IV, δ13C = −9.3‰1105 ± 40

Comment (M. V.): this site was dated previously to the late and terminal Maya Classic periods from about AD 700–1000. Excavations at the site were undertaken by Institut für Altamerikanistik und Ethnologie (IAE), University of Bonn, Germany and the Instituto Nacional de Antropología e Historiy, Centro Regional Yucatan (INAH-CRY) from 1974–1991. OxA-7655 came from bones found near the ceremonial area of the town. This result, together with other dates from nearby burials, confirmed an early dating of the building at the beginnings of the eighth century AD. For further discussion see Andrews (1995), Lucke (1994), Prem (1991, 1995), Reindel (1997), Schlegel (1997), Vallo (2000).


Burton Ramie

Samples of bone from Barton Ramie, Belize River (17:18N 89:00W), submitted by N. Hammond. Dept. Archaeol., Boston Univ., USA.

OxA-3298human bone, Md123/Bur16/N-8857–5, δ13C = −11.4‰1230 ± 65
OxA-3299human bone, Md123/Br19/N-8857–40, δ13C = −10.9‰1415 ± 70
OxA-3426human bone, Md123/Br31/N-8857–56, δ13C = −11.1‰1685 ± 65
OxA-3427human bone, Md154/Bur1/N-8857–97, δ13C = −7.8‰1950 ± 65
OxA-3665human bone, Md123/20/N8857–41, δ13C = −14.2‰2290 ± 65
OxA-3666human bone, Md123/30/N8857–45, δ13C = −9.8‰2040 ± 70

Comment (N. H.): Barton Ramie is one of several sites, also including Holmul, Guatemala, and Nohmul and Tzimin Kax, Belize, that have yielded burials accompanied by pottery vessels in Holmul I or Protoclassic style, marked by early polychrome decoration and innovative vessel forms. The phenomenon has been ascribed to both invasion/immigration from the Maya highland zone and to autochthonous development; while placed somewhere in the early centuries AD, the Protoclassic has never been closely dated.

The Barton Ramie dates were run on human bone from burials from Mound 123 stocked with Protoclassic vessels (Holmul individuals lacked surviving collagen and associated bone was not available from other sites), together with one earlier (Mound 154 Burial 1, OxA-3427) and one later (Mound 123 Burial 16, OxA-3298) interment with vessels of well-dated style as controls. Both yielded dates in accordance with their associated ceramics, while the four Protoclassic burials gave dates ranging from the third century BC to the sixth century AD. This long span indicates that the Protoclassic is not a short-lived ‘horizon’ phenomenon such as would be associated with an invasive episode. It suggests that the style developed during the Late Preclassic, and continued in parallel with the related Tzakol polychrome style in the Peten region just to the west. It also confirms Adams's (1971) assignment of an extended period (AD 150–450) to the Protoclassic Salinas phase at Altar de Sacrificios in the southern Peten. A local development of Protoclassic style, embedded within the established lowland Maya ceramic tradition, is confirmed. See also Brady et al. (1998).


Samples of bone from Cuello (18.05N 88:35W), northern Belize, submitted by N. Hammond, Dept. Archaeol., Boston Univ., USA, funded by the NERC-funded ORADS facility and the SERC.

OxA-3084human bone, 148 (4280), δ13C = −15.9‰3070 ± 80
OxA-3085human bone, 154 (5093), δ13C = −10.8‰2020 ± 80
OxA-3664human bone, 4521, δ13C = −13.8‰2510 ± 65
OxA-4452manioc root charcoal, 4458, δ13C = −22.5‰2540 ± 70
OxA-4453charcoal, 4662, δ13C = −24.7‰2485 ± 70
OxA-4454charcoal, 4663, δ13C = −26.8‰2800 ± 70
OxA-4455charcoal, 4682, δ13C = −23.7‰2745 ± 75
OxA-4456fruit/nut kernel charcoal, 4683, δ13C = −24.3‰2535 ± 70
OxA-4457charcoal, 4685, δ13C = −24.4‰2625 ± 75
OxA-4458human bone, B.176, C4465, δ13C = −13.5‰2600 ± 75
OxA-4459human bone, B.177, δ13C = −12.7‰2545 ± 70
OxA-4460human bone, B.178, δ13C = −14.4‰2620 ± 75
OxA-4461human bone, B.180, δ13C = −13.5‰3040 ± 80
OxA-4542charcoal, 4569, δ13C = −24.8‰2650 ± 60
OxA-5020human bone, 159/167, δ13C = −13.3‰2720 ± 55
OxA-5021human bone, 171, δ13C = −11.6‰2590 ± 55
OxA-5037human bone, 174, δ13C = −13.9‰2715 ± 75
OxA-5312human bone, 173, δ13C = −11.6‰3215 ± 65
OxA-5776human bone, Burial 27, F79, δ13C = −14.5‰2510 ± 55

Comment (N. H.): the early village community of Cuello yielded the first, longest and most securely-stratified Preclassic radiocarbon chronology in the Maya Area (Hammond et al. 1976, 1977). based on congruent dales from two laboratories (Cambridge (Q) and Los Angeles (UCLA)) that suggested a beginning for village life before 2000 BC. Subsequent dates for other sites and further dates for Cuello itself from Cambridge and La Jolla (LJ), as well as an early Arizona AMS date (AA-458) suggested this chronology was too early (Hammond 1991). A collaborative programme with ORAU utilised bone collagen from human burials, well-stratified time-capsules with associated grave goods including stylistically dated pottery vessels, in place of the burnt wood hitherto used for dating (Housley et al. 1991).

The research also included assessing the effect of conservation chemicals on the dating of bone (Law et al. 1991), and was later extended to include the dating of the first macrofossils of the root crop manioc to be identified in the Maya Area (OxA-4452: Hather and Hammond 1994) and the dating of carbonised food remains from an apparent ritual feast dumped into a subterranean chultun chamber along with brewing jars, drinking cups and an odd assortment of burnt flints (OxA-4662, −4663, −4682, −4683, −4685), which confirmed the early first millennium B.C. date suggested by the Bladen Xe phase ceramics (900–600 BC: Kosakowsky and Pring 1998). A single date (OxA-4542) on accumulated debris from the old land surface suggested that this continued to build up into the Bladen phase.

Further dating of Bladen phase burials from the excavations of 1990–1993 (Hammond et al. 1991, 1992, 1995: OxA-3084, −3664, −4458, −4459, −4460, −5021, −5037, −5312) confirmed this placement in all but one case: OxA-5312 yielded a date (1640–1370 cal BC at 95.4% prob.) centuries older than either the stratigraphy or the accompanying Consejo Red Estrella variety bowl of the Bladen Xe ceramic complex indicate. OxA-5037 adjacent to it and apparently coeval yielded a date (1050–760 cal BC at 95.4% prob.) within expected limits, while also resolving a stratigraphic conundrum and simultaneously dating the oldest known occurrence of Romero's Type A-1 dental modification in the Maya Area (Hammond et al. 1995, 122).

The dates for Swasey phase (1200–900 BC) burials (OxA-4461, −5020) provided welcome confirmation that this period begins in the late second millennium B.C. after some expressed doubts (Andrews in Andrews et al. 1990, 579). The single determination (OxA-3085) for the Lopez Mamom phase Burial 154 was slightly later than the expected upper limit of 400–300 BC, while that for the early Cocos Chicanel phase Burial 27, a sacrificed adolescent buried on the axis of the first temple pyramid (OxA-5776) was more than a century older than expected from the stratigraphy and accompanying pottery.

Overall, the collaborative research on dating the early Maya community at Cuello has been extremely productive and satisfactory, demonstrating that dating of the Maya themselves through their skeletal remains is a useful alternative to dating the debris from their domestic and ritual activities. The δ13C measurements, ranging from −10.8‰ to −14.5‰, have shown a rate of maize consumption consistent through time and across the age/sex range, and comparable with specific research on this topic carried out by Tykot et al. (1996).

La Milpa

A sample of bone and a sample of charcoal from the site of La Milpa (17:50N 89:03W) which is in northwestern Belize, submitted by N. Hammond, Dept. Archaeol., Boston Univ., USA, funded by the NERC-funded ORADS facility.

OxA-6810human bone, B 11.67, δ13C = −17.0‰1750 ± 50
OxA-8079charcoal, B88.04, δ13C = −25.6‰1220 ± 35

Comment (N. H.): La Milpa is a major Classic Maya city in northwestern Belize, but on the 20 stelae only one legible hieroglyphic date, of AD 780, survives. Ox A-6810 dates a royal tomb (Op. B11) with depauperate grave assemblage but elaborate construction and closing ritual, of a 35–50 year-old male (Hammond et al. 1996). Pottery vessels in the tomb were assigned to AD 450 ± 40 on stylistic grounds: OxA-6810 suggests that the true dale may fall at the beginning of this range, although this conflicts with the assignment of a vessel identical to one in B11, from Tikal Burial 94 (Culbert 1993: fig. 32e), to a date late in the Manik ceramic complex (AD 250–550), i.e. AD 500 ± 50. Theδ13Cof −17.0‰ suggests a surprisingly low ingestion of maize for any Maya, especially one of such high status.

OxA-8079 dates pine wood (Pinus caribaea) from a burning rite accompanying the infilling of the throne-room in Structure 38 of the palace acropolis at La Milpa (Hammond et al. 1998). The higher OxCal likelihood of 760–890 cal AD (80% prob.) spans the known period of La Milpa's Late/Terminal Classic prosperity, and in the absence of directly-associated ceramics places the construction, use and disposal of the throne room within this phase.

Toledo District

A sample of wood from Toledo District (16:30N 88:45W), southern Belize, submitted by N. Hammond, Dept, Archaeol., Boston Univ., USA, funded by the SERC.

OxA-2209?Manilkara zapota wood, M5–1. δ13C = −26.8‰1650 ± 70

Comment (N. H.): in the 1980s a seated human figure carved in the tropical hardwood chicozapote was found in a cave in the Maya Mountains of southern Belize, on the headwaters of the Monkey River. Wooden artefacts are extremely rare survivals from the Classic period of Maya civilisation: this figure was stylistically Early Classic (AD 250–600) and resembled a better-preserved but looted example reported by Ekholm (1964). OxA-2209 is important in confirming an Early Classic date for such seated human figures, and by extension also indicates Early Classic occupation of this poorly-explored part of the Maya Lowlands and coeval use of caves for ritual (Stuart and Housley 1999).


Santa Rosa

Samples of wood from Santa Rosa (13:50N 90:20W), Pacific coast of southeastern Guatemala, submitted by N. Hammond, Dept. Archaeol., Boston Univ., USA. The excavations were directed by F. Estrada Belli.

OxA-7779wood, TP5, δ13C = −27.4‰2935 ± 65
OxA-7780wood, TP7, δ13C = −24.6‰2890 ± 65
OxA-7781wood, TP8, δ13C = −23.5‰1440 ± 60

Comment (N. H.): OxA-7779, −7780 provided the first secure ehronometnc data for anchoring the late Early Formative beginning at Chuquihuitan of a long stratigraphic and ceramic chronology for the entire southeast Guatemalan coast, a region archaeologically unexplored until the 1990s. OxA-7781 provides good correlation for the cross-dating of the Middle Classic pottery at the principal centre of this period. La Maquina, at about AD 550 with similar ceramics in the Escuintla region to the west. Together the three determinations anchor the Santa Rosa regional chronology and show it to be congruent with the sequences from the central and western areas of Pacific Guatemala (Estrada Belli 1999).

Naj Tunich

Samples from Naj Tunich, Peten (16:16N 89.15W), Guatemala, submitted by J. E. Brady, Dept. Anthropol., California State Univ., Los Angeles, USA.

OxA-5832charcoal, food residue, IV-42, δ13C = −12.8‰1835 ± 50
OxA-5833charcoal, food residue, IV-43, δ13C = −15.3‰1890 ± 45
OxA-5834charcoal, food residue, VI-3c, δ13C = −16.1‰1770 ± 45
OxA-5835charcoal, food residue, VI-3h, δ13C = −16.6‰1960 ± 45

Comment (J. E. B.): Naj Tunich, located in southeastern Peten, Guatemala, is a cave site famous for the large corpus of inscriptions and drawing in its 3.5 km of tunnels (Brady and Stone 1986). The entrance chamber is over 100 m long and has a natural rise at the eastern end that has been modified through filling and levelling into a two-tiered balcony structure. The ceramic analysis indicated that much of this construction occurred during the poorly dated protoclassic.

Four carbon samples were submitted to ORAU in an attempt to define better the Protoclassic chronologically. Two samples were submitted from each of two different excavations. The sample from Lot IV-42 (OxA-5832) came from the first 50 cm of a test pit in a small earthen platform on the upper level of the balcony. Lot IV-43 (OxA-5833) came from the level immediately below Lot IV-42 and was terminated at bedrock at 72 cm. Except for a few later sherds found on the surface, all the ceramics from both lots were protoclassic in date.

A second test pit was excavated in the floor of the entrance chamber, just in front of the balcony structure. Lot VI-3C (OxA-5834) came from a layer of yellow clay 23–29 cm below datum that overlay a thin black lens that may have been a use-floor. Lot VI-3H (OxA-5835) came from the bottom of the pit at 117 cm. Once again, except for later sherds in the surface lot, all the ceramics from the pit were protoclassic in date.

All four samples were composed of carbonized material burnt to the inside of ceramic sherds. Combustion of other samples produced an aromatic fragrance suggesting copal incense; a tree resin perhaps mixed with maize. In both test pits, the radiocarbon results were consistent with the samples’ stratigraphic position and, as expected, OxA-5835 yielded the earliest date. OxA-5832, −5833, and −5834 fell very much where they were expected. OxA-5835 falls earlier than most archaeologists would accept but, given its deep stratigraphic position, I suspect it is correct. For an extended discussion of these and other protoclassic dates see Brady et al. (1998).


Lapo Do Sol

Samples of charcoal from Lapa Do Sol, Santarem, Brazil (02:27S 54:22W), submitted by M. Otte, Préhistoire, Univ. Liège, Belgium to date previously undated Brazilian art.

OxA-7959charcoal, 1, δ13C = −25.3‰1260 ± 80
OxA-7960charcoal, 2, δ13C = −23.0‰65 ± 60
OxA-7961charcoal, 3, δ13C = −23.5‰65 ± 70

Comment (M. O.): OxA-7960 and −7961 are clearly contaminated. For comment on OxA-7959 see Otte and Etchevrane (in press).



Sample of bone from Wariwilka, Mantero Valley, Peru, submitted by Dr. K. Reinhard, Dept. Anthropol., Univ. Nebraska, Lincoln, NE 68588–0368, USA.

OxA-7630human bone, WW-1, δ13C = −10.3‰1115 ± 35

Comment (K. R.): OxA-7630 is important because it is from a sacrificed individual in the Andes. It is probably the oldest high Andean sacrifice. The archaeological context was pre-Inca, but it was previously not possible to determine whether the sacrifice was of the Waanka culture or older. It is older than the Waanka culture and is, therefore, much older than anticipated.

Falkland Islands

West Point Island

Samples of Phalcobaenus australis bone from West Point Island (51:21S 60:42W), Falkland Islands, submitted by M. Adams, Bird Group, The Natural History Museum, Tring, Hertfordshire.

OxA-7731Phalcobaenus australis bone, WPI 1, δ13C = −12.1‰5335 ± 50
OxA-7732Phalcobaenus australis bone, WPI 2, δ13C = −10.7‰5250 ± 50
OxA-7733Phalcobaenus australis bone, WPI 3, δ13C = −12.1‰5255 ± 45
OxA-7734Phalcobaenus australis bone, WPI 4, δ13C = −13.0‰5425 ± 55

Comment (M. A.): four bone samples from the species Phalcobaenus australis were submitted for dating in November 1997. These samples were part of a collection of over 200 bones and bone fragments collected from a single peat-bog in December 1996. The remaining samples have been cleaned and preserved, and the majority successfully identified to species level. It is intended to publish a full report of this research to include the nature of the site, a list of species involved and of course the age of the bones. The age of the material is of key significance as this far predates any known human activity on the Falkland Islands.

Asia and Australasia


Wupu cemetery, Hami

Sample of wool from the site Wupu cemetery. Hami, Xinjiang, (42:37N 93:32E), China, submitted by I. Good, Center for the Study of Ancient Textiles, Westwood, MA 02090, USA.

OxA-5771wool textile, Grave 25, δ13C = −20.1‰2525 ± 60

Comment (I. G.): OxA-5771 was from a burial in the Wupu Cemetery in Xinjiang. It was derived from a woollen textile fragment found on the partially mummified mortuary remains of that grave. This radiocarbon sample was the only independently tested sample (outside of Chinese laboratories), and demonstrated a date at considerable variance with published dates from the Chinese. A reasonable explanation is that there is a considerable timespan within the Wupu Cemetery, but only one or two samples were taken to provide dates published in the Chinese archaeological literature, which date between 3030 ± 85 BP and 2760 ± 85 BP (Mallory and Mair 2000).

This sampling was a very important step towards understanding the complexity of the Xinjiang desert oasis sites, and emphasises the need to better understand the chronology, independently of typological studies. The reader is referred to Good (1995; 1997).


Noen U-Loke and Spirit Cave

This series of dates was acquired using samples of natural plant resins applied as coatings on prehistoric ceramics from the sites of Noen U-Loke, Mun River Valley (15:15N 102:16E), and Spirit Cave, Mae Hongson Prov, (19:34N 98.07E), Thailand. The resin coatings have been analysed using GC-MS and results show significant similarities to modern ‘dammar’ resins from the Dipterocarpaceae family of trees. The sherds were examined under a binocular microscope and the resin samples collected using a scalpel to remove the visible organic residues from the surface of the potsherds. The samples were submitted by C. Lampert and C. Heron, Dept. Archaeol. Sciences, Univ. Bradford. W. Yorkshire, through the NERC-funded ORADS facility.

Noen U-Loke  
OxA-10268pottery residue, BradSEA 01/NUL#208 A2 4:5 4, δ13C= −27.9‰1900 ± 37
OxA-10269pottery residue, BradSEA 02/NUL#208 A2 4:5 4, δ13C= −28.1‰1861 ± 35
OxA-10270pottery residue, BradSEA 03/NUL#351 B1 4:9.δ13C = −28.0‰2149 ± 35

Comment (C. L. and C. H.): three determinations (OxA-10268. −10269 and −10270) were obtained for this Iron Age site in the Mun River Valley, using interior resin coatings on two potsherds. The purpose in submitting this material for AMS dating was to assess the validity of radiocarbon determinations acquired using plant resin samples by comparing them with a series of well stratified dates from the site. All three determinations correlate well with the existing series (Higham and Thosarat 1998) and appear to indicate that resinous coatings on potsherds may offer a direct method of dating the ceramics, since a resin coating is unlikely to be applied except during the functional lifetime of a pot. Two determinations obtained trom separate samples taken from a single sherd (OxA-10268 and −10269) are in good agreement with each other, suggesting that the resin coating is reasonably homogeneous.

Spirit Cave

OxA-10271pottery residue. BradSEA 04/SC T19 c1-c2. 8N, δ13C = −25.4‰3042 ± 37
OxA-10272pottery residue. BradSEA 05/SC T19 c1-c2. 8N.δ13C = −25.4‰2995 ± 40

Comment (C. L. and C. H.): OxA-10271 and −10272, from this rockshelter site in north-west Thailand. were obtained from a single sherd. The first of these determinations an exterior resin coating, the other an interior resin coating. Again, the two determinations are in good agreement with one another, supporting the view that both coatings were applied contemporaneously. The resin was submitted for dating to try to resolve uncertainties concerning an early date originally suggested for these ceramics. The dates obtained from the resin samples are not in agreement with a published series of bamboo-charcoal dates (Gorman 1972), which suggested that the ceramics were of late Hoabinhian origin. Neither does it appear that the potsherds were introduced to the site by disturbance of later log coffins (thought to be Iron Age), but suggest a continuation in the use, possibly intermittent, of the rockshelter during the Late Neolithic-Bronze Age. Further determinations from resin-coated ceramics are needed before these results can be fully supported or refuted. The reader is referred to Lampert et al. (in press).

Khok Phanom Di

Samples of plant remains from the site of Khok Phanom Di (13:35N 100:8E), Thailand, submitted by B. Vincent, PO Box 5165, Dunedin, New Zealand.

OxA-7777rice husks, KPD CP41, δ13C = −25.1‰4095 ± 70
OxA-7778rice husks, KPD CP42, δ13C = −24.9‰4285 ± 55

Comment (B. V.): the excavation of Khok Phanom Di, central Thailand in 1984–5 was directed by C. F. W. Higham and R. Thosarat. Two sets of conventional radiocarbon dates were obtained. Those from the Institute of Nuclear Sciences, Wellington, were internally inconsistent. The larger series from the Australian National University was internally consistent, and indicated initial settlement at about 2000 BC, and the end of the use of the cemetery five centuries later. No charcoal was recovered from the uppermost 1.2 m of 6.8 m of prehistoric deposits. Typical rice-tempered sherds from this level, which contained a potting workshop, were therefore submitted by me for AMS dating.

In my opinion, the results will require careful consideration to be given to the Khok Phanom Di sequence as it stands because pottery recovered from the lowest levels of the site parallels that from the earliest phase of Nong Nor, a nearby site which gave earlier determinations of about 2450 BC. Either the entire sequence from Khok Phanom Di is too late or the Nong Nor Phase 1 determinations are too early. The alternative, of course, is that the OxA-7777 and −7778 determinations are too early.


Cemetery Beach, Norfolk Island

Samples of bone from Cemetery Beach, Norfolk Island, Australia (29:02S 167:57E), submitted by A. Anderson, Centre for Archaeol. Research, Australian National Univ., Canberra ACT 0200, Australia.

OxA-5781Rattus exulans bone, Nfi 001, δ13C = −19.2‰495 ± 55

Comment (A. J. A.): OxA-5781 is from unit C4, 140–155 cm. The approximate age of this unit on multiple charcoal samples is twelfth century AD (Meredith et al. 1985).

New Zealand

Shag River Mouth

Sample of bone from Shag River Mouth, New Zealand (45:30S 170:50E), submitted by A. Anderson, Centre for Archaeol. Research, Australian National Univ., Canberra ACT 0200, Australia.

OxA-5780Rattus exulans bone, SM 001, layer 5, area 3, δ13C = −16.8‰900 ± 55
OxA-6020Rattus exulans bone, BR 071–02, layer 6, area C, δ13C = −18.8‰930 ± 65

Comment (A. J. A.): OxA-5780 and −6020: the approximate age on multiple samples of other materials is fourteenth century AD (Anderson et al. 1996). The agreement of the rat collagen samples with measurements on other materials is not precise, but not too far away either. This is very encouraging in view of some other measurements obtained on rat bone collagen.

Historic Scotland

The Oxford laboratory provided Historic Scotland with its radiocarbon determinations between July 1999 and June 2001. Most determinations were obtained from sites dug during landscape-oriented campaigns in areas threatened by natural forces, such as marine and sub-aerial erosion in the Western Isles, or insidious threats such as ploughing in eastern Scotland. A significant number were from sites excavated in advance of trunk road building. A further source of evidence came from dates obtained during the writing up of old excavations and re-dating of material from significant published excavations. Some dates were obtained for material from excavations which were not otherwise sponsored by Historic Scotland, where the evidence appeared to illuminate that from active Historic Scotland projects and radiocarbon determinationswould not otherwise have been obtained. Several palaeoenvironmental sequences were dated in association with archaeological work on nearby sites.

The vast majority of archaeological dates were from single pieces of charred material or bone (Ashmore 1999, 24–30). Where possible more than one sample was dated from each context. The determinations demonstrate how common residuality is on Scottish sites, and throw doubt on results obtained from a mixture of several pieces of charcoal. New dates from samples from Knap of Howar, Stones of Stenness and Lundin Links suggest that Baillie's contention that the error estimates attached to some 14C measurements obtained in the early 1980s were underestimated by a factor of around four may be correct (Baillie 1990, 36–6). However the evidence from re-dating of samples from Lundin Links and Knap of Howar may equally well suggest that errors were not normally distributed: most errors may have been underestimated by a factor between one and two, with a few determinations having markedly greater errors.

The dating of poorly preserved bone improved during the period of the contract, and the use of stable isotope ratios to assess success (in minimising the retention of carbon from material unrelated to that in the bone at the time of death of the organism providing the sample) allowed some date estimates to be corrected by re-measurement with more appropriate pre-treatment.

The known problems associated with the dating of cremated bone and pot residues using traditional fractions meant that few such samples were submitted for dating. That the three cremated bone determinations from Fordhouse reflect contamination with later carbon was confirmed by a date subsequently obtained from the carbonate fraction of bio-apatite at the Groningen laboratory by the National Museums of Scotland (Sheridan pers. comm.). Bone carbonate dating seems to be the best way forward, at least for highly fired bone. It is difficult to tell whether all of the pot residue measurements from Littleour, Deer's Den and Ceann Nan Clachan are accurate. The pot residue from Ceann Nan Clachan produced a date compatible with those from barley grains. Two of the three results from Littleour are in agreement with dates from charcoal but the other is significantly different. Two determinations from pot residues in a small pit at Deer's Den produced significantly different dates from each other and from dates from charcoal in other seemingly similar pits. One confounding factor in assessing what these apparent discrepancies mean is the possibility that people in these complex societies kept heirlooms. Nevertheless, it does not seem safe to rely solely on pot residue dates, although there is little doubt that some provide reliable information about the dates of the pots and of the contexts in which they were found.

The Historic Scotland dating programme includes one or two sequences which seem difficult to explain because it is not possible to distinguish between the possibilities of residual or intrusive material, or errors in excavation or post-excavation, or failings in administrative, pre-treatment or measurement procedures or statistical flukes. The dates from An Dunan (part of the Uig landscape project, Arch. List 30) are odd. The dating reversals from the Dunadd, Cille na Creige, Tarsuinn and Polmsh pollen columns (see Arch. List 30) are easily explained in general terms by inclusion of residual material in the sequences, but it is less easy to prove any particular explanation. However, the vast majority of the 14C ages are entirely credible (in the sense that there are no reasons to suspect errors other than the normal error of measurement) particularly if it be remembered that about 20 of the ages in this list will be wrong by more than twice their errors and four by more than thrice.

The dating programme demonstrates conclusively that it is essential to obtain multiple single-entity determinations rather than determinations for bulked-together bone or charcoal. The danger of misinterpretation arising from measurements from residual material is far less than that of obtaining ‘average’ dates from pieces of organic material of very different date. Many more AMS analyses will be needed before we can be truly confident that the dated sites are typical of their areas, and there are many remaining chronological and geographical problems to be resolved, but overall the dating programme has been hugely successful, throwing light where previously there was, if not darkness, a fog.

The period before about 4000 BC

The dates obtained from Oxford for hunter-gatherer sites, notably those from Cramond (Edinburgh), suggest that human occupation in Scotland began well before the end of the ninth millennium BC and probably nearer its middle. Indeed, human occupation in Scotland seems to have been earlier than the spread of most tree species apart from birch, hazel and willow. Edwards and Whittington (1997, 66) suggest that pollen evidence shows that hazel spread from the far west from about the middle of the tenth millennium BC, very soon after the end of the Loch Lomond Stadial. Hazel pollen is however difficult to distinguish from that of bog myrtle. Direct dating of hazel wood and nut shells from archaeological contexts, including those from Cramond (Edinburgh), Fordhouse (Angus) and Camais Darach (Skye), currently suggest that people's use of it spread from the south-east about the middle of the ninth millennium BC. More generally, the determinations measured by ORAU during this contract reverse the previous tendency to suppose that hunter gatherers spread up the west coast before they exploited the east. But the total amount of evidence, and the difference in date, are small, and expansion of populations on a broad front remains a possibility.

Determinations from the Scotland's First Settlers project do not confirm a hypothesis that hunter gatherers settled down in the fifth millennium BC and adopted farming: the dates from Scotland's First Settlers: Sand (Lochalsh) and Loch A Squir, Raasay (Skye) come from the eighth and seventh millennia BC, the latter site also producing dates from the first millennium BC. The sites at Ben Lawers (Perth and Kinross), Home Farm, Castle Menzies (Perth and Kinross) and Chapelfield, Cowie (Stirling) confirm a pattern in Scotland of early use of sites used for much later structures. Between them they demonstrate inland activities from the eighth to the fifth millennia BC. The cave site at Raschoille near Oban (Argyll and Bute) contains evidence for a variety of activities from the seventh to the fifth millennium BC, after which the cave was used for depositing human skeletal material through and beyond the fourth millennium BC.

Fourth millennium BC

The determinations from the timber alignment and arc of postholes at Home Farm, Castle Menzies in central Scotland, are amongst a small number of reliable dates from the first quarter of the fourth millennium BC. In this period people who had traditionally not included farming in their subsistence strategies may have occupied the land along with those who did. The dated structures are different from the structures usually associated with the concept of an early “Neolithic Period”. The dates of the second quarter of the fourth millennium and later from Raschoille Cave near Oban (Argyll and Bute) demonstrate that the local people chose to finally dispose of at least some of the remains of their dead in much earlier midden under a natural overhang, rather than in the putatively contemporary chambered tombs of the area. The practice was continued for at least half a millennium. From the other side of the country, Fordhouse barrow near Dun (Angus) shows a site with much evidence for pre-agricultural activity used for formal burials in a chambered mound in a traditionally “Neolithic” style as early as any other mounds in Scotland. The results obtained by ORAU for midden at Knap of Howar (Orkney) can now be seen as showing that the site belongs to the mid fourth millennium BC, and is clearly separate in time from sites with grooved ware, resolving a long-lasting ambiguity created by previously obtained dates (Ashmore 2000, 300–3). The AMS results from Holm of Papa Westray (Orkney) appear to demonstrate that some sheep in the islands had a partly marine diet. They also show closure of a stalled tomb at a date similar to those from animal bones in three tombs with broadly similar characteristics on another Orkney Island, Rousay (Renfrew 1979, 206). A date was obtained for a hearth at Garvald Burn (Scottish Borders) near a ‘late Mesolithic’ chert scatter and knapping floor. It may suggest a continuation of hunter-gatherer activities into the first half of the fourth millennium BC in some parts of southern Scotland.

The occurrence of early fourth millennium BC pits with pottery and lithics characteristic of this period at Kintore Bypass, Deer's Den, (Aberdeenshire) (see Arch. List 28) suggests that farmers occupied the same lowland areas as later peoples and hints at the potential for reinterpretation of settlement in this period. The house at Kinbeachie (Ross and Cromarty) (see Arch. List 28) is broadly comparable to some of the numerous and roughly contemporary houses know in Ireland and suggests that further excavation in the plough-lands should demonstrate that there was to some extent a ‘settled Neolithic’ in Scotland, whatever the situation in southern England. The evidence from Ben Lawers is complex: a date of this period from the upper part of a later pit may be connected with Neolithic pottery found nearby. The settlement on Eilean Domhnuill (Western Isles) (see Arch. List 30) can be interpreted in at least two ways, as a very individual but not atypical settlement, or as a special place, and it suggests a need for reinvestigation of long-known sites such as Eilean an Tighe (Scott 1953). The crude timber platform at Parks of Garden (Stirling) suggests exploitation of marshy environments in central Scotland, though it is not clear whether there was settlement of this period in the wetlands themselves.

The new determinations provide strong support for the idea that there were many regional ‘Neolithics’ in Scotland and that the ‘Neolithics’ of the plough-lands have not been explored adequately in most interpretations of Scottish prehistory (Barclay 2000, 275). The discoveries at Raschoille suggest that the same may be true of the west.

Third millennium BC

The animal bones at the base of the ditch at Stones of Stenness in Orkney were dated afresh, showing that the monument belonged around 3000 BC and was, as previously believed, early amongst major ceremonial sites with Grooved Ware. By way of contrast, the dates from Littleour (Perthshire) suggested that Grooved Ware continued in use until well after the middle of the millennium. That late date is supported by ages from another laboratory for a site at Auchlishie in Angus (Dick 2000). The cist from Mill Road, Linlithgow (see Arch. List 28) containing six inhumations is very securely dated to the last quarter of the third millennium BC. Although there are five earlier dates from cists in Scotland, all of them except that from the multiple burial in a massive cist at Sumburgh, Shetland, have question-marks attached to them. Perhaps these multiple burials support the idea that burial rituals common in the fourth millennium were not replaced rapidly by single burial customs; but it is also notable that the dates are roughly contemporary with the earliest seemingly reliable dates for inhumations with beakers in cists. Amongst these is that from the Haddington Mains cist, which is surprisingly early, given the typology of the beaker, even if comparable to the previously earliest seemingly credible date for a beaker (with a burial in a pit) from Sorisdale, Coll. Admittedly the Ben Lawers date discussed above, with dates from sites of the fourth millennium and that obtained many years ago from a cist at Skateraw in East Lothian, are earlier; but the former date is not directly for the beaker and the latter does seem to require re-measurement. Perhaps however a more fruitful exercise might be a large-scale re-examination of Scottish beaker typologies using AMS techniques to obtain smaller errors than were available for Kinnes and his collaborators’ study (Kinnes et al. 1991). The dates for the late third millennium (or very early second millennium) activity at Silgenach provide the first reliable multi-date set for this period in the Western Isles, although a single date from the house at Kildonan (Zvelebil 1989, 1991) suggests it may be of around the same date. The ages from the midden at Northton and from agricultural activity at Rosinish require re-measurement; the former requirement should be satisfied this year in association with production of a full report.

Second millennium BC

The Callanish kerb cairn lies on the line of the avenue of the Calanais stone setting, which has dates suggesting periods of activity involving creation of charred material in the period between 2900 and 2600 BC and between about 1950 and 1750 BC. It is possible that the charred grain relates to agricultural activity at the time of the earlier set of dates from the main setting. The main use of the cairn could be roughly contemporary with the second period of activity at the main setting, or later. It may have been contemporary or slightly earlier than a very different set of activities at West Water, at the northern end of the Scottish Borders, involving burials in pits. West Water is particularly celebrated for the discovery of lead beads. A house from a settlement at Bayanne is one of the very few so far securely dated to this period in Shetland. The function of the ring ditch at Kintore Bypass, Deer's Den (see Arch. List 28) in eastern Scotland is difficult to ascertain. It is contemporary with some of the round houses at Lintshie Gutter in Clydesdale and some of the round houses at Achany Glen, Lairg in Northern Scotland The dates for late cremations at Fordhouse barrow in Eastern Scotland are included here, despite their being measured by the Oxford laboratory as belonging to the end of the first millennium BC, because re-dating at Groningen suggests a much more credible date in the second quarter of the second millennium BC. Pits at Silgenach (Western Isles) produced dates from barley grains which may have been residual because the pottery in the pits is of a style thought to be over half a millennium later.

The mid-millennium dates from Coast Road, Longniddry (East Lothian) for crouched inhumations are important in helping to fill a 500 14C year gap in the record of burials in East Lothian but are very broadly comparable to dates from crouched burials in cists at Dryburn Bridge, East Lothian and Grainfoot, Longniddry. Lothian burial customs appear often to be different from those in other parts of Scotland, in particular in the 1st millennium BC and the easy assumption that these burials have a close affinity with crouched burials accompanied by beakers in cists should be challenged. One of the sets of human remains from St Nicholas Farm (Fife) (see Arch. List 30) was of this period; surprisingly early, for the bones had been supposed to relate to use of the site as a hospital in the medieval period. An urn from Glennan (Argyll and Bute) should have belonged to this period but instead produced a much later date, and to further confound expectations a second date obtained from another laboratory produced an anomalously early date.

Both of the burnt mounds, at Tangwick (Shetland) (see Arch. List 30) and Cleuchbrae, Annandale (Dumfries and Galloway) appear to be multi-period. Broadly speaking their dates are similar to one another, neither showing a high probability of activity in the middle of the millennium. Although their dates can be matched by those of other burnt mounds elsewhere in Scotland there is little suggestion that their diachronic patterns are dominant elsewhere, although they certainly occur, but it will obviously be important to obtain multiple single entity dates from all burnt mounds in future.

First millennium BC

This period saw major changes in Scotland. The archaeology of the period between about 1000 and 800 BC is (very incompletely) known from dated structures in various parts of Scotland: substantial houses in Perthshire and Jura, settlements of round houses on hillside platforms in the south and burnt mounds in the north, and late use of ceremonial sites near the Moray firth. In Edinburgh the Castle rock was occupied and a settlement had started up on what was later to be a very substantial occupation site in Orkney. Altogether it suggests many societies of the first quarter of the first millennium were not very different from their predecessors of the latter part of the second millennium BC but others presaged societies of the subsequent period. However, none of the dates obtained during this contract belong securely to this fascinating first fifth of the first millennium BC.

The earlier Iron Age radiocarbon calibration plateau is almost matched by a later plateau; but at least that means that we can distinguish earlier and later periods. The dates from between the eight and sixth centuries BC include those from the scooped house at Kintore Bypass: Deer's Den (Aberdeenshire) (see Arch. List 28) and fort at the Brown Caterthun (Angus), a large multivallate structure formed by interrupted banks. In the Western Isles occupation was resumed at Silgenach (Western Isles) and started at Uig Landscape Project: Gob Hirer (Western Isles). The cellular buildings at Ceann Nan Clachan, Vallay Strand (Western Isles), are important in showing that this building style was present much earlier than the well known cellular buildings found in settlements centring on Atlantic roundhouses. The dates from after the gap between the two dating plateaux include a subrectangular house from Uig Landscape Project: An Dunan (Western Isles) and a round house at Kintore bypass: Tavelty (see Arch. List 28), probably reflecting regional differences in house styles. Two dumbbell shaped ovens from Kintore bypass: Rosebank (see Arch. List 28) have dates significantly different from each other. They may relate to the Kintore Roman temporary marching camp.

First millennium AD

Although there are extremely important dating sequences from settlements such as those at Kintore bypass: Deer's Den, Kintore bypass: Rosebank (see Arch. List 28) and Auchlishie in the east, Bayanne on Shetland and Bornais in the Western Isles, the major revelation provided by the ORAU detrminations during this contract has been the dates from cemeteries, mostly but not exclusively in eastern and south-eastern Scotland.

The radiocarbon-dated burials include those from a long cist cemetery at Galson in the Western Isles with a span ending in the mid sixth century judging by dates obtained before this contract; the new dates from Oxford suggest that burials started at a date between the first and third centuries AD. This is remarkably early. Other new evidence provided by Oxford for various small and large sites, most notably Thorneybank near Dalkeith (Midlothian), suggests that the immediate origins of eastern Scottish long cist cemeteries lay in south-east Scotland, in an area not traditionally associated with Picts, very probably by the fourth century AD. The small cemetery encountered during work in advance of A1 Dualling, Dunbar (East Lothian) may be early compared to other eastern of long cists cemeteries, although there are concerns about possible contamination. The later fifth to early seventh century cairn cemetery at Lundin Links (Fife) may reflect burial traditions emanating from Angus, because one of the round barrows from Red Castle (Angus) predated the fifth century. It seems likely on the present limited evidence that the two traditions, of long cist and of cairn cemeteries, met around the late fifth century at Lundin Links, and that by the late seventh century AD the practice of burying people under cairns became unpopular there, although it may have remained in favour in northern Scotland until about the end of the millennium. Other burials were dated in various parts of Scotland. At Loch Borralie (Caithness), a burial possibly representing a variant on the tradition of burying people in small cairns. A burial in a long cist and another in a round cist at An Corran, Boreray (Western Isles) were roughly contemporary with one another. A date from an inhumation at Cnip (Western Isles) incorporates a significant marine effect; and may be Norse despite a date nominally too early. The long cist burial at Ben Lawers (Perth and Kinross) is interesting as one of the most westerly of the eastern Scottish distribution. Three dated burials from St Nicholas Farm (Fife) showed that the hospital seems to have been built in an area previously used to dispose of human remains.

Another important theme is the appearance of dated Norse structures succeeding re-use of a wheelhouse at Bornais (Western Isles); evidence for the Norse is discussed in more depth below with sites of the second millennium BC. The metalworking site at Meadows, Dornoch ((Easter Ross) is clearly of very considerable interest and may have some bearing on the subsequent creation of the medieval burgh.

Lastly, an otherwise completely undated timber platform at Kilmore (Argyll and Bute) was shown to belong in this period while a date from Berryhill (Aberdeenshire) suggests that an enclosure which might have been a plantation or other recent boundary may well be of the first millennium AD.

Second millennium AD

The determinations from Bornais and Smoo cave probably reflect Norse activity, helping to fill out knowledge of their activities in areas where a decade ago few archaeological traces of the Norse were known. The discoveries at Bornais are of particular importance because of the richness of the artefact assemblages.

Some determinations from middens sampled in the Scotland's first Settlers project provided evidence for much later activity than had been expected. The midden at Scotland's first Settlers: Ashaig 1 is at least in part medieval while that at Scotland's First Settlers: Crowlin 1, Lochalsh (Highland) is early first millennium at least in part and a limpet scoop is late medieval in date.

Many of the other second millennium AD dates come from cemeteries, perhaps reflecting the relative ease with which such sites are recognised. They confirm the prevalence of burial in long cist or long grave cemeteries, often centred on chapels, and there are no traces amongst the dated burial places at Newark Bay (Orkney), Cille Bhrea, Lemlair (Easter Ross), Stirling Castle (see Arch. List 30), Wishart Avenue, Montrose (Angus) (see Arch. List 30) or Colonsay House (Argyll and Bute) (see Arch. List 28) of continuing pagan practices.

The Scottish bloomeries project has been very important in revealing details of medieval iron industries in the west and north-west, reflecting perhaps the ease of recognition of sites in this area but confirming that even at this early period fuel supplies helped to control the location of bloomeries. The Ben Lawers and to an even greater extent the Uig landscape project (see Arch. List 30) are starting to reveal medieval rural settlement, albeit in marginal areas, still very difficult to locate with any certainty from field remains alone. Lastly, a date from a scatter of shells at Newbigging, Burntisland (Fife) showed that a possible midden was not early prehistoric but, presumably, related to fairly modern agriculture.

A1 Dualling

Sample of bone from the site A1 Dualling, Dunbar, East Lothian (NGR NT 68257729) Scotland, submitted by T. Neighbour and M. Cressey, Centre for Field Archaeology, Old High School. 12 Infirmary Street. Edinburgh, EH1 1LT, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9378human bone, grave 3/SPOT, δ13C= −20.7‰1850 ± 45

Comment (P. A.): a single human bone from Grave 3 of a small long cist cemetery was dated to between the first and early fourth centuries AD. Although the date is credible because small long cist cemeteries may be early, the bone was not well preserved and experience with dating similarly poorly preserved bone from other long cist cemeteries in eastern Scotland suggests that it is easily contaminated by soil carbon.

An Corran

Samples of bone from the site An Corran, Boreray, Western Isles (NGR NF 85768051), Scotland, submitted by J. Downes, ARCUS, Univ. Sheffield, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-8784cattle bone, 001/27, δ13C = −20.7‰1640 ± 50
OxA-8802human bone, 002/7, δ13C = −19.4‰1875 ± 45
OxA-8803human bone, 003/8, δ13C = −19.4‰1815 ± 40

Comment (P. A.): three determinations from a small complex coastal eroding site suggested that a corbelled non-funerary structure dates between the late third and early sixth centuries AD, that a human interment in a long cist dates to between the early first and mid third centuries AD, and that another burial in a round cist dates to between the late first and early fourth century AD. The long cist burial is of broadly similar date to burials at Galson in Lewis, but in Scotland as a whole these dates are all earlier than those for long cist burials in eastern Scotland and, probably, earlier than those in Orkney, although one burial at Buckquoy has produced a broadly similar date.


Samples from the site Auchlishie, Kirriemuir, Angus (NGR NO 38705786), Scotland, submitted by A. M. Dick, 3, Grampian Crescent, Kirriemuir, Angus, Scotland. DD8 4TW, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-8773Corylus avellana, charred shell, 457.6, δ13C = −25.4‰1960 ± 35
OxA-8774Pinus sylvestus wood, 552.1, δ13C = −23.0‰290 ± 40
OxA-8796Alnus glutinosa charcoal, 608.2, δ13C = −26.5‰1885 ± 40
OxA-8797Corylus avellana, charcoal, 622.4, δ13C = −23.0‰1820 ± 35
OxA-8798Alnus glutinosa charcoal, 638.2, δ13C = −27.3‰1865 ± 45
OxA-8799Alnus glutinosa charcoal, 904.2, δ13C = −28.3‰1985 ± 40

Comment (P. A.): six determinations from a complex and long-used domestic site at Auchlishie in Angus help to provide a dating sequence which largely conforms with dating by artefacts. Two dates suggest that a scooped ring ditch house and a Dalladies-type souterrain belong between the mid to late first centuries AD and the earlier third of the second century AD, and three results from material in pits attributed to a post-ring house show that it probably dates to somewhere between the first and early fourth centuries AD. The other determination showed that robbing of a souterrain probably took place somewhat before the expected date of around 1800 AD.

Bayanne House, Shetland

Samples of charred barley grains from Bayanne House, Yell, Shetland (NGR HU 51959777), submitted by H. Moore and A. G. Wilson, East Archaeological Consultants, Unit 8, Abbeymount Techbase, 2 Easter Road, Edinburgh EH7 5AN, through Historic Scotland.

OxA-9894Hordeum sp. charred seeds, BY97 F.20(i), δ13C = −23.8‰1704 ± 34
OxA-9895Hordeum sp. charred seeds, BY97 F.561(i), δ13C = −24.4‰2980 ± 34
OxA-9896Hordeum sp. charred seeds, BY97 F.633(i).δ13C = −23.4‰2689 ± 35
OxA-9897Hordeum sp. charred seeds, BY97 F.641(i), δ13C = −25.6‰2752 ± 37
OxA-9898Hordeum sp. charred seeds, BY97 F.650(1), δ13C = −22.3‰2300 ± 38
OxA-9899Hordeum sp. charred seeds, BY97 F.6550), δ13C = −25.8‰2936 ± 35
OxA-9900Hordeum sp. charred seeds, BY97 F.7180), δ13C = −23.5‰3417 ± 38
OxA-9901Hordeum sp. charred seeds, BY97 F.7740), δ13C = −25.2‰3015 ± 40
OxA-9902Hordeum sp. charred seeds, BY97 F.775A(i), δ13C = −24.8‰3126 ± 36
OxA-9928Hordeum sp. charred seeds, BY97 F.699(id13C = −23.9‰1875 ± 60
OxA-9951Hordeum sp. charred seeds, BY97 F.764(i), δ13C = −23.1‰3216 ± 37

Comments (H. M. and A. G. W.): the complex, deeply stratified settlement at Bayanne, Yell, Shetland is, given the generally acidic environments of northern Shetland, remarkable for its preservation of bones and shells. Eleven AMS determinations from charred barley grains suggest that it was first occupied around 1700 BC. House 3 was occupied between about 1500 and 1250 BC and was abandoned by about 800 BC, although an associated outhouse was still in use at this time. House 1 was in use about about 800 BC and a third, House 2, was in use about 450 AD. The ages fit in well with the site sequence and they are particularly important in providing dates for pottery (there are very few dates securely associated with pottery in Shetland) and economic activities.

Bay of Moaness

Samples from the site Bay of Moaness, Rousay, Orkney Islands, (NGR HY 37752928) Scotland, submitted by K. J. Edwards, Dept. of Geog., Univ. Aberdeen, through Historic Scotland. Comments by P. Ashmore. Historic Scotland.

OxA-9011red deer bone, BM-1, δ13C = −21.4‰2890 ± 45
OxA-9012sheep/goat bone, BM-2, δ13C = −21.7‰3000 ± 50
OxA-9013red deer bone, BM-3, δ13C = −20.9‰2445 ± 45
OxA-9014sheep/goat bone, BM-4, δ13C = −20.1‰2770 ± 45
OxA-9015sheep/goat bone, BM-5.δ13C = −20.2‰1720 ± 40
OxA-9016ungulate bone, BM-6, δ13C = −21.5‰1780 ± 45
OxA-9072peat, BM-7, δ13C = −26.6‰5870 ± 45
OxA-9073peat, BM-8, δ13C = −27.4‰4285 ± 50
OxA-9074peat, BM-9, δ13C = −28.8‰1895 ± 40
OxA-9075peat, BM-10, δ13C = −28.0‰1320 ± 35
OxA-9076peat, BM-12, δ13C = −26.1‰5320 ± 40
OxA-9077peat, BM-13, δ13C = −27.6‰3425 ± 50
OxA-9078peat, BM-14, δ13C = −28.8‰1215 ± 35
OxA-9128peat, BM-11, δ13C = −28.1‰6885 ± 55

Comment: (P. A.): fourteen determinations from intercalated deposits of marl and peat on the coast of Rousay provided a sequence of events in what are now banded clay and peat intertidal deposits. The onset of peat formation was dated to the earlier half of the sixth millennium BC, with massive peat formation in the earlier half of the fifth millennium. There was a fall in woodland pollen around the last quarter of the fifth millennium, which may suggest human intervention half a millennium earlier than we have evidence in Orkney for structures typical of farming communities. An expansion of herbaceous taxa at 3080 to 2700 BC is of unknown significance in terms of human occupation nearby Four determinations suggest that between the late second and mid first millennia BC remains of butchered animals were deposited in the area, and two other determinations suggest something similar happening in the mid second to early fourth centuries AD. There was a renewed expansion of herbaceous taxa between the mid seventh and late eighth centuries AD followed by a soil capping some time between the late seventh and mid tenth centuries AD. The sequence is intriguing for at least two reasons. The data can be used to help test the model favoured by Lambeck (1995) of Orkney being on a fore-bulge during the glacial period and subsequently sinking consistently. Pollen and macroplant analysis should allow a fuller view of land use both prior to the first farming in Orkney and during the later fourth and earlier third millennia when several chambered tombs were in use on the southern flanks of the island.

Ben Lawers

Samples of charcoal from the site Ben Lawers, Perth and Kinross, (NN 61293928) Scotland, submitted by J. Atkinson, GUARD, The Gregory Building, Lilybank Gardens, Univ. Glasgow, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-8964Betula charcoal, 5, 13042.δ13C = −27.6‰250 ± 38
OxA-8965Betula charcoal, 5AS, 16032, δ13C = −26.7‰309 ± 39
OxA-8966Calluna charcoal, 14, 16065, δ13C = −25.4‰308 ± 36
OxA-8967Salix charcoal, 16, 16066, δ13C = −24.9‰8045 ± 55
OxA-8968Betula charcoal, 1, 16011, δ13C = −27.4‰349 ± 38
OxA-8969Betula charcoal, 5AS2, 16032, δ13C = −26.5‰323 ± 36
OxA-8970Betula charcoal, 4, 16053, δ13C = −26.4‰274 ± 39
OxA-8971Betula charcoal, 3, 16048, δ13C = −25.4‰263 ± 36
OxA-8972Betula charcoal, 8, 17024, δ13C = −26.0‰1344 ± 36
OxA-8973Corylus charcoal, 4, 17007, δ13C = −27.4‰5055 ± 45
OxA-9035Prunus sp. charcoal, 11, 10025, δ13C = −27.7‰329 ± 35

Comment (P. A.): eleven determinations from various sites near Ben Lawers, North of Loch Tay, demonstrated use of the area from the centuries around 7000 BC until the late medieval or post-medieval period, including an early Christian cemetery. There was clearly a large amount of disturbance, some no doubt due to use of turf as a building material in the post-medieval period, but some more difficult to explain.

Willow charcoal from the fill of a pit partly sealed by a phase 2 bank of a cellular turf structure was dated to between the last third of the eight and the first third of the seventh millennium BC.

Hazel charcoal from the upper fill of a small pit containing an AOC beaker and fragments of calcined bone was dated to the first quarter of the fourth millennium BC, about 1500 to 2000 years earlier than expected. The most likely explanation is that the hazel was residual, with an origin perhaps around the time that Neolithic pottery was in use in the vicinity.

Six determinations of from charcoal were associated with a cellular turf structure. Three came from fire spots, one from a hearth and two from a substantial occupation horizon containing early Neolithic pottery and worked quartz. All implied dates after the late fifteenth century AD; four of them suggested dates before the mid seventeenth century AD while one which might have been as late as the end of the eighteenth century AD and the other could even have been modern. The cellular structure is much more likely to have been built afresh in the Late Medieval period, and thereafter reused, than to have been a re-used Neolithic structure. It is an interesting speculation that without the radiocarbon determinations from the occupation layer the structure might have been interpreted as early Neolithic.

Blackthorn charcoal from an early hearth at a shieling hut was dated to between the late fifteenth and mid seventeenth centuries AD. Birch charcoal from the lower fill of a large deep posthole at the shieling hut dated to some time after the beginning of the sixteenth century AD.

A single determinations from flax from the bottom of a pit suggests that it was used for retting some time after 1660 AD.


A sample of charcoal from the site Berryhill, Oyne 2000, (NGR NJ 66792503), Aberdeenshire, Scotland, submitted by H. Murray, Hill of Belnagoak, Methlick, Ellon, Aberdeenshire, AB41 7JN, through Historic Scotland.

OxA-10386Ilex aquifolium charcoal, 304/1, δ13C = −24.1‰1953 ± 33

Comment (H. M.): the sample came from below a hilltop stone enclosure wall. It had appeared possible that this wall could have been an eighteenth century improvement enclosure, rather than part of the prehistoric settlement. The probable first century AD date of the charcoal, although based on a single entity sample, suggest a prehistoric date for the wall. With caution it might be used to view the settlement in the context of the first century Agricolan campaigns in north-east Scotland.

Bornais, South Uist

Samples of bone from Bornais, South Uist, (NGR NF 729302) Scotland, submitted by N. Sharpies, HISAR, Cardiff Univ., PO Box 909, Cardiff, CF1 3XU, through Historic Scotland. Comments by P. Ashmore.

Historic Scotland.  
OxA-9638cattle bone, BO99/8157/558, δ13C = −20.9‰1052 ± 34
OxA-9639cattle bone, BO99/8155/557, δ13C = −21.1‰1317 ± 39
OxA-9640cattle bone, BO99/2035/482, δ13C = −20.2‰1699 ± 34
OxA-9641cattle bone, BO99/8565/456, δ13C = −20.4‰1790 ± 40
OxA-9642sheep/goat bone, BO99/8152/182, δ13C = −20.4‰1011 ± 39
OxA-9643cattle bone, BO99/9170/495, δ13C = −20.7‰1686 ± 33
OxA-9665red deer bone, BO99/9169/495, δ13C = −21.6‰1705 ± 50
OxA-9677red deer bone, BO99/9105/453, δ13C = −21.4‰1580 ± 60
OxA-10273sheep bone, BO99/5854/214, δ13C = −21.0‰1065 ± 35
OxA-10274cattle bone, BO99/5896/215, δ13C = −21.4‰1004 ± 32
OxA-10275Hordeum sp, charred seed, BO99/5906/604, δ13C = −22.5‰880 ± 32
OxA-10276Avena sp, charred seed, BO99/5964/269, δ13C = −25.8‰537 ± 34
OxA-10277Avena sp, charred seed, BO99/5971/269, δ13C = −25.9‰521 ± 32
OxA-10278Avena sp, charred seed, BO99/8629/276, δ13C = −25.8‰563 ± 33
OxA-10279cattle bone, BO99/8707/675, δ13CC= −22.5‰863 ± 35
OxA-10291Avena sp. charred seed, BO99/5909/604, δ13C = −22.9‰580 ± 70
OxA-10292Avena sp. charred seed, BO99/8045/614, δ13C = −24.8‰590 ± 50
OxA-10304Avena sp. charred seed, BO99/8077/614, δ13C = −26.0‰660 ± 50
OxA-10305Avena sp. charred seed, B099/8633/276, δ13C = −24.2‰705 ± 50

Comment (P. A.): this settlement site included three mounds which appear to result from successive occupations ranging from before the mid first millennium AD to, probably, some time around the fourteenth century AD. It is extremely important because it provides much information about both the pre-Norse and Norse uses of the area. The site has many rich assemblages of artefacts which appear to represent, amongst other things, on-site manufacture. The economic information from the site is of considerable importance.

Three determinations of bones in middens at the edge of mound 1 suggest use from perhaps as early as the second but more likely at some time between the third and early fifth centuries AD. One of an arc of cattle metapodials surrounding a hearth associated with a rebuilt Late Iron Age structure in mound 1 was dated to between the mid third and early fifth centuries AD while the final phase of the stone-lined hearth dated to between the mid fourth and early seventh centuries cal AD.

Individual bones from secondary occupation of a large bow shaped house of Norse type provided two dates the earlier of which is from a residual bone of 650 to 780 cal AD. The other bone suggests occupation at 890 to 1030 cal AD while another bone from an abandonment deposit dates to between 900 and 1160 cal AD.

Two sand layers underlying the main house on Mound 3 and immediately above the floor of a house were dated respectively to between the late ninth and early eleventh and the late tenth and mid twelfth centuries AD. The three successive floors of the house produced three determinations from oat grains between the late thirteenth and mid fifteenth centuries AD and one earlier date, from a barley grain, between the early eleventh and mid thirteenth centuries AD. A grain drying kiln was possibly in use as early as the mid eleventh to mid thirteenth centuries AD but certainly was in use at some time between the early early and mid fifteenth centuries AD.

Calanais fields Project

Samples of peat from Calanais fields, Isle of Lewis (NGR NB 21253265), Scotland, submitted by C. Flitcroft, M. Johnson and G. Coles, Dept. Archaeol., Univ. Edinburgh, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

The Calanais Fields Project has completed two seasons of excavation, in 1999 and 2000, at Calanais Farm on the Isle of Lewis. The results of this fieldwork have demonstrated the presence of an extensive agricultural landscape buried beneath up to 2 m of peat, consisting of a variety of different features including field walls, clearance heaps and structures, and the widespread presence of a prehistoric land surface with fragments of rig-and-furrow surviving. This fragment of field system is likely to constitute part of an organised agricultural landscape that may extend over a much wider area, and is beginning to provide significant new information on the farming practices of prehistoric people in the islands.

OxA-10089peat, 5011, 20–21cm, δ13C = −27.9‰234 ± 34
OxA-10090peat, 5012, 59–60cm, δ13C = −27.2‰1628 ± 37
OxA-10091peat, 5009, 104–105cm, δ13C = −28.7‰2222 ± 37
OxA-10118peat, 14004, 20–2lcm, δ13C = −26.4‰1370 ± 40
OxA-10119peat, 14005, 34–35cm, δ13C = −28.2‰1805 ± 45
OxA-10120peat, 14007, 52–53cm, δ13C = −29.9‰2380 ± 40

Comment (P. A.): preliminary pollen analysis indicates a transition from cereal growing to mixed farming, to a final phase of reliance only on animal stocking, as the climate deteriorated and peat growth began, followed by the abandonment of the field system. The AMS determinations from peat columns west of the main stone setting at Calanais (Callanish) suggests that blanket peat did not start to form there until about some time in the second or third quarter of the first millennium BC, sealing underlying stone structure and fields.

This site will be significant for interpreting the inception of complex Atlantic roundhouses, and the reorganisation of the social and economic systems this entailed in the first millennium BC. The surviving peat appears to include a record stretching at least until after the middle of the first millennium AD.

Callanish Kerb Cairn

Samples of charred seeds from Callanish Kerb Cairn, (NGR NB 21793473) Scotland, submitted by T. Neighbour, Centre for Field Archaeology, Old High School, Infirmary Street, Edinburgh, EH1 1LT, through Historic Scotland.

OxA-9865Hordeum sp. charred seed, C.121b, δ13C = −22.5‰3310 ± 45
OxA-9866Hordeum sp. charred seed, C.129a, δ13C = −23.3‰3325 ± 40
OxA-9867Hordeum sp. charred seed, C.135a, δ13C = −23.4‰3295 ± 40
OxA-9868Hordeum sp. charred seed, C.135b, δ13C = −24.5‰3385 ± 45
OxA-9869Hordeum sp. charred seed, C.181a, δ13C = −21.5‰3355 ± 45
OxA-9870Hordeum sp. charred seed, C.181b, δ13C = −22.2‰3433 ± 39
OxA-9903Hordeum sp. charred seed, C.121a, δ13C = −24.1‰3418 ± 38
OxA-9931Hordeum sp. charred seed, C.129b, δ13C = −22.6‰4225 ± 85

Comment (T. N.): three of the eight charred barley ages from the Callanish kerb cairn suggest that the main period of kerb cairn construction dates to around the end of the first quarter of the second millennium BC, while two dates around the second quarter suggest that the site, if not the kerb cairn, remained in use for at least a few generations. A date in the first half of the third millennium BC from a grain in the central cist may suggest residual material from an earlier use of the site. The ages can be used to provide dates for the abundant pottery of the site and are of added interest given the large number of other funerary or ceremonial sites in the area, particularly since the cairn lies on the axis of the avenue of the main stone setting at Callanish a kilometre away.

Camas Daraich

Samples of charred seeds from the site Camas Daraich, Skye, Highland (NGR NG 56500050) Scotland, submitted by M. Cressey, Dept. Archaeol., Univ. Edinburgh, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9782Corylus avellana charred seed, Sample 5, δ13C = −24.2‰7670 ± 55
OxA-9783Corylus avellana charred seed, Sample 6, δ13C = −25.1‰7985 ± 50
OxA-9784Corylus avellana charred seed, CD 15(B), δ13C = −25.4‰7545 ± 55
OxA-9971Corylus avellana charred seed, CD 15(A), δ13C = −27.4‰7575 ± 75

Comment (P. A.): three determinations each from a hazel nutshell, from hearth deposits and a layer full of similar material, suggest that this part of the hunter-gatherer site belongs around the third quarter of the seventh millennium BC. They and another date from fire-blackened soil rich in lithics from around the first quarter of the seventh millennium suggest that the flint technology is a local variant rather than unusually early. Although this site is not part of the regional ‘First Settlers’ project, which is discussed below, it helps to build a regional picture of hunter-gatherer industries, subsistence and mobility in the coastal areas of the south-west Highlands.

Ceann Nan Clachan

Samples from the site Ceann Nan Clachan, Vallay Strand, N Uist, Western Isles, (NGR NF 771739) Scotland, submitted by A. Dunwell, Centre for Field Archaeology, Univ. Edinburgh, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9831Hordeum sp. charred seed, CNC SB C134, δ13C = −24.2‰2469 ± 37
OxA-9969Hordeum sp. charred seed, CNC SA C134, δ13C = −24.4‰2475 ± 50
OxA-9970Hordeum sp. charred seed, CNC SC C134, δ13C = −25.3‰2360 ± 50
OxA-9985pot residue, CNC SD C108, δ13C = −23.9‰2450 ± 34

Comment (P. A.): three determinations from a single grain of barley and a date from a pot residue suggest that a cellular building dates to between the early eighth and late fifth centuries BC. The dating is important because it shows that cellular buildings existed here in the early Iron Age.

Chapelfield, Cowie

Samples from the site Chapelfield, Cowie, Stirling, (NGR NS 8363 8951) Scotland, submitted by J. Atkinson, GUARD, Univ. Glasgow, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9233Hordeum vulgare charred seed, 437/694d, δ13C = −23.2‰136 ± 38
OxA-9234Corylus nutshell, 432/694d, δ13C = −24.6‰5085 ± 45
OxA-9235charred seed, 267/008, δ13C = −23.1‰214 ± 38
OxA-9298Corylus charcoal, 235/018, δ13C = −26.4‰7220 ± 80
OxA-9750Conlus seed, CE95 267/008, δ13C = −24.5‰5590 ± 55

Comment (P. A.): eight determinations from Chapelfield, Cowie had previously been obtained from bulk samples each consisting of charcoal from wood of a single species. It seemed possible, analysing the results, that the contexts contained residual or intrusive material or both. Barley in features with fifth and sixth millennium BC dates from charcoal required explanation.

Five single entity determinations were obtained from ORAU. Large oak lumps from Pit I had previously suggested a date in the first half of the fifth millennium BC, but ORAU determinations showed that a piece of hazel dated to the third quarter of the fifth millennium BC, and that a grain of barley dated to after 1500 AD. A pine date from Pit I I had previously suggested a dale around the middle of the sixth millennium BC but two ORAU results suggested that it contained hazel dated to the first quarter of the fifth millennium and that a grain of barley dated to after 1670 AD. A large piece of hazel charcoal from the stake-lined Pit V was dated to around the end of the seventh millennium BC.

There was activity in several different periods at Chapelfield. The ORAU results make it clear that the barley which might otherwise have been taken to be unusually early was fairly recent and intrusive, and suggested that oak charcoal dated previously had been from the interior of an old tree.

Cille Bhrea

Samples of bone from the site Cille Bhrea, Lemlair, Highland (NGR NH 57646148) Scotland, submitted by T. Rees, AOC Archaeology Group, The Schoolhouse, 4, Lochend Road, Leith, Edinburgh. EH6 8BR. through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9882human bone, 16/1, δ13C = −20.2‰220 ± 35
OxA-9883human bone, 27/1, δ13C = −20.5‰149 ± 38
OxA-9884human bone. 32/1, δ13C = −20.2‰235 ± 36
OxA-9885human bone, 37/1, δ13C = −20.6‰226 ± 37
OxA-9886human bone, 94/1, δ13C = −20.4‰149 ± 35
OxA-9887human bone. 102/1, δ13C = −20.9‰250 ± 35
OxA-9917human bone. 105/1, δ13C = −20.6‰246 ± 35
OxA-10010human bone. 63/1, δ13C = −20.9‰1230 ± 45

Comment (P. A.): seven of eight determinations from human hone from a cemetery in a non-rectilinear enclosure at a small chapel on the coast suggested that most of the skeletons dale to after 1500 AD. One burial was dated to between the late seventh and mid tenth centuries AD. All of the bones had δ13C values suggesting a wholly terrestrial diet.

Cleuchbrae, Annandale

Samples of charcoal and wood from the site Cleuchbrae, Annandale, Dumfries and Galloway (NGR NY 10359330) Scotland, submitted by J. Duncan, GUARD, Dept. Archaeol, Univ. Glasgow, through Historic-Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-8800Betula wood. 104, δ13C = −28.5‰2885 ± 45
OxA-8801Corylus charcoal, 17, δ13C = −26.5‰3115 ± 40
OxA-8832Quercus wood, 1012, δ13C = −26.9‰3055 ± 45
OxA-8833Corylus charcoal, 19, δ13C = −25.4‰3380 ± 40

Comment (P. A.): four determinations from a burnt mound with a wooden trough suggested activity of at least two different periods. Hazel charcoal from an early context suggested a date in the second quarter of the second millennium BC, the outer rings of the trough itself date to the third quarter or early in the last quarter while a nearby slender birch stake dates to the last quarter or slightly later and its date is significantly different from that of the trough. The burnt mound dates are similar to those of earlier burnt mounds elsewhere in Scotland and comparable to the earliest dates from platform settlements.

Cnip, Isle of Lewis

A sample of bone from the site Cnip, Isle of Lewis, Western Isles (NGR NB 099 364) Scotland, submitted by T. Neighbour, Centre for Field Archaeology, Old High School, Infirmary Street, Edinburgh, EH1 1LT, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9604human bone, burial A, δ13C = −17.0‰1332 ± 40

Comment (P. A.): this determination from an inhumation (burial A) in the machair has an enriched δ13C value (–17‰) suggesting that it incorporates a significant marine effect; it is thus at 640–780 cal AD probably earlier than it should be.

Coast Road, Longniddry

Samples of bone from the site Coast Road, Longniddry, East Lothian (NGR NT 44187709) Scotland, submitted by L. Baker, Headland Archaeology, 13 Jane Street, Edinburgh EH6 SHE, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-10034human bone, skeleton 4/017, δ13C = −20.2‰3274 ± 39
OxA-10088human bone, skeleton 2/009, δ13C = −20.4‰3152 ± 39

Comment (P. A.): two determinations from human bone, from a small cemetery of crouched burials, suggested that a crouched burial in a cist dated to between the early seventeenth and mid fifteenth centuries BC, while a crouched burial in a grave dated to between the late sixteenth and late fourteenth centuries BC.

Cramond, Edinburgh

Samples of seeds from the site Cramond, Edinburgh (NGR NT 18987697), Scotland, submitted by J. Lawson, City of Edinburgh Council, Archaeology Service, 10 Broughton Market, Edinburgh, EH3 6NU, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-10143hazelnut shell, CR95/74/1409, δ13C = −23.5‰9150 ± 45
OxA-10144hazelnut shell. CR95/283/1402 M, δ13C = −23.1‰9110 ± 60
OxA-10145hazelnut shell, CR95/291/1409, δ13C = −24.9‰9230 ± 50
OxA-10178hazelnut shell, CR95/956/1426 M, δ13C = −23.3‰9105 ± 65
OxA-10179hazelnut shell, CR95/958/1426 K, δ13C = −23.9‰9130 ± 65
OxA-10180hazelnut shell, CR95/1066/1431, δ13C = −26.0‰9250 ± 60

Comment (P. A.): six determinations each determined from single hazel nutshells showed that this is by a small margin the earliest known securely dated hunter-gatherer site in Scotland. The calendar dates suggest a short period of occupation of this coastal site, very likely to be earlier than 8300 BC and maybe as early as 8600 BC with activities including flint chipping and pit-digging. Comparable, albeit somewhat late, dates have been obtained from near Biggar and from fife Ness, with Kinloch on Rum probably slightly later.

Dunadd Environs, Kilmartin

Twelve plant macrofossil samples, monocotyledon fragments, leaf and twigs of bog myrtle and Betulaceae, and fruits of Carex rostrata, from a peat core in an abandoned river meander of the River Add. Situated about 500 m west of the early medieval defended royal centre of Dunadd in mid-Argyll, the c. 3.5 m of organic sedimentation provides a high resolution palaeoenvironmental record for the last two millennia in the Kilmartin Glen (NGR NR 83189363). Recovered in June 1999, the samples were submitted for dating in 2000 by J. J. Miller, Glasgow Univ. Archaeol. Res. Div. (GUARD) and R. A. Housley, Dept. Archaeol., Univ. Glasgow, through Historic Scotland.

OxA-9359core 3, 24–25cm, monocotyledon fragments, δ13C = −28.6‰831 ± 35
OxA-9405core 3, 56–57cm, monocotyledon fragments, δ13C = −28.1‰556 ± 36
OxA-9360core 3, 84–85cm, monocotyledon fragments, δ13C = −27.3‰945 ± 45
OxA-9598core 3, 104–105cm, Myrica gale twigs, δ13C = −27.6‰1255 ± 40
OxA-9599core 3, 132–133cm, monocotyledon fragments, δ13C = −25.4‰1260 ± 34
OxA-9600core 3, 164–165cm, monocotyledon fragments, δ13C = −25.8‰1525 ± 37
OxA-9601core 3, 188–189cm, monocotyledon fragments, δ13C = −21.0‰1642 ± 36
OxA-9406core 3, 204–205cm. Carex roatrala, δ13C = −27.0‰1807 ± 39
OxA-9381core 3, 228–229cm, δicotyledon fragments, δ13C = −27.6‰2125 ± 45
OxA-9382core 3, 268–269cm, monocotyledon fragments, δ13C = −28.0‰2380 ± 45
OxA-9407core 3, 300–301cm, Betulaceae leaf, δ13C = −27.5‰1940 ± 39
Ox A-10006core 3, 336–337cm, Betulaceae twig, δ13C = −28.2‰1850 ± 37

Comment (R. A. H.): the high sediment accumulation rate (6–8 calendar years per cm depth) of the former river channel provides the setting for a detailed pollen (core 2) and plant macrofossil (core 3) study (Housley et al, forthcoming) that focuses on human impact, land use and vegetation history in the last two millennia.

In general, once mineral deposition had ceased above 210 cm depth the rate of peat accumulation was relatively stable and the individual AMS determinations form a coherent chronological sequence. Problems with two inverted measurements (OxA-9381 and −9382) may reflect a degree of re-working associated with rapid mineral inwash before the final abandonment of the meander; the river current being responsible for transporting older re-worked organic material and alluvium from higher in the catchment. A similar dating inversion problem was observed at Buiston crannog (Crone 2000, 36–41), again associated with minerogenic sedimentation and possibly for the same reasons. A third problematic age measurement from the top of the sequence (OxA-9359) may be the result of recent activity on the sampling site, possibly animal trampling in that the sample came from a shallow depth from core 3. Disturbance from shallow drainage ditches is also visible on a 1948 aerial photograph. Given the soft condition of the meander surface deposits, some degree of physical disturbance could have occurred. However the pollen and plant macrofossil evidence that was taken from a separate monolith adjacent to the coring locality does not show disturbance, suggesting the extent of any such trampling was restricted.

The plant macrofossils show very localised vegetational changes associated with the hydrology reflecting, at least in part, the hydroseral succession but also the general fluvial history of the River Add and relative sea-levels. The pollen is more regional and shows that the Kilmartin Glen had already undergone considerable woodland clearance by the start of the first millennium AD. There is a notable decrease in woodland taxa, particularly alder, immediately prior to the start of the main occupation period of Dunadd (c. 600–900 AD) which is marked by minor increases in microscopic charcoal and a peak in the pollen of willow. Until the last two centuries land use indicators are overwhelmingly pastoral rather than arable, although this is not surprising given the locally wet conditions and the close proximity of the Moine Mhor (a large lowland raised mire). Although there are minor vegetational changes contemporary with the use of Dunadd as the inaugural site of the rulers of the Scottish kingdom of Dál Riata, the overall impression is one of land use continuity both during and after this period of royal use. In fact what appears to have had far more impact on the Kilmartin area was the opening up of the Highlands for commercial exploitation from the eighteenth century when major environmental changes are observable.

Fordhouse Barrow, Dun

Samples from Fordhouse Barrow, Dun, Angus (NGR NO 66586053), Scotland, submitted by E. Proudfoot, St. Andrews Heritage Services, 12 Wardlow Gardens, St. Andrews, fife. KY16 9DW, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-8222Quercus sp. charcoal, 2781, δ13C= −24.4‰5035 ± 40
OxA-8223Quercus sp. charcoal, 2815, δ13C = −24.9‰4920 ± 45
OxA-8224Quercus sp. charcoal. 2451, δ13C = −26.4‰4965 ± 40
OxA-8225charred hazelnut shell, 2952, δ13C = −23.1‰8100 ± 45
OxA-8226charred hazelnut shell. 2860, δ13C = −24.6‰5660 ± 40
OxA-8593human bone, 2942, δ13C = −25.5‰2165 ± 70
OxA-8777human bone, 3135, δ13C = −26.3‰2050 ± 110
OxA-8844human bone, 3060, δ13C = −26.6‰2010 ± 120
OxA-10057charred hazelnut shell, 2909/338, δ13C = −23.9‰7890 ± 50
OxA-10058charred hazelnut shell, 2950/330, δ13C = −25.1‰7920 ± 50
OxA-10059charred hazelnut shell, 2951/341, δ13C = −23.2‰8255 ± 55
OxA-10060charred hazelnut shell, 3049/602, δ13C = −23.5‰5565 ± 45

Comment (P. A.): twelve determinations from this important site which included a chambered burial mound and other structures relating to funerary activity showed that the site had been used many times over a period of about 5 to 6 millennia.

AMS results from hazel nutshell samples showed activity between the last half of the eighth millennium BC and the earlier third of the seventh millennium. However, a hazel nutshell in one of a series of Phase I pits with seven Arran pitchstone blades and a bifacial knife dated to the centuries either side of the middle of the fifth millennium BC suggested that this early archaeological phase contained material of different periods. This pre-cairn activity could be compared in date to that under a small long mound at Biggar Common in S Lanarkshire or to later activity under long cairns at Port Charlotte on Islay and Camster in Caithness, in that a major burial monument appears to have been built on an area previously used for purposes which are not clearly funerary.

Three determinations from oak planks used to build a structure in the phase 3B mound were in only slight disagreement with each other and suggest that it was built between 3800 and 3700 cal BC. This date suggests that Fordhouse is as early as any other well-dated burial mound in Scotland, and comparable in date to the early massive hall at Balbridie on Deeside. It contributes to a perception that current evidence is best interpreted as reflecting that in Scotland the first building of large structures, occurred a couple of centuries after 4000 BC rather than before 4000 BC. Less confidently, because of some earlier bulk charcoal dates for the use of pottery, it adds to the impression from single entity dates that the first widespread use of pottery also belongs about eight generations after 4000 BC.

A hazel nutshell fragment from the blocking of the chambered cairn passage was dated to the two centuries after the middle of the fifth millennium BC and was probably residual from activities contemporary with the Phase I pits.

Three determinations in the last few centuries BC were obtained from cremated human bone. However, one was in a complete collared urn in a cairn of phase 6 and the other two in contexts which also seemed likely to belong much earlier, and it seems certain that all three dates reflect contamination with later carbon. An AMS result subsequently obtained from the Groningen laboratory by the National Museums of Scotland, from the carbonate fraction of bio-apatite in cremated bone from the urn (Sheridan pers. comm.), suggests a date in the second quarter of the second millennium BC.

Galson, Isle of Lewis

Samples of bone from Galson, Isle of Lewis, Western Isles (NGR NB 436594), Scotland, submitted by T. Neighbour, Centre for field Archaeology, Old High School, Infirmary Street, Edinburgh, EH1 1LT, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-10164human bone, Gals II, δ13C = −20.1‰1895 ± 36
OxA-10165human bone, Gals III, δ13C = −19.8‰1848 ± 36
OxA-10166human bone. Gals IV, δ13C = −20.1‰1858 ± 34

Comment (P. A.): re-dating of human bone from three inhumations within a long cist cemetery provided results 136, 2137 and 136 14C years later than those originally obtained. In two cases they suggest burial in the first to early third centuries AD, and in the other burial between the first and early fourth centuries AD. The three new measurements are conformable with three others obtained earlier from other burials. The results are particularly interesting because they suggest that the burials in this long cist cemetery were as early as or earlier than burials in long cist cemeteries in eastern Scotland. They also support the possibility that a date from an inhumation at Buckquoy in Orkney may be similarly early, although the matter must remain open pending re-dating of the Buckquoy skeleton.

Garvald Burn

Samples of peat and charcoal from Garvald Burn, East Lothian (NGR NT 10154865) Scotland, submitted by C. Barrowman, Archaeol. Dept., Gregory Building, Lilybank Gardens, Univ. Glasgow, through Historic-Scotland.

OxA-10280peat, Monocot fragment GBOO/Monolith tin 3, δ13C = −27.8‰5000 ± 45
OxA-10449charcoal, Betula GBOO/CS4/402, δ13C = −24.4‰4865 ± 40

Comment (C. B.): the date from the charcoal which relates to the knapping floor fits in well with the original hypothesis of a late Mesolithic assemblage based on typological grounds from a cursory analysis of the assemblage. The date from the monolith tin collected from the peat bog immediately adjacent to the site, indicates that peat formation started around the same time as the late Mesolithic activity at Garvald Burn. Further environmental analysis from the peat sample will therefore give an indication of the natural environment during the late Mesolithic at Garvald Burn. Further dates are obviously necessary to confirm these results, and samples will be submitted post-excavation.

Glennan, Kilmartin

Sample of charcoal from the site Glennan, Kilmartin, Argyll and Bute (NGR NM 86220097) Scotland, submitted by G. MacGregor, GUARD, Univ. Glasgow, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

Ox A-10281Ericaceae charcoal, GUA 920.2/GLE001/004, δ13C = −24.9‰700 ± 33

Comment (P. A.): a single charred heather twig from sample, apparently part of the charcoal with a cremation deposit in a broken inverted urn in a small shelter in scree, produced a medieval date. Subsequent dating of a piece of charred hazel roundwood by another laboratory produced an equally curious result in the latter half of the fourth millennium BC. It is notable that if a date had been obtained from a bulked together sample of charcoal a ‘credible’ date for the pot might have been produced. The heather may reflect disturbance in the medieval period or later. Could the hazel date reflect use of ancient waterlogged wattle from a wetland deposit?

Home Farm, Castle Menzies

Samples of charcoal and charred seeds from Home Farm, Castle Menzies, Perth and Kinross (NGR NN 83054935), Scotland, submitted by S. Carter, Headland Archaeology, 13 Jane Street, Edinburgh EH6 SHE. through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9811Corylus sp. charcoal, HCM99 001/027, δ13C = −24.7‰4701 ± 38
OxA-9812charred barley seeds, HCM99 011/071, δ13C = −24.4‰1571 ± 36
OxA-9813Quercus sp. charcoal, HCM99 024/124, δ13C = −24.5‰5130 ± 40
OxA-9814Quercus sp. charcoal, HCM99 047/433, δ13C = −24.5‰5010 ± 40
OxA-9815Quercus sp. charcoal, HCM99 067/453, δ13C = −24.0‰7030 ± 45
OxA-9816Quercus sp. charcoal, HCM99 073/466, δ13C = −25.6‰5035 ± 70
OxA-9854charred barley seeds, HCM99 005/067, δ13C = −24.4‰2705 ± 50
OxA-9987Quercus sp. charcoal, HCM99 003/037, δ13C = −24.9‰5093 ± 39

Comment (P. A.): eight determinations from single pieces of charred material from pits and postholes suggested at least five different periods of activity. Four pits which did not form part of recognisable structures produced a variety of results, in the first quarter of the sixth millennium BC (charcoal), broadly around the middle of the fourth millennium BC (charcoal), the first quarter of the first millennium BC (barley) and the middle of the first millennium AD (barley). Four probable postholes forming parts of alignments and an arc of postholes produced broadly similar dates in the first quarter of the fifth millennium BC. Technically the difference between these latter overall is not statistically significant. The earliest age (from an alignment) and the latest (from the arc) are significantly different if compared to each other in isolation from the rest of the ages. Combining the three ages from the arc of postholes produces a weighted mean age which is not significantly different from that of the alignment.

Haddington Mains Cist

Sample from Haddington Mains Cist, East Lothian (NGR NT 35366751) Scotland, submitted by J. Lawson, City of Edinburgh Archaeology Service, 10 Broughton Market, Edinburgh, EH3 6NU, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-10254human bone, HAF99/005, δ13C = −21.0‰3945 ± 40

Comment (P. A.): a bone from an adult female crouched inhumation in a short cist, accompanied by a beaker and a shoulder-joint of pork, produced a calibrated result around the third quarter of the third millennium BC. One interest of the burial is the occurrence of the pork joint, as seen in several other short cist burials in the east and south-east of Scotland. Another is that the typology of the beaker would traditionally have suggested a considerably later date.

Holm of Papa Westray North

Samples from Holm of Papa Westray, Orkney Islands (NGR HY 50445228), Scotland, submitted by Anna Ritchie, 50/1 Spylaw Road, Edinburgh, EH10 5BL, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9752red deer antler, Trench V, layer 1, δ13C = −20.4‰4250 ± 45
OxA-9753sheep bone, Trench V, layer 2, δ13C = −18.8‰4225 ± 50
OxA-9832red deer antler, Trench 15, layer 1, δ13C = −20.7‰4235 ± 45
OxA-9833bone, sheep, Trench 15, layer 3, δ13C = −12.8‰4585 ± 40
OxA-9834bone, sheep, Trench IV, layer 1(2), δ13C = −14.6‰4440 ± 40
OxA-9871bone, otter, Trench 15, layer 2, δ13C = −11.3‰4680 ± 50
OxA-9872bone, red deer, Trench IV, layer 1(1), δ13C = −21.0‰3855 ± 45

Comment (P. A.): previously obtained determinations had suggested that burials took place in the tomb at some time during the last third of the fourth millennium or the first century of the third millennium BC.

Seven determinations were obtained, related to late use of the chambered tomb. Two of the three ages related to filling of the end chamber of the tomb were from animals which had a wholly or partially marine diet. The other date, from a red deer bone, suggests that the end chamber was filled up in the first third of the third millennium BC. A sheep bone from the forecourt shows a slight marine effect but ages from it and a piece of red deer antler both suggest a similar date for use of the forecourt. Similarly one of the two ages from a midden adjacent to the kerb of the back of the chambered cairn containing grooved ware and beaker was from a sheep which had a wholly or partially marine diet. The other age, from a red deer bone, suggests that the midden, which contained grooved ware and beaker, accumulated, at least in part, centuries later in the third quarter of the third millennium BC or slightly later.

These determinations from animal bones are of interest in showing closure of a stalled tomb at a date similar to those from animal bones in three tombs with broadly similar characteristics on another Orkney Island, Rousay (Renfrew 1979, 206). At Point of Cott, on Westray, animal bones from the tomb were dated consistently later than adult human bones. Intriguing questions about the late use of tombs arise from these observations.

Some of the animal bones produced δ13C measurements of particular interest for their implications about animal diet. An otter, with a δ13C value of −11.3‰, had perhaps unsurprisingly probably lived largely on marine fish and possibly also shellfish while one sheep from the end-cell, compartment 5 of the cairn, with a value of −12.8‰, and the juvenile sheep with a value of −14.6‰ from the midden, had probably suckled ewes which had a high component of sea weed in their diets. Two bones from red deer produced δ13C values implying a wholly terrestrial diet. The otter and sheep ages should be corrected for the marine effect but it is uncertain whether a proportion of Harkness's (1983) correction of 405 years can be applied, not only because of possible variations in the marine effect but also because it is based on comparisons between terrestrial material and shellfish, and shellfish have different metabolic pathways from fish, otters, seaweed and sheep.

Kelvin Valley

Geoarchaeological recording of valley floor sediment fills along the middle course of the River Kelvin, between Kirkintilloch and Kilsyth in central Scotland was undertaken in 1998 and 1999 to investigate the potential for the preservation of archaeological features beneath the floodplain surface, to advise on the development of mitigation strategies in future civil engineering works, and to identify the palaeoenvironmental value of the deposits, five detailed transects across the floodplain were made between Inchbelle Farm (NGR NS 666752) and Twechar (NGR NS 699759), from west to east: at Inchbelle Farm (NGR NS 666752), Bridgend Farm I (NGR NS 671752), Bridgend Farm II (NGR NS 673752), Twechar I (NGR NS 692759), Twechar II (NGR NS 695761).

A Slitz corer with chamber length of 1 m (internal diameter 6 cm) powered by a Cobra petrol-driven engine was used to extract most of the cores. Organic units within the deep alluvial fills were logged and critically evaluated as to context, in situ occurrence, stratigraphic significance, composition and organic content: 38 organic contexts logged were reduced to 12 secure chronological index-points. Samples for AMS 14C assays on bulk organic sediment were submitted by R. Tipping, Dept. Environ. Sci., Univ. Stirling, through Historic Scotland. Comments by P. Ashmore. Historic Scotland.

OxA-9808Inchbelle, very humified peat, 6a(i), c.34.50 m OD, δ13C = −27.5‰6315 ± 45
OxA-9986Inchbelle, very humified peat, 6a(ii), c.36.20 m OD, δ13C = −27.5‰8575 ± 55
OxA-9786Bridgend I, large % wood fragments and small number of seeds within humified material, 2b(iii), c. 33.80 m OD, δ13C = −28.4‰9540 ± 55
OxA-9787Bridgend I, large % wood fragments and small number of seeds within humified material, 2e(iii). c. 34.60 m OD, δ13C = −26.7‰8155 ± 55
OxA-9807Bridgend I, very humified peat, 2b(ii), c. 34.70 m OD, δ13C = −27.3‰5990 ± 45
OxA-9785Bridgend I, very humified peat. 2a(ii), c. 34.95 m OD, δ13C = −27.9‰4970 ± 45
OxA-9789Bridgend II, very humified peat, 4a(iv), c. 32.85 m OD, δ13C = −30.8‰10020 ± 60
OxA-9791Bridgend II, very humified peat. 4c(iii), c. 33.20 m OD, δ13C = −27.9‰8505 ± 50
OxA-9788Bridgend II, poorly humified peat with a few seeds. 4a(iii), c. 33.45 m OD, δ13C = −26.6‰7870 ± 55
OxA-9790Bridgend II, poorly humified peat, 4b(in), c. 33.90 m OD, δ13C = −27.6‰9890 ± 90
OxA-9809Twechar I, very humified peat. IIa, c. 33.10 m OD, δ13C = −27.7‰9325 ± 55
OxA-9810Twechar I, very humified peat with occasional wood fragments, IIa(iii), c. 34.50 m OD, δ13C = −27.0‰5775 ± 45

Comment (R. T.): four AMS determinations from Bridgend I showed initiation of valley-floor peat at various ages between about the last quarter of the tenth millennium BC or first third of the ninth millennium and the first third of the fourth millennium BC. At Bridgend II the earliest laterally extensive valley-floor peat was dated to the first three-quarters of the tenth millennium BC while a later phase of peat was dated to around the middle of the eight millennium. Laterally extensive minerogenie over-bank sediments were dated to the fist half of the seventh millennium BC. At Inchbelle, replacement of valley floor peat by mineral sediments indicating increased fluvial activity was dated to the second quarter of the eighth millennium BC and replacement of alluvial sediments by peat and a reduction in fluvial activity was dated to the last half of the sixth millennium BC. At Twechar I the onset of laterally extensive valley floor peat was dated to the middle third of the ninth millennium BC, or slightly earlier, and a reduction of minerogenic over-bank fluvial deposition was dated to the second quarter of the fifth millennium BC.

All the peats 14C dated in the Kelvin Valley are of early-mid Holocene age. Many peats near the bases of the sampled sediment sequences are dated to the first 1000 or so years after the Younger Dryas Sladial, following 14–15 m of incision from fluvioglacial terraces (OxA-9786, −9789, −9790, −9809). This is a very significant result because it shows that in this region peat accumulated on valley floors from the earliest Holocene, and was probably originally of great lateral extent. Above this wide early Holocene floodplain surface, the transects suggest that no significant net aggradation of sediment occurred in the early-mid Holocene. 14C samples at comparable depths have different ages and erosion may have more-or-less equalled sediment accumulation. The uppermost organic units are of mid-Holocene age (OxA-9808, −9807, −9785, −9810), and above these c. 2–2.5 m of fine-grained minerogenie sediments have accumulated, not noticeably faster than in the early Holocene.

The sequence shows that fluvial sediments just below the present floodplain surface have considerable antiquity, and are not recent deposits. They have a high potential for the preservation of archaeological features, although very few have been recovered in recent excavation.

Kilmore, Oban

Samples of wood from Kilmore, Oban, Argyll and Bute (NGR NM 88202528), Scotland submitted by J. C. Bonsall, Dept. Archueol., Univ. Edinburgh, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-8437Betula sp. wood, ObKIL/01, δ13C= −26.0‰1400 ± 40
OxA-8534Fraxinus excelsior wood, ObKIL/02, δ13C= −24.9‰1585 ± 45

Comment (P. A.): two determinations from ash and from birch showed that an otherwise completely undated timber platform belonged in the third quarter of the first millennium AD.

Knap of Howar

Samples from Knap of Howar, Papa Westray, Orkney Islands (NGR HY 48305180) Scotland, submitted by A. Ritchie, 50/1 Spylaw Road, Edinburgh, EH10 5BL, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9754sheep bone, House 1, layer 9, δ13C = −20.5‰4720 ± 50
OxA-9755sheep/goat bone, House 1, layer 16, δ13C = −18.8‰4630 ± 50
OxA-9756Caprinus bone, House 2, passage B layer 4, δ13C = −19.9‰4495 ± 50
OxA-9757cattle bone, House 2, layer 7, δ13C = −20.6‰4680 ± 50
OxA-9758Caprinus bone, House 2, layer 12, δ13C = −19.7‰4570 ± 50
OxA-9759sheep bone, Trench III, layer 3, δ13C = −18.9‰4800 ± 45
OxA-9760pig bone, Trench III, layer 4, δ13C = −20.1‰4750 ± 50
OxA-9761Caprinus bone, Trench V, layer 2, δ13C = −20.7‰4610 ± 50

Comments (P. A.): eight determinations were obtained from single bones. Two from the secondary floors of House 2 show that the material in the floor included residual bones of significantly older date than the youngest bones in the floor.

Theoretically, since all of the determinations have to be treated as termini post quern, the house could have been built and used at any time later than the radiocarbon ages imply, perhaps entirely within the period when grooved ware was in use at Barnhouse, Skara Brae 1 and Stenness (Ashmore 2000, figure 26.2). All that can be said is that there is absolutely no evidence from the new determination that it was. The most plausible interpretation of the new set of results is that Knap of Howar was built about 3500 BC and continued in use until some time between 3400 and 3100 BC.

Because the primary midden at Knap of Howar is older than the structure, even the early house contexts may contain bones which were old at the time the structure was built. A previous set of radiocarbon determinations (Birmingham and SRR) consisted of ages each from combined bone samples. Any of those analyses may thus represent a fairly meaningless average of the ages of bones of significantly different date. Also, errors were higher than realised at the time. An attempt to find new samples directly relatable to the old ones was not very successful. The overall impression from the new suite of determinations is that the old ones included much large random errors than those quoted but no significantly large systematic error.


Samples from the site Littleour, Perth and Kinross (NGR NO 17344024) Scotland, submitted by G. Barclay, Historic Scotland, Edinburgh. Comments by P. Ashmore, Historic Scotland.

OxA-8992pot residue, L23, Pot 3, find II, δ13C = −27.9‰4110 ± 55
OxA-8993pot residue, L23, Pot 6, find 33, δ13C = −26.6‰3845 ± 75
OxA-8994pot residue, L23, Pot 2, find 28+10, δ13C = −26.1‰3880 ± 55

Comment: (P. A.): three determinations were obtained from carbonised organic encrustation on three sherds of Grooved Ware from a pit possibly related to a timber enclosure previously dated from charcoal to the last third of the third millennium BC. Two of the new ages suggested dates in the last half of the third millennium agreeing well with the charcoal determination and suggesting that the pit may date to between about 2400 and 2150 BC to the nearest half century. The other however is statistically highly incompatible with the others, and suggests a date in the first half of the third millennium BC. There seem to be two possibilities. One is that the pit incorporated potsherds significantly older than the pit, and the other is that residues can pick up environmental carbon which will provide an age older than the true age of the residue. It is difficult to choose between these possibilities, because of the presence of the possibly related timber enclosure which (judging by a date from charcoal interpreted as the charred outer skin of an oak post) may date to the second half of the fourth millennium BC, and which is of a type associated with broadly similar pottery.

Loch Borralie, Durness

A sample of bone from Loch Borralie, Durness, Highland (NGR NN 37906761) Scotland, submitted by G. MacGregor, GUARD, Univ. Glasgow. Comments by P. Ashmore, Historic Scotland.

OxA-10253human bone, GUARD 918.2/006, δ13C = −20.2‰1931 ± 37

Comment (P. A.): an extended inhumation at a multi-phase cairn under sand was dated to between the second half of the first century AD and the beginning of the third.

Lundin Links

Samples of bone from the site Lundin Links Pictish Cemetery, fife (NGR NO 405023) Scotland, submitted by P. Ashmore, Historic Scotland, Longmore House, Salisbury Place, Edinburgh, EH9 1SH.

OxA-8895human bone, 1980–728 LL5, δ13C = −20.6‰1560 ± 40
OxA-8896human bone, 1973–871 LL6, δ13C = −20.4‰1455 ± 35
OxA-8897human bone, 1980–718 LL8, δ13C = −20.6‰1550 ± 35
OxA-8898human bone, 1980–721 LL9, δ13C = −21.1‰1600 ± 30
OxA-8899human bone, 1980–720 LL10, δ13C = −20.8‰1540 ± 35
OxA-8900human bone, 1980–725 LL11, δ13C = −20.6‰1565 ± 35
OxA-8901human bone, 1973–861 LL13, δ13C = −20.8‰1555 ± 35
OxA-8902human bone, 1980–726 LL14, δ13C = −21.0‰1465 ± 35
OxA-8903human bone. 1980–727 LL18, δ13C = −20.8‰1535 ± 35
OxA-8904human bone. 1973–870 LL19, δ13C = −20.3‰1610 ± 40

Comment (P. A.): ten determinations were obtained from burials in a cemetery discovered in sand dunes near the shore of the firth of Forth. Some were from cists under low round cairns, others from cists under low rectilinear cairns and some were from isolated cists. Earlier attempts to date the cemetery had not been wholly successful. They gave rise to speculations about possible Iron Age origins and possible late first millennium AD survival of the cemetery. Analysis suggests that no significant information is lost by ignoring all of the dates obtained before 1999. The new determinations do not show any pattern of difference in results between the various types of burial. They imply that the cemetery was in use both before and after AD 540 and probably for about five to ten generations. None of the δ13C values from burials from this cemetery suggest any marine effect. Numerous early accounts of discoveries of burials to the east of the cemetery suggest that there was a long cist cemetery nearby.

Meadows. Dornoch

Samples from the site Meadows, Dornoch. Highland (NGR NH 797895) Scotland, submitted by R. Coleman. SUAT Ltd., 55 South Methven Street, Perth, PH1 5NX, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9349charred barley seeds. DM01/97 03/17, δ13C = −24.8‰1139 ± 37
OxA-9350charred hazelnut shell, DM01/97 05/38, δ13C = −23.3‰113 ± 31
OxA-9351charred barley seeds, DM01/97 07/45, δ13C = −24.1‰1121 ± 35
OxA-9352charred barley seeds, DM01/97 08/53, δ13C = −23.6‰1129 ± 33
OxA-9353charred barley seeds, DM01/97 13/56, δ13C = −23.0‰1247 ± 32
OxA-9513charred barley seeds, DM01/97 10/58, δ13C = −22.9‰1055 ± 55

Comment (P. A.): six determinations were obtained from single hazelnut shells and grains of barley for a site including an enclosure containing a building and evidence for iron-working. One date gave a terminus post quern probably of the eighth or ninth century AD for the fill of a narrow ditch or gully cutting a pit full of slag and charcoal. Three other dates gave terminus post querns each in the ninth to tenth century for rough paving slabs marking the entrance into a building in an enclosure, a layer of metal-working debris which was cut by the enclosure ditch and the fill of a curvilinear ditch. A rectangular pit full of slag and charcoal and possibly hammer-scale, in an outer enclosure, contained a barley grain dated to between the late ninth century and the mid twelfth century AD and a possible anvil base in the enclosure contained a hazel nut shell of the fourteenth to mid fifteenth century. The site thus appears to have been long-lived and to have been used for metal-working both before and after it was enclosed.

Newark Bay, Deerness

Samples of human bone from Newark Bay, Deerness, Orkney Islands (NGR HY 5746 0413), Scotland, submitted by D. Brothwell, Dept. Archaeol., Univ. York, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-10407human bone, SK001, δ13C= −16.8‰1070 ± 36
OxA-10408human bone, SK004, δ13C= −18.4‰1068 ± 36
OxA-10409human bone, 99(6), δ13C= −16.8‰1015 ± 37
OxA-10410human bone, 69(87), δ13C= −14.2‰1123 ± 36

Comment (P. A.): four determinations were obtained from the cemetery associated with a chapel on the coast. The burial with the oldest 14C age had a δ13C value of about 14‰, suggesting a very high element of marine food in the diet. The two with intermediate 14C ages had δ13C values below 17‰ showing a marked element of marine food in the diet, and the fourth, with the youngest 14C age had about half as strong an element. The ages are therefore too old, but it is uncertain what proportion of Harkness's (1983) correction of 405 years can be applied, not only because of possible variations in the marine effect but also because it is based on comparisons between terrestrial material and shellfish, and shellfish have different metabolic pathways from fish and people. The bodies may all belong to the tenth to twelfth centuries, like the youngest one.


Sample of Patella vulgata shell Newbigging, Burntisland, fife (NGR NT 21378634), Scotland submitted by A. R. Rees, CFA, Old High School, 12 Infirmary Street, Edinburgh, EH1 1LT, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-8843Patella vulgata shell, AOC 1947/1, δ13C = 1.4‰600 ± 35

Comment (P. A.): a single AMS determination from a mussel shell showed that a midden scatter was post-Medieval in date.

Parks of Garden

Samples from Parks of Garden, Stirling (NGR NS 60389682), Scotland, submitted by C. Ellis, AOC Archaeology Group, The Schoolhouse, 4, Lochend Road, Edinburgh, EH6 8BR, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9289Alnus glutinosa peat, 1/173, δ13C= −27.5‰5153 ± 40
OxA-9613Quercus sp., wood, AOC 3166 W67 F149, δ13C= −26.1‰5080 ± 40
OxA-9751Querent sp., wood, AOC 3166 W38 F106, δ13C = −23.3‰4475 ± 45

Comment (P. A.): nine radiocarbon determinations were obtained to improve understanding of a timber platform on the margin of Flanders Moss (see also Arch. List 28). Three results suggest that wood found in peat under the platform died in the first quarter of the fourth millennium BC, including an oak which may have fallen naturally. However, peat near the surface on which the platform was built dated to after the middle of the fourth millennium. Five determinations suggest that the platform itself and a large timber lying outside but overlapping it dated to the last third of the fourth millennium or the first century of the third.

Raschoille, Oban

Samples of various materials from Raschoille, Oban, Argyll and Bute (NGR NM 85472888), Scotland, submitted by J. C. Bonsall, Dept. Archaeol., Univ. Edinburgh, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-8396Cervus elaphus bone, OhRC/03, δ13C = −21.8‰7640 ± 80
OxA-8839Cerasloderma edule shell, ObRC/07a+07b, δ13C =3.0‰7580 ± 45
OxA-8397Cervus elaphus bone. ObRC/04, δ13C = −21.5‰7575 ± 75
OxA-8395Lynx lynx bone, ObRC/01, δ13C= −19.9‰7495 ± 50
OxA-8398Cervus elaphus bone, bevel ended tool, ObRC/05, δ13C = −21.6‰7480 ± 75
OxA-8501Cerastoderma edule shell. ObRC707a+07h, δ13C = 1.4‰7390 ± 55
OxA-8840Cerasloderma edule shell, ObRC/07a+07b, δ13C =2.6‰7300 ± 50
OxA-8535Cervus elaphus bone, bevel ended tool, ObRC/06, δ13C = −21.4‰7265 ± 80
OxA-8439charred Corylus avellana shell, ObRC/09, δ13C = −25.1‰7250 ± 55
OxA-8538Cervus elaphus bone, ObRC/02, δ13C = −22.1‰6460 ± 180
OxA-8438Corylus avellana charcoal, ObRC/08, δ13C = −26.3‰5115 ± 55
OxA-8440charred Corylus avellana shell, ObRC/10, δ13C = −21.8‰4995 ± 45
OxA-8432human bone, ObRC/18, δ13C = −20.4‰4980 ± 50
OxA-8431human bone, ObRC/16, δ13C = −20.6‰4930 ± 50
OxA-8433human bone, ObRC/19, δ13C = −20.2‰4920 ± 50
OxA-8441human bone, ObRC/21, δ13C = −21.2‰4900 ± 45
OxA-8442human bone. ObRC/22, δ13C = −21.0‰4890 ± 45
OxA-8536Corylus(?) charcoal. ObRC/11, δ13C = −27.2‰4880 ± 60
OxA-8404human bone. ObRC/17, δ13C = −21.6‰4850 ± 70
OxA-8443human bone, ObRC/23, δ13C = −20.4‰4825 ± 55
OxA-8434human bone, ObRC/20, δ13C = −21.1‰4720 ± 50
OxA-8444human bone, ObRC/24, δ13C = −21.1‰4715 ± 45
OxA-8435human bone. ObRC/25, δ13C = −22.5‰4680 ± 50
OxA-8400human bone, ObRC/14, δ13C = −20.3‰4640 ± 65
OxA-8399human bone, ObRC/12, δ13C = −21.4‰4630 ± 65
OxA-8401human bone, ObRC/15, δ13C = −21.1‰4565 ± 65
OxA-8537human bone, ObRC/13, δ13C = −21.8‰4535 ± 50

Comment (P. A.): the site is a cave close to the head of what would have been a marine bay in the earlier part of the Holocene. Twenty-seven determinations were obtained, three from marine shell, three from hazel. 14 from human bones, four from animal bones and two from bevel-ended tools. The deposits were roughly divided into lower and upper. When ordered from oldest to youngest the first 12 determinations came from the lowest deposits. All of these except the latest two dated to various ages between the last half of the seventh millennium and 5000 BC.

The three earliest results, excluding the cockle shell (Cerastoderma edule), are from animal bones (two from red deer and one from a lynx) and are not significantly ditferent from each other. They could all date to the third quarter of the seventh millennium BC. The next two, again excluding the Cerastoderma edule shells, are from the two bevel-ended tools. The difference between their ages is just significant, and they could both belong in the last third of the seventh millennium BC though one could be several decades earlier and the other several decades later. Their dates suggest that they are about a millennium younger than the oldest known Scottish examples. The next six determinations are all from hazel or cockle shell. The strength of the marine offset at this time is not known, but even if 400 14C years were subtracted from the cockle shell determinations they would still be earlier than the last two of these six (OxA-8438 and −8440). They were not significantly different from each other, and may both have dated to the first quarter of the fifth millennium BC.

The next I I determinations in order of descending radiocarbon age were all from human burials from upper deposits with the exception of a hazel twig from the lower deposits dating to the second quarter of the fourth millennium or slightly earlier. The human burials cannot all have been deposited at the same time as each other and indeed they almost give the impression of being successive burials over the period after the first quarter and before the last third of the fourth millennium BC.

The latest four determinations are also from human bones, three from the lower deposits and one from the upper. The two older bones probably come from the middle half of the fourth millennium and the two late from the second half (the plateau in the calibration curve between about 3400 and 3100 BC extends their calibrated ranges to the last century of the fourth millennium). Overall, the burials appear to be contemporary with chambered tombs.

Red Castle

Samples of human tooth and bone from Red Castle, Lunan Bay, Angus (NGR NO 68785085), Scotland submitted by D. Alexander, Centre for Field Archaeology, Old High School, 12 Infirmary Street, Edinburgh EH1 1LT, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-10162human bone, DS3 SB2 C087, δ13C = −20.5‰1426 ± 36
OxA-10163human bone, DS6 G030 C068, δ13C = −21.1‰1661 ± 36
OxA-10167human tooth, DS7 G030 C068, δ13C = −20.7‰1544 ± 36

Comment (P. A.): the site was a cemetery containing cropmark square and round barrows with cist burials under them and also cist burials without any evidence for covering mounds. Initially eight dates were obtained from human bone and teeth (see Arch. List 28) and then three more determinations were obtained to check the results.

The burial under round barrow 1 was dated to between the last quarter of the first century AD and the middle of the fourth century AD. That under round barrow 2 was dated to between the mid fifth century and the mid seventh century AD

Two determinations of bone from the burial under square barrow 1 were virtually identical and suggest that the burial dates to between the second quarter of the fifth century and the middle of the sixth century AD. The burial under square barrow 2 produced a significantly later date and a check date produced a virtually identical result suggesting that the burial dates to between the mid sixth century and the mid seventh century, and more likely the first half of the seventh century than earlier. The burial under square barrow 3 was dated to the seventh century AD.

A cist burial without any evidence for a covering mound initially produced conflicting results. A tooth had a significantly older age than a femur. Two fresh ages were obtained. One of the new ages, from a tooth (OxA-10167), may have been contaminated with younger material judging by stable isotope measurements. The other age, from a bone (OxA-10163), was very similar to the previous date from a tooth (OxA-8413), and they suggest that the burial dates to between the second half of the third and the middle of the fifth centuries AD. However, it is difficult to know which dates to trust. Since tooth enamel is suspected of producing aberrant dates the burial is possibly best dated by the second femur date (OxA-10163), for which stable isotope ratios were optimised, which suggests that it belongs between the mid third and mid sixth centuries AD.

The results from Red Castle suggest that it is sensible to obtain more than one date from burials with poorly preserved bone and that stable isotope ratios must be used to control the extraction of the fraction used for dating.

None of the δ13C values from burials from this cemetery suggest any marine effect.

Scotland's First Settlers project

The dates from Sand, Crowlin 1, Loch A Squir and Ashaig middens in SE Skye and the facing mainland are important not only in dating the sites and helping understanding of the stratigraphy (which had been expected to include redeposition at Sand and at Loch A Squir), but also in showing that the sites do not represent a fifth millennium increase in local populations which, it had been hypothesised, might have adopted a more sedentary way of life at about that time, perhaps leading to adoption of agriculture. Samples from Crowlin 1, Lochalsh, (NGR NG 691338), Loch A Squir, Raasay, (NGR NG 60845286), Ashaig 1, Lochalsh (NGR NG 2420) and Sand, Lochalsh, (NGR NG 68414934), Highland. Scotland, submitted by M. Cressey, Dept. Archaeol., Univ. Edinburgh, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

Ashaig 1

OxA-9277Betula sp. charcoal, SFS99 1/12, δ13C = −25.6‰769 ± 36
OxA-9278Corylus avellana charcoal. SFS99 1/4, δ13C = −26.4‰771 ± 32
OxA-9279Betula sp. charcoal, SFS99 1/6, δ13C = −27.1‰723 ± 33

Comment (P. A.): three pieces of charcoal were dated. They all probably belong to the thirteenth century (one possibly the fourteenth century) AD.

Crowlin 1  
OxA-9250Betula sp. charcoal, SFS99 3/4, δ13C = −27.7‰1296 ± 39
OxA-9251Betula sp. charcoal, SFS99 1/11, δ13C = −26.6‰99 ± 37
OxA-9252Betula sp. charcoal, SFS99 1/6, δ13C = −26.4‰477 ± 35
OxA-9253deer bone, SFS99 N11, δ13C = −20.9‰316 ± 39

Comment (P. A.): three determinations from single pieces of birch charcoal suggest early and middle first millennium AD and also fifteenth century AD activity. A result from a limpet scoop implies it dates to the fifteenth century AD. These samples had been expected to date to before about 4000 BC.

Loch A Squir  
OxA-9254Betula sp. charcoal, SFS99 1/6, δ13C = −26.5‰2055 ± 39
OxA-9255deer bone, SFS99 N25, δ13C = −21.6‰7245 ± 55
OxA-9305Betula sp. charcoal, SFS99 1/3, δ13C = −26.6‰7620 ± 75

Comment (P. A.): two of three determinations from this midden suggest seventh millennium activity at the site while the third from the lowest level suggests activity in the second or first century BC. It seems likely that the upper midden has been redeposited.

OxA-9280antler, SFS99 9/8, δ13C = −21.8‰7520 ± 50
OxA-9281deer bone, SFS99 N19, δ13C = −21.3‰7715 ± 55
OxA-9282deer bone, SFS99 N18, δ13C = −20.8‰7545 ± 50
OxA-9343Betula sp. charcoal, SFS99 9/8, δ13C = −24.6‰7765 ± 50
OxA-10152mammal bone, B24A NE spit 8/013/N62, δ13C = −22.1‰8470 ± 90
OxA-10175mammal bone, B24B NE spit 7/013/N60, δ13C = −21.1‰7825 ± 55
OxA-10176mammal bone, A1B NE spit 9/022, δ13C = −20.9‰6605 ± 50
OxA-10177mammal bone, A2B SW spit 10/022, δ13C = −21.8‰6485 ± 55
OxA-10384mammal bone, 4A NE spit 4/013/N70, δ13C = −21.1‰7855 ± 60

Comment (P. A.): seven bevel-ended tools were dated. Three came from a loose unconsolidated limpet midden overlying a rockfall and covered by crushed shell and turf. One tool dates from the second or third quarter of the eighth millennium BC. Two tools date from the first half of the seventh millennium BC. Two tools from a shell-free organic midden overlying a sterile palaeosol and covered by the main shell midden are younger; the agreement between their ages is not very good but they suggest the midden dates to around the middle of the sixth millennium BC. Another tool from the midden dates to the third quarter of the seventh millennium. The seventh was dated to the centuries around the middle of the seventh millennium, while an antler and a birch sample from the same part of the midden dated to the centuries before and those after the middle of the seventh millennium.

Scottish bloomeries

Ten determinations were obtained for bloomeries and associated remains: Cladh Nan Sassunach, Highland (NGR NH 00706595), Fasagh Ironworks, Highland (NGR NH 011654), and Letterewe Ironworks, Loch Maree, Highland (NGR NG 958707), Allt Na Ceardaich, Argyll and Bute (NGR NS 142927) and Tamheich Burn, Argyll and Bute (NGR NS 03108225). Samples submitted by J. A. Atkinson, GUARD, Univ. Glasgow, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

Cladh Nan Sassunach

OxA-9322Pinus sylvestris charcoal, SBP/CS96/003/3008, δ13C= −23.8‰428 ± 33

Comment (P. A.): pine charcoal from a thick band of carbonised material in the coffin chamber of a grave in a graveyard, Cladh Nan Sassunach, by Fasagh ironworks was dated to between the second quarter of the fifteenth century and the first quarter of the seventeenth century AD.

Fasagh ironworks  
OxA-9283Pinus sylvestris charcoal, SBP/FA96/141/088, δ13C = −25.3‰718 ± 36
OxA-9284Pinus sylvestris charcoal, SBP/FA96/123/091, δ13C = −25.3‰494 ± 32
OxA-9320Betula sp. charcoal, SBP/FA96/001/055, δ13C = −25.1‰539 ± 35
OxA-9321Alnus charcoal, SBP/FA96/019/TPB2, δ13C = −28.5‰337 ± 33

Comment (P. A.): four 14C determinations were obtained. Pine charcoal (OxA-9283) from a band of silt (088) deposited as makeup in the central forge area of Fasagh ironworks was dated to the thirteenth or fourteenth centuries AD. Birch charcoal from the fill of a posthole below the central forge area and pine charcoal from a floor layer in the central forge area of Fasagh ironworks were dated to the fourteenth century or the first half of the fifteenth. Alder charcoal from a final dumping layer of the northern slag mound of Fasagh ironworks was dated between the last quarter of the fifteenth century and a decade before the middle of the seventeenth century AD.

The dates from these four sites demonstrate a flourishing later medieval primary iron-working industry in western Scotland. Their location presumably reflects the abundance of woodlands available for charcoal manufacture.

Letterewe ironworks

OxA-9346Betula sp. charcoal, SBP/LA98/011/103/104, δ13C = −26.1‰437 ± 32

Comment (P. A.): a single piece of birch charcoal (sample 011) from a thick band of charcoal and silt in a charcoal store south of the early blast furnace was dated to between the fifteenth and early seventeenth centuries AD.

Allt Na Ceardaich  
OxA-9324Alnus charcoal, SBP/AC96/028/020, δ13C = −26.5‰452 ± 35
OxA-9344Betula sp. charcoal, SBP/AC96/003/013, δ13C = −26.3‰652 ± 34
OxA-9345Betula sp. charcoal, SBP/AC96/10/50, δ13C = −27.9‰680 ± 31

Comment (P. A.): three determinations were obtained. Birch charcoal from the basal layer of the southern slag heap at the large bloomery furnace and another piece from a layer of material in a charcoal store at a large bloomery furnace were both dated to between the last quarter of the thirteenth and the end of the fourteenth centuries AD. Alder charcoal from a thin band of smithing waste from a hearth was dated to the fifteenth century.

Tamheich Burn

OxA-9323Alnus charcoal, SBP/BP95/005/1014, δ13C = −26.8‰410 ± 32

Comment (P. A.): alder charcoal (sample 005) from the basal fill (1014) of a bloomery furnace was dated to between the second third of the fifteenth century and the first third of the seventeenth centuries AD.


Samples of bone and seeds from Silgenach, Western Isles (NGR NF725288), Scotland submitted by N. Sharpies, HISAR, Cardiff Univ., through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-8880cattle bone, 1656, δ13C = −20.4‰2385 ± 40
OxA-8881cattle bone, 1660, δ13C = −20.5‰2485 ± 35
OxA-8905cattle bone, 1396, δ13C = −21.1‰3875 ± 35
OxA-8920cattle bone, 1318, δ13C = −20.9‰3710 ± 45
OxA-8921cattle hone, 1344, δ13C = −22.4‰3665 ± 45
OxA-8922cattle bone, 1632, δ13C = −20.4‰2485 ± 40
OxA-8923cattle bone, 1642, δ13C = −20.6‰2410 ± 40
OxA-8924red deer bone, 1648, δ13C = −21.2‰2340 ± 55
OxA-8925barley seed, 9002, δ13C = −23.3‰3655 ± 45
OxA-8926barley seed, 9021, δ13C = −23.0‰3490 ± 40
OxA-8927barley seed, 9023, δ13C = −23.1‰3520 ± 50
OxA-8928barley seed, 9031, δ13C = −23.6‰3715 ± 45
OxA-8989sheep bone, 1615, δ13C = −21.3‰3565 ± 70
OxA-9006cattle bone, 1392, δ13C = −20.7‰3665 ± 45

Comment (P. A.): fourteen determinations for single animal bones and barley grains from an area of deflating machair sites suggested dates for cultivation, occupation and butchering of animals. A cattle metapodial from soil overlying ardmarks of the earliest cultivation level on this part of the machair suggested a terminus post quern of around the last quarter of the third millennium BC or slightly later. A barley grain provides a terminus post quern for soil overlying ardmarks of the second cultivation level on this part of the machair in the quarter millennium centring on 2000 BC, although a sheep bone from the same layer was much older and presumably residual. Judging by a date from a barley grain, and provided the grain is not residual, an occupation layer with an animal butchery deposit may be contemporary with either of these two episodes of agriculture. Ages broadly of the last quarter of the third millennium BC or the first century of the second millennium BC, were obtained from a bone in a sandy occupation layer including pottery of Early Bronze Age style and from a bone in a sand deposit, probably of butchery waste. Another butchery layer was dated to the last century and a half of the third millennium or the first third of the second millennium.

Two barley grain determinations from pits with pottery of Late Bronze Age or Early Iron Age style produced surprisingly early results both in the first third of the second millennium BC. The grains may have been residual. The last five ages, all from butchery layers or sandy layers with animal bones in them were not significantly different from each other as a group. The earliest of these ages from butchery waste came from a bone overlying the floor of a house belonged between the early eight and late fifth centuries BC. A brown sand of identical age contained pottery of Late Bronze Age style. The youngest age was from a red deer bone from a dump of animal bones which may have been as early as the start of the eight century BC or as late as the end of the third century BC.

Thus, the area seems to have been used sporadically or even possibly for fairly long spans of time for a variety of purposes from the centuries around the beginning of the second millennium to after the eighth century BC, a period of over 1000 years.

Smoo Caves

Samples of charcoal from Smoo Caves, Highland (NGR NC 41856730), Scotland, submitted by A. Pollard. GUARD, Dept. Archaeol., Univ. Glasgow, through Historic Scotland. Comments by P. Ashmore. Historic Scotland.

OxA-8210Betula/Salix charcoal. GKC Column Spit 2, δ13C = −25.7‰1030 ± 40
OxA-8211mixed charcoal. GKC Column Spit 15, δ13C = −27.1‰1160 ± 35
OxA-8212mixed charcoal. GKC Column Spit 33, δ13C = −25.7‰1120 ± 30

Comment (P. A.): threedeterminations were obtained from charcoal in laminated middens in the smaller of two sea-washed caves on the northern coast of Sutherland. Pottery sherds and other artefacts were found at various levels and suggest Late Norse and post-Norse activity in the upper levels and possible Iron Age activity in the lower levels. The earliest date falls between the last quarter of the eighth century AD and the third quarter of the tenth century and the latest suggests occupation between the tenth and middle third of the twelfth centuries AD.

Solway Phase 3

Sample of oak wood from Solway Phase 3, Dumfries and Galloway (NGR NY 164653), Scotland, submitted by M. Cressey, Dept. Archaeol., Univ. Edinburgh. Comments by P. Ashmore, Historic Scotland.

OxA-8959Quenus sp. wood, BKI, δ13C = −25.‰7020 ± 50

Comment (P. A.): oak wood from a submerged forest was dated to the first quarter of the sixth millennium BC.

Stones of Stenness

Samples of bone from Stones of Stenness, Orkney (NCR HY 30671252), Scotland submitted by J. N. G. and A. Ritchie, 50/1 Spylaw Road, Edinburgh EH10 5BL, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9762Canis lupus bone, basal ditch-fill 1 B16, δ13C = −20.4‰4240 ± 45
OxA-9763cattle bone, basal ditch-fill 2 B17, δ13C = −21.1‰4425 ± 50
OxA-9764cattle bone, basal ditch-fill 4, δ13C = −21.1‰4390 ± 50
OxA-9765cattle bone, basal ditch-fill 5 B13, δ13C = −21.1‰4405 ± 50
OxA-9904cattle bone, basal ditch-fill 3 B25, δ13C = −21.0‰4360 ± 40

Comment (P. A.): four samples from cattle bones and one from a wolf bone in the organic basal ditch fills of the henge ditch surrounding the standing stones were dated. Had the cattle bones been exactly contemporary with one another they would have implied a date in a period a century either side of 3000 BC. However, it is not possible to prove that the bones were exactly contemporary with one another, because the wolf bone is significantly later, so their ages cannot be combined. They therefore show only that the ditch fill started to accumulate between about 3300 and 2900 BC. The wolf bone was processed with the others, and its activity was measured with two standards and a background sample which all produced normal results. Its yield was good and its C/N ratio suggests that it was not contaminated with carbon of a different age from that of the bone. Archaeological explanations must be sought for the difference; the likeliest is that the ditch remained open for (at least) many decades rather than that the miscellaneous cattle bones had been curated for at least a few generations.

Thorneybank cemetery

Samples of human bone from Thorneybank cemetery, Midlothian (NGR NT 34806890), Scotland, submitted by A. R. Rees, CFA, Old High School, 12 Infirmary Street, Edinburgh EH1 1LT, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-8152human bone, DS1 G1/C238, δ13C = −20.5‰1705 ± 45
OxA-8153human bone, DS2 G2/C249, δ13C = −20.5‰1600 ± 40
OxA-8154human bone, DS5 G5, δ13C = −20.6‰1590 ± 40
OxA-8155human bone, DS9 G14/C491, δ13C = −20.2‰1255 ± 40
OxA-8186human bone, DS7 G10/C490, δ13C = −20.0‰1595 ± 35
OxA-8187human bone, DS12 G45/C068, δ13C = −20.7‰1410 ± 35
OxA-8188human bone, DS15 G34/C064, δ13C = −20.5‰1460 ± 40
OxA-8189human bone, DS19 G48/C109, δ13C = −20.6‰1430 ± 40
OxA-8190human bone, DS25 G60/C181, δ13C = −20.6‰1480 ± 35
OxA-8191human tooth, DS26 G67/C226, δ13C = −20.7‰1395 ± 35
OxA-8192human bone, DS27 G68/C217, δ13C = −20.4‰1610 ± 35
OxA-8193human bone, DS28 G72/C133, δ13C C= −20.7‰1600 ± 40
OxA-8194human bone, DS29 G73/C144, δ13C = −20.7‰1485 ± 35
OxA-8201Betula sp. charcoal, S119 C303, δ13C = −24.9‰3625 ± 40
OxA-8653human bone, Gr9 c.489 (DS6), δ13C = −20.1‰1525 ± 30
OxA-8654human bone, Gr46 c.097 (DS18), δ13C = −19.9‰1550 ± 35
OxA-8655human bone, Gr90 (DS32), δ13C = −20.4‰1560 ± 30
OxA-8664human bone, Gr3 c.208 (DS3), δ13C = −20.1‰1560 ± 35
OxA-8665human bone, Gr4 c.201 (DS4), δ13C = −20.4‰1490 ± 40
OxA-8666human bone, Gr52 c.130 (DS20), δ13C = −20.1‰1475 ± 35
OxA-8667human bone, Gr53 c.116 (DS21), δ13C = −20.1‰1495 ± 40
OxA-8668human bone, Gr99 (DS33), δ13C = −20.4‰1650 ± 40
OxA-8690human bone, Gr106 (DS34), δ13C = −20.6‰1510 ± 35
OxA-8723human bone. Gr23 c.052 (DS10), δ13C = −20.1‰1425 ± 55
OxA-8724human bone, Gr28 (DS12), δ13C = −20.4‰1500 ± 45
OxA-8787human bone, Gr37 c.097 (DS16), δ13C = −20.4‰1555 ± 40
OxA-8788human bone, Gr54 c.119 (DS22), δ13C = −20.5‰1545 ± 35
OxA-8789human bone, Gr56 c.158 (DS23),b, δ13C = −20.3‰1610 ± 40
OxA-8935human bone, Gr32 c.044 (DS14), δ13C = −20.6‰1599 ± 38
OxA-10160human bone, DS31 G27, δ13C = −20.3‰1581 ± 36
OxA-10161human bone, Gr58 c.183 (DS24), δ13C = −20.4‰1578 ± 34

Comment (P. A.): thirty-one determinations were obtained from a large long cist and grave cemetery. All were from human bone (and one tooth), except for one from birch roundwood charcoal from a large pit with fire cracked stones. This was dated to between the last century and a half of the third millennium BC and the first quarter of the nineteenth century BC.

The determinations were obtained mostly in two batches. The two oldest determinations obtained were checked and replaced. The origins of the cemetery may lie as early as the fourth or even the third century BC but it remains possible that the remaining two oldest dates (after re-dating of the originally oldest two burials) are ‘too early’. Certainly, the vast majority of the dates are in the fifth and sixth centuries AD though it is conceivable that the two latest might be early seventh century AD.

West Water reservoir

Samples of charcoal from West Water reservoir, Scottish Borders (NGR NT 116524), Scotland, submitted by F. Hunter, Dept. Archaeol., National Museum of Scotland, Edinburgh, through Historic Scotland. Comments by P. Ashmore, Historic Scotland.

OxA-9547Corylus avellana charcoal, WWR 94 / 013A, δ13C= −26.2‰3299 ± 35
OxA-9548Sorbus sp. charcoal, WWR 94/013B, δ13C= −25.7‰3298 ± 35

Comment (P. A.): two AMS determinations, one from a piece of rowan and another from a piece of hazel, were obtained from a pit containing charcoal and heat-cracked stones at a cist cemetery partially submerged by a reservoir. They were almost identical and suggest that the pit dates from the seventeenth or sixteenth centuries BC. The nearby cist cemetery included a burial accompanied by either a two-strand necklace or two separate necklaces. One included 31 cylindrical lead beads and seem to provide the earliest evidence for the use of metallic lead in Britain.


Great Britain

Star Carr

Samples from Star Carr, N. Yorkshire (NGR TA 028809), submitted by P. Mellars, Dept. Archaeol., Downing Street, Cambridge, CB2 3DZ, through the NERC-funded ORADS facility.

OxA-3342plant remains, ST C13a/10–15, δ13C = −29.2‰9390 ± 70
OxA-3343plant remains, ST C12a/60–65, δ13C = −28.4‰9420 ± 70
OxA-3344plant remains, ST C11a/115–120, δ13C = −28.1‰9360 ± 70
OxA-3345charcoal, ST C10/170–175, δ13C = −26.3‰9580 ± 70
OxA-3346wood. ST C8/210–220, δ13C = −27.3‰9560 ± 70
OxA-3347wood. ST C7/260–265, δ13C = −28.7‰9680 ± 70
OxA-3348charcoal, ST C6b/315–320, δ13C = −25.7‰9700 ± 70
OxA-3349wood, ST C3/350–355, δ13C = −26.6‰9640 ± 70
OxA-3350plant remains, ST C2/400–410, δ13C = −27.4‰9500 ± 70
OxA-3351plant remains, ST C1/460–470, δ13C = −27.2‰9630 ± 100
OxA-4041wood, ST CB 1, 206–210 cm, δ13C = −28.2‰6515 ± 85
OxA-4042wood, ST CB 2, 242–245 cm, δ13C = −26.4‰7640 ± 85
OxA-4043plant remains, STCB 4, 326–330 cm, δ13C = −16.4‰10940 ± 100
OxA-4044carbonate. ST CB 4, 326–330 cm, δ13C = −0.7‰10700 ± 100
OxA-4045plant remains, ST CB 6, 356–358 cm, δ13C = −14.2‰11040 ± 100
OxA-4046carbonate, ST CB 6, 356–358 cm, δ13C = −0.5‰11250 ± 100
OxA-4376plant remains, ST C M1(B)1, δ13C = −28.5‰9385 ± 115
OxA-4377plant remains. ST C MI(B)2, δ13C = −28.5‰8940 ± 90

Comment (P. M.): the dated samples were collected in stratigraphic order from the basal 65 cm of organic lake-edge deposits, approximately 30 m to the east of the original excavations of G. Clark at the early Mesolithic (‘Proto-Maglemosian’) site of Star Carr, North Yorkshire. Plotting the dates against depth in the deposits reveals a clear plateau and associated ‘wiggle’ pattern which matches closely that recorded by Becker and Kromer in dendrochronologically dated pine samples around 9600 (radiocarbon years) BP. Calibration of these dates against the tree-ring data shows that the main phases of human activity, and associated burning of the local reedswamp, at Star Carr span a period of at least 300 years, from c. 10 400–10 700 BP in absolute terms (Mellars and Dark 1998).

There were two more samples from the same stratigraphic column dated later (OxA-4376, and −4377) which come from the immediately overlying levels.

Star Carr

Samples of charcoal from Star Carr, N. Yorkshire (NGR TA 028809), submitted by P. Dark, Dept. Archaeol., Univ. Reading, Whiteknights, Reading, through the NERC-funded ORADS facility.

OxA-4797Phragmites australis charcoal, CLK 1, δ13C = −27.7‰9385 ± 80
OxA-4798Phragmites australis charcoal, CLK 2, δ13C = −27.6‰9260 ± 100
OxA-4799Phragmites australis charcoal, CLK 3, δ13C = −26.1‰9500 ± 75

Comment (P. D.): the samples were designed to provide a chronology for a pollen and charcoal sequence from a sediment core taken in 1992, immediately adjacent to G. Clark's excavations of the early Mesolithic site at Star Carr (Clark 1954). The sequence showed two major phases of burning, and radiocarbon determinations were required to indicate the date of these for comparison with two burning episodes identified in palaeoecological sequences from the recently-excavated ‘Trench A’, 20 m to the east (Mellars and Dark 1998).

OxA-4797 was associated with the second phase of burning and with the beginning of the local population expansion of Corylus avellana (hazel). The central point of the determination is identical to that for the Corylus rise in monolith I (MI) from Trench A (OxA-4376). The coincidence with the Corylus rise suggests that this charcoal phase post-dates both burning episodes in MI, as the main charcoal phases in MI precede the Corylus rise.

OxA-4799 and −4798 were designed to date the beginning and end respectively of the first phase of burning. OxA-4798 is younger than expected, although its range overlaps with that of OxA-4799. For this first phase of burning the determinations are of little assistance in clarifying the relationship with Ml because OxA-4799 lies on the plateau in the radiocarbon timescale centred on 9600 BP (Day and Mellars 1994). Comparison with the pollen record suggests, however, that this phase is most likely to relate to the first episode of burning in Ml (Mellars and Dark 1998).

Broken Cavern

Samples from Broken Cavern, Torbryan, Devon (NGR SX 81506750), submitted by C. Gleed-Owen, Centre for Quaternary Science, School of Sci. and Env., Coventry Univ., through the NERC-funded ORADS facility.

OxA-6120Emys orbicularis bone, 3, BRK 525, δ13C = −24.3‰6125 ± 50
OxA-6121Rana temporana bone, 4, BRKSA 254, δ13C = −23.2‰860 ± 180
OxA-6122Rana temporaria bone, 5, BRKSA 250, δ13C = −26.0‰4600 ± 90

Comment (C. G.-O.): Arch. List 26 reported the first ever radiocarbon measurements on herpetofaunal (amphibian and reptile) remains from Britain, requested on species with palaeoclimatic and biostratigraphic significance, fifteen AMS dates were reported then but for various reasons several measurements were not reported until now. OxA-6120 was on indeterminate mammal bone and not on Emys orbicularis as had been thought. Nevertheless, the date provides extra information concerning the age of context 25, showing its initiation was significantly earlier than suggested by the Neolithic archaeology and previous AMS date of c. 4650 BP (OxA-3889). The samples for OxA-6121 and OxA-6122 were expected to produce Late Glacial ages, but both were unwittingly exposed to contamination during post-excavation storage in gelatine tubes (thus rendering the measurements invalid).

Upper Hayes Farm

An interdisciplinary investigation into the geomorphology and sedimentology, with focus on the formation and extent of the peat deposits, was undertaken at Loton Loop in Shropshire. It was hoped that the investigation would provide data for Shropshire Wildlife Trust, the Environment Agency and Severn-Trent Water to inform their investigations into flood management in the area. ‘Re-wetting’ of the surrounding agricultural land and other habitats in the Loop has been considered as part of the proposals for flood alleviation. A series of cores were taken in the area and it was considered that a date from the peat deposits would provide information on the approximate period of peat formation. A sample of peat from Upper Hayes Farm, Alberbury (NGR SJ 353164), England, was submitted by J. Tucker, Shropshire Wildlife Trust, Shrewsbury, comments by A. Wheeler, School of Applied Sciences, Univ. Wolverhampton.

OxA-7166peat, SJ31.J, δ13C = −29.6‰5560 ± 65

Comment (A. W.): although the findings have not been published, the data and a speculative environmental history of the Loton Loop area were presented at an open meeting about the results of the investigation. The meeting was held at Severn-Trent headquarters in Shrewsbury in May 1998 and attended by delegates from University of Wolverhampton, Shropshire Wildlife Trust, Environment Agency, Severn-Trent, Powys Archaeological Trust and other interested parties.

Roger's Cave

Samples from Roger's Cave, Wye Valley, Gloucestershire (NGR SO 56141550), submitted by C. Gleed-Owen, Centre for Quaternary Science. School of Sci. and Env., Coventry Univ., through the NERC-funded ORADS facility.

OxA-6123Bufo bufo bone. 8, 183, δ13C = −21.2‰1455 ± 55
OxA-6124Anguis fragilis bone, 12, 268, δ13C = −22.6‰168.4 ± 1.4 % mod

Comment (C. G.-O.): Late Glacial dates were expected but both samples were contaminated by storage in gelatine tubes (cf. Broken Cavern above). Although contamination is quite obvious for OxA-6124, stratigraphic difficulties at the site present another possibility for the young age of OxA-6123.


A sample from Flixton, Vale of Pickering, N. Yorks (54:12N 0:26W), submitted by J. B. Innes, Dept. Geography, Univ. Durham.

OxA-3734organic clay-mud sediment, F-1/50cm, δ13C = −27.7‰8930 ± 85

Comment (J. B. 1.): the sample selected for dating is from a thin layer of organic mud resting upon stiff basal clay and overlain by 20 cm sand containing Mesolithic flints. Above the sand is a peat profile dated 8340 ± 105 BP (Hv-17828) at the base, with a mid-Flandrian 1 pollen assemblage of Betula, Corylus, Ulmus and Quercus. The thin organic mud contains an earlier pollen assemblage dominated by Corylus, with Betula and Ulmus but no Quercus. The organic mud was dated to provide ages for the start and duration of sand deposition and so also limiting ages for its contained Mesolithic industry. The pollen data suggest a date in pollen zone Flandrian lb for the mud and so in a range around 9000 BP. OxA-3734 agrees very well with the age estimated from the pollen data and also conforms with the radiocarbon determination on the base of the upper peat.

Staines channel

Samples from Staines Channel, (NGR TQ 0270072570), England, submitted by S. Ford, Thames Valley Archaeol. Services, 47–49 Beauvoir Road, Reading.

OxA-6469peat, SMI 95 179–181, δ13C = −30.3‰9710 ± 75
OxA-6470peat, 89–91, δ13C = −30.9‰8710 ± 70
OxA-6474peat, SBC94 TVAS94/14, δ13C = −27.9‰8460 ± 65

Comment (S. F.): OxA-6469, −6470 and −6474 dated deep peat deposits which were also subjected to pollen analysis. A full report has been published and relevance of dates presented in Keith-Lucas (2000)


Samples from Bradfield, (NGR SV 60007260) England, submitted by S. Ford, Thames Valley Archaeological Services, 47–49 Beauvoir Road. Reading.

OxA-6834peat, BR 95 230, δ13C = −30.3‰11 030 ± 90
OxA-6835peat, BR 95 80, δ13C = −29.1‰8040 ± 70

Comment (S. F.): two dates on a deep peat deposit located in the Pang valley, West Berkshire during construction of the Theale-Bradfield pipeline. The location of the deposit is important as the site is close to the Berkshire Downs chalkland and provides palaeoenvironmental information for a region where there are few sources available.

River Tawe, Swansea

This was a sample of bark taken from an oak(?) tree trunk found in current bedded sandy gravels in a deep excavation to recover a tunnelling machine, under the River Tawe, Swansea (NGR SS 66269294). The elevation of the sample with respect to Ordnance Datum was 6 m B.O.D. Submitted by T. S. Smith, Thyssen Geotechnical, Thyssen House, Bynea, Llanelli, Carmarthenshire, SA14 9SU. Comments by J. E. Bonney, Thyssen Geotechnical

OxA-8065oak(?) wood, M83976, δ13C = −25.2‰6015 ± 55

Comment (J. E. B.): the need for dating arose from a contractual claim in respect of the depth of the made ground on the riverbank. It was claimed that the quite substantial tree trunk indicated the gravels were late Victorian fill. The bedded nature of the gravels and dating of the included timber confirmed otherwise. From the geological point of view the dating is interesting as these are the gravel deposits over which the River Tawe now flows.

Anderby Creek

Sample of wood from Anderby Creek, Lincolnshire (NGR TF 544763), submitted by T. Clare, Biological and Earth Sciences, Liverpool John Moores Univ., Liverpool.

OxA-6791oak wood, δ13C = −26.9‰4310 ± 50

Comment (T. C): this determination was obtained from biogenic deposits as part of a study of submerged forests (Clapham et al. 1997; Clapham 1999). Two previous dates (OxA-5963: 4480 ± 55 BP and OxA-5964: 4625 ± 55 BP, Arch. List 22) are sufficiently similar to suggest that the recorded deposit developed over some four hundred years or more.

High Hyton

Sample of wood from High Hyton, Cumbria, (NGR SD 102874), submitted by T. Clare, Biological and Earth Sciences, Liverpool John Moores Univ., Liverpool.

OxA-7184wood, HH94/3, δ13C = −25.6‰4485 ± 60

Comment (T. C): the determination was obtained from wood which lay at the side of a shallow valley and on the upper surface of the peat deposit described in Clare (1995). It is consistent with the dates obtained from the upper part of the peat obtained in the centre of the valley (OxA-5086 and −5087, Arch. List 22) and appears to correspond to a period of lower sea level (Clare 2000, fig.3). More importantly, however, the macro-fossil remains obtained from the side of the valley, which include uncharred finds of an emmer spikelet fork (Triticum dicoccum) and a flax seed (Linum ustatissimum), can be considered late Neolithic in date.

Kendal Castle Hill

Sample of wood from Kendal Castle Hill, Cumbria (NGR SD 523927), submitted by T. Clare, Biological and Earth Sciences, Liverpool John Moores Univ., Liverpool.

OxA-6902wood, KCI, δ13C = −26.0‰8185 ± 55

Comment (T. C): a small wetland area lies to the east and at the base of the drumlin on which the castle stands. This wetland area is the only visible remains of a once more extensive tarn which was, presumably, exploited as part of the medieval park which stretched eastwards from the castle. A core taken for pollen analysis was over 3.3 m deep showing that the sides of the tarn are remarkably steep. Nevertheless, the pollen analysis (King 1993) suggests that broadleaved woodland was the dominant vegetation for much of the time although at least one phase of clearance could be recognised. It was assumed that the broadleaved woodland must predate the main occupation of the castle and to confirm this a sample was taken from a woody layer at the top of a second core, some 30 cm below the present ground surface. The determination obtained is, therefore, difficult to reconcile with the stratigraphic position of the sample and it is concluded that the upper part of the wetland represents deposition of material obtained from a greater depth, perhaps when the present brook was created at the back of domestic properties.


A sample from Portinscale, Cumbria (54:36N 3:10W), submitted by T. Clare, Biological and Earth Sciences, Liverpool John Moores Univ., Liverpool.

OxA-6215twig or bark, Port 2, δ13C = −29.3‰4025 ± 45

Comment (T. C): in 1901 during construction of a house and gardens four ‘roughout’ stone axes, some chippings and an upright section of trunk were found at the edge of a wetland area (Cumberland and Westmorland Antiquarian and Archaeological Society ns.2, 418–9). A small-scale excavation of the site was undertaken in 1992 but no artefacts were found. However, organic remains were encountered in coring the hillside below the house. A subsequent small excavation, undertaken in 1994, showed that the thin organic layer contained twigs, birch bark and some leaves, including holly, which had previously been interpreted as modern garden refuse. This interpretation was, however, difficult to reconcile with the stratigraphy and two samples were sent for radiocarbon analysis. The date given here is consistent with that obtained from the second sample (UB-3641: 4392 ± 59 BP) and shows that a land surface contemporary with the axes survives beneath part of the present domestic garden.


Samples from Coulererach, Isle of Islay (NGR NR 208652) Scotland, submitted by S. Mithen, Dept. Archaeol., Univ. Reading, through the NERC-funded ORADS facility. The samples were submitted to gain absolute dates on a Mesolithic occupation sealed beneath 1.5 m of peat at Coulererach, and on a pollen sequence from within the peat. Coulererach is located on the west coast of Islay and test excavations were undertaken there by the Southern Hebrides Mesolithic project in 1993, after the local crofter alerted S. Mithen to flint artefacts eroding from her ditches.

OxA-4921ericaceous twig, Coul-1, δ13C = −30.6‰4010 ± 75
OxA-4922monocotyledon leaf/stem, Coul-2, δ13C = −29.2‰3860 ± 110
OxA-4923plant remains, Coul-3, δ13C = −31.1‰4895 ± 80
OxA-5040monocotyledon/Eupa plant remains, Coul-4, δ13C = −27.1‰4615 ± 70
OxA-5153monocotyledon plant remains, Coul-5, δ13C = −27.1‰4700 ± 100
OxA-4924charcoal, Coul-6, δ13C = −26.4‰7530 ± 80

Comment (S. M.): OxA-4924 was on charcoal from amidst a dense lithic scatter. The result agrees with the typological affinities of the microliths and provides the only absolute date from the settlement. OxA-4923 comes from the base of the peat immediately above the artefacts showing that there was either a hiatus between human occupation and the start of peat formation, or the loss of sediments due to erosion. The other determinations show that the peat accumulated rapidly, and have provided significant insights into vegetation development on Islay. Full discussion of the Mesolithic occupation at Coulererach is provided in Mithen and Finlay (2000), and of the vegetation history in Bunting et al. (2000).

Broad Ditch

The sample of wood was found in 1987 during the construction of a new sewer from the Wisley Sewage Works (NGR TQ 057602), Surrey. The line cut across a feature known as Broad Ditch, which marks the parish boundary and is generally accepted as an earlier course of the River Wey. Various pieces of wood were recovered under salvage conditions including a number of stakes and a substantial beam about 1.78 m in length, pierced at each end by pairs of sub-rectangular holes. There was no associated dating evidence, hence the decision to seek radiocarbon dating. Submitted by D. G. Bird, County Planning Dept., Surrey County Council, Kingston-upon-Thames, Surrey.

OxA-3166oak wood, δ13C = −26.0‰1180 ± 95

Comment (D. G. B.): a draft report has been prepared by D. Williams, who prepared record drawings of the material, but completion of the report has so far foundered on the need for additional information from the original finder. It is suggested that the timber is closely similar to one from a structure excavated at Anslow's Cottages, Burghfield, Berkshire, dated to the Late Saxon period (Butterworth and Lobb, 1992, esp fig 36 and pl 13). The Burghfield beam was found across the line of a channel and is considered to be possibly a part of a sluice-gate mechanism, perhaps for controlling water flow to a system of water meadows. This would make sense in terms of the surroundings of the Broad Ditch.

Cross Glebe, Ness

Samples of plant remains from Cross Glebe, Ness, Isle of Lewis (NGR NB 503627), Scotland, submitted by R. N. Hubbard, Paleobiology Research Unit, Univ. East London.

OxA-7123pollen, MDB 09/96, 36, δ13C= −27.1‰7455 ± 55
OxA-9125organic core, sample 12, d13C = −28.6‰104.3 ± 0.5% mod
OxA-9188organic core, sample 7, δ13C = −28.9‰100 ± 45
OxA-9189organic core, sample 23, δ13C = −28.4‰390 ± 55

Comment (R. N. H.): of the four AMS determinations from the mire deposits at Cross Glebe, two are readily understood, and explain the remaining two determinations.

The initial age determination, OxA-7123, from just above the base of the very shallow mire deposits, was unexpectedly old given its depth of 58 cm, but makes very good sense in retrospect. The sample corresponds to the point at which the percentage of Salix pollen ceases to decline, and similar vegetational events have been dated to similar times elsewhere in north-western Scotland.

OxA-9125, from a depth of about 19 cm, should have dated a marked change in the herbaceous vegetation that must relate to an important change in land use that was estimated (on the evidence of OxA-7123) to have occurred 2000–2500 BP. The ‘future’ date can only have arisen from material photosynthesised in the late AD 1950s—early 1960s, when atmospheric nuclear tests enriched the environment in 14C to an extent that swamped the fossil fuel effect. What has been dated is clearly an intrusion, presumably an Eriophorum or Agropyron rhizome fragment.

OxA-9188 and −9189 should have dated (respectively) a pronounced increase in sedge pollen at an estimated age of 1100–1500 BP and the first signs of Neolithic settlement in the locality around 4600 BP. Taken at face value, OxA-9188 implies a major ecological change almost within living memory, and OxA-9189 implies four major ecological changes within the past 300–400 years: neither is remotely tenable. Obviously recent plant rhizomes have consituted a large proportion of the material dated.

Disappointing as some of these results are, they point to ways in which the sub-sampling and dating of such featureless matrices could be improved. Whereas an object like a leaf or a seed is unlikely to be intrusive, any woody material probably should be regarded with suspicion. A sensible procedure would be to disaggregate the material in dilute alkali, and to sieve the dispersed constituents. Objects large enough to be visible to the naked eye, or retained on a 2 mm sieve, are likely to be seeds or leaves (prime material for AMS dating) or intrusive roots or rhizomes that should be avoided at all costs. Material in the 2000–250 μm range is more likely to be dominated by moss leaves that should reflect the primary accumulation.


Irish Quaternary fauna project

These results form part of the Irish Quaternary fauna project which was funded by the National Heritage Council. Stage 1 and 2 were published in Arch. Lists 23 and 24 respectively. Results from Edenvale Cave (52.40N 09:00W) and Kilgreany Cave (52:05N 07:44W) were omitted from Arch. List 24. The samples were submitted by P. C. Woodman, Dept. Archaeol., Univ. College Cork, Ireland.

OxA-3700Equus caballus, Edenvale Cave, F21099, bone, δ13C = −19.6‰1675 ± 60
OxA-4241Megaloceros giganleus, Kilgreany Cave, F21155, bone, δ13C = −20.4‰10 960 ± 110

Comment (P. C. W.): OxA-3700 is obviously part of a series of Iron Age/Early Medieval determinations from the Edenvale Cave complex. OxA-4241, from Kilgreany Cave is interestingly contemporaneous with a reindeer date from the same cave and reinforces the evidence of a Late Glacial faunal presence at the site.

The Burren

Sample of charcoal from The Burren, Co. Clare (c. 53N 9W), Ireland, submitted by N. R. Moles, School of Geosciences, Queen's Univ., Belfast.

OxA-7784charcoal, NM94B, δ13C = −25.5‰3070 ± 35

Comment (N. R. M.): while previously reported evidence from lacustrine sediments has linked Burren soil erosion to anthropogenic forest clearance, there have been no previous accounts of dateable episodes of prehistoric slope instability based on evidence from terrestrial sites. Moles et al. (1999) describe the occurrence of charcoal buried to a depth of c. 2 m within diamicton at a site on Knockanes hill close to Mullach Mór hill in the south-eastern Burren. OxA-7784 places the formation of this charcoal in the Bronze Age, contemporary with exceptionally high rates of soil erosion and forest clearance in a nearby lake catchment. EMS images indicate that the charcoal is of fine-grained deciduous wood, most probably hazel (Corylus avellana) but perhaps birch (Betula sp.) Results of analyses of diamictons present in the vicinity of the charcoal support the view that the pattern of occurrence of diamictons in the present day Burren landscape is in part the product of prehistoric mass movement.


Stor Amyran

Samples of plant remains from Stor Amyren (6:45N 20:05E), Sweden, submitted by F. Oldfield, Dept. Geography, Univ. Liverpool, through the NERC.

OxA-3767Sphagnum peat, B/20–24 cm, δ13C = −25.8‰55 ± 55
OxA-3768Sphagnum peat. B/25–29 cm, δ13C = −24.8‰145 ± 55
OxA-3769Sphagnum peat, B/30–34 cm, δ13C = −24.8‰45 ± 55
OxA-3770Sphagnum peat B/35–39 cm, δ13C = −25.3‰100 ± 50
OxA-3771Sphagnum peat. B/39–43 cm, δ13C = −25.1‰295 ± 60
OxA-3772Sphagnum peat, B/43–47 cm, δ13C = −24.6‰485 ± 55
OxA-3773Sphagnum peat, B/47–51 cm, δ13C = −25.6‰605 ± 55
OxA-3774Sphagnum peat. B/51–55 cm, δ13C = −24.7‰615 ± 60
OxA-3775Sphagnum peat. B/55–59 cm, δ13C = −26.1‰625 ± 55
OxA-3776Sphagnum peat, B/59–62.3 cm, δ13C = −24.4‰820 ± 55

Comment (F. O.): the dates listed were on hand picked leaves and stems from slices of Sphagnum peat taken from a 19 cm diameter, 70 cm long core, as part of a project designed to assess possible links between recent climate variability at high latitudes and the rate of net sequestration of carbon in ombrotrophic peat. The results from AMS dating were compared with high resolution dates based on bulk slices. These latter were carried out in the Queen's University of Belfast laboratory. The aim of the comparison was to indicate whether AMS determinations with larger errors based on hand-picked material were more suitable for our purpose than the high resolution dates which have better statistical precision, but more uncertainty regarding contributions from rootlet penetration or from the time transgressive character of a large diameter slice.


Biniai Nou, Menorca

Samples of Ateleryx algirus bones, the Algerian hedgehog, from Biniai Nou, Menorca (39:54:25N 4:12:49W), Spain. Submitted by A. Morales, Laboratorio de Arqueozoología, Departamento de Biología, Universidad Autónoma de Madrid, Spain.

OxA-10469Aleleryx algirus bone. BIN 1, d13C = −20.8‰634 ± 34
OxA-10487Ateleryx algirus bone, BIN 2, d13C = −21.8‰355 ± 75

Comment (A. M.): Ateleryx algirus presently occupies the littoral fringe of northern Africa, from Morocco to Libya, as well as the Iberian Mediterranean all the way to Provence, the easternmost of the Canary islands, plus a series of islands in the Mediterranean, including Malta and the Balearic archipelago. Its status outside Africa has been alternately described as ‘natural’ (most probably) or ‘introduced by man’ but, in the absence of any palaeontological or archaeozoological record, such labels await confirmation. The remains dated originated in a local Bronze Age communal grave on the island of Menorca featuring large quantities of microvertebrates, mostly rodents, which gave the impression of being intrusives, as indeed these Middle Age dates now confirm. These dates point to an introduction of the species in a far more recent moment than hitherto postulated, probably carried out by northern African peoples who at that moment occupied not only these islands but most of the Iberian peninsula as well. It remains to be seen whether earlier dates are found in areas outside its endemic (ie. African) region.


Venice Lagoon

The following samples of wood and plant remains come from two islands in the northern part of the Venetian Lagoon, Sant'Erasmo and San Francesco del Deserto, Italy (45:32N 12:24E). They constitute part of an extensive series of radiocarbon determinations undertaken at ORAL) on samples from early sites in Venice and the Lagoon (Ammerman et al. 1999). Submitted by A. J. Ammerman, Dept. Classics, Colgate Univ., New York, USA, in conjunction with the Soprintendenza per I Beni Ambientali e Architettonici for Venice.


In the early Middle Ages, Sant'Erasmo (like the island known today as the Lido, which occurs further to the south) once formed one of the barrier islands along the coast between the Adriatic Sea and the Venetian Lagoon. In 1997, several cores were made at the north-east end of Sant'Erasmo in order to study the evolution of the island's geomorphology. The sample of wood in core 6 comes from a depth of 420 cm, which corresponds to an elevation of 1.39 m below sea level (with reference to the 1897 sea-level standard commonly used in Venice).

OxA-7439wood, core 6, 25, 420cm-l.39masl, δ13C = −27.1‰1205 ± 50

Comment (A. J. A.): the sample was recovered in the context of a sandy channel fill. The determination, as expected, shows that the morphology of the north-east end of the island was quite different in the time before the tenth century AD.

S. Francesco del Deserto

A series of five AMS determinations were previously determined on samples of wood and reeds recovered during the course of archaeological work on the island (OxA-4512 through −4516; see Arch. List 19, pp.211–2; Ammerman et al. 1995). OxA-6784 comes from a hollow piece of wood (US 2036; 13 cm diameter, 68 cm long) recovered from the excavation on the north side of the island; it is interpreted by the excavator, M. De Min, as a late Roman drainpipe or fistula made of wood on the basis of its stratigraphic position and its associated ceramic finds. OxA-6770 comes from a lens of plant remains found within a horizon of lagoonal sediment having a distinctive bluish colour; it was recovered in archaeological core 11 on the west side of the island. OxA-6771 consists of a thin lens of plant remains found at the boundary that occurs at the base of the sequence of lagoonal sediments in the same core. On the basis of measurements undertaken previously on this boundary at other sites in Venice, a calibrated age of around 6000 years ago was expected for this sample. Again the elevations are given with reference to the 1897 standard in Venice.

OxA-6770plant remains, 7, δ13C = −26.0‰750 ± 45
OxA-6771plant remains, 8, δ13C = −26.2‰4495 ± 60
OxA-6784wood, 6, δ13C = −22.8‰1785 ± 50

Comment (A. J. A.): all three determinations are in good agreement with what was expected. OxA-6784 documents occupation going back to late Roman times on the island. OxA-6771 gives an age much the same as three other determinations (OxA-4817, −5895, −5896, Arch. List 19, 22) for the mid-Holocene transgression that led to the formation of the Venetian Lagoon (Ammerman et al. 1999).



All of the samples selected for AMS dating were taken from critical points in archaeological and palaeoenvironmental sequences in the northern part of the Epirus region of north-west Greece, in order to provide greater precision for the dating of Late Glacial/Early Postglacial environmental changes and their correlation with changes in the sequence of human activities in neighbouring rockshelter deposits. Submitted by Professor G.N. Bailey, Dept. Archaeol., Newcastle Univ. through the NERC-funded ORADS facility.

Boila Rockshelter, 20:34N 39:59E  
OxA-5241charred bone 4, δ13C= −21.2‰12 480 ± 120
OxA-5242charred bone 5, δ13C = −26.7‰13 240 ± 110
OxA-5243charred bone 8, δ13C = −21.1‰10 190 ± 90
OxA-5244charred bone 10, δ13C = −26.3‰70 ± 55
OxA-5245charred bone 28, δ13C = −21.2‰10 560 ± 110
OxA-5246charred bone 35, δ13C = −20.6‰13 810 ± 130
Gramousti, 40:00N 20:40E  
OxA-3486charcoal, 3:0.90m/2/3, δ13C = −19.1‰4770 ± 80
OxA-3487charcoal, 2:0.90–1, 10m/2/1, δ13C = −23.4‰4510 ± 75
OxA-3488charcoal, 21:1.35m/5/7, δ13C = −23.9‰4560 ± 70
OxA-3489charcoal, 10:1.40m/5/7, δ13C = −25.6‰4740 ± 75
Grava rockshelter, 39:28N 19:53E  
OxA-3401charred bone G7, 4a, δ13C = −17.2‰8840 ± 85
OxA-3402charred bone G8 4d, δ13C = −16.6‰9710 ± 95
Kalamas Basin, 39:58N 20:41E  
OxA-3942shell, D-87–30 PA53, δ13C = −9.0‰6260 ± 70
OxA-3943shell, D-87–11 PA30, δ13C = −9.0‰9160 ± 80
Klithi, 39:58N 20:41E  
OxA-2970charred bone. F0948, 39, δ13C = −24.2‰4290 ± 140
OxA-2971charred bone, F0962, 22, δ13C = −22.4‰16 650 ± 190
OxA-2972charred bone, F8531, 26, δ13C = −24.2‰16 140 ± 150
OxA-3732charred bone, F8504, V25, δ13C = −19.3‰14 570 ± 130
OxA-3941charred bone, D1248, sample 17, δ13C = −17.1‰13940 ± 110
Megalakkos, 39:58N 20:41E  
OxA-2973charred bone. Unit N (c), δ13C = −24.8‰100.2 ± 0.8 % mod

Comment (G. N. B.): all the results have been fully published and commented on as follows: for general comments and interpretation of all dates (other than Badanj) see Gowlett et al. (1997). For more detailed comment and interpretation of the Boila rockshelter dates see Kotjabopoulou et al. (1997). For more detailed comment and interpretation of the Gramousti sequence (lake pollen sequence) see Willis (1997) and Turner and Sanchez-Goni (1997). For more detailed comment and interpretation of the Grava rockshelter see Bailey (1997a; 1997b). For more detailed comment and interpretation of the Kalamas Basin dates see King et al. (1997). The main point of these Kalamas determinations was to test the hypothesis, based on one thermoluminescence date and some potsherds, that a 30 m high river terrace was of Postglacial date (and therefore evidence of subsequent tectonic movement, rather than of Glacial Maximum date (and therefore most probably evidence for river incision resulting from climatic change. The dates, being on freshwater shells which must be contemporaneous with the formation of the river sediments (rather than intrusions), support the hypothesis of a Postglacial date. For more detailed comment and interpretation of Klithi dates see Bailey and Woodward (1997). For more detailed comment and interpretation of Megalakkos see Sinclair (1997).


Gesher Benot Ya'aqov

Samples of charcoal from the site Gesher Benot Ya'aqov, (32:37N 35:33E) Israel, submitted by N. Goren-Inbar, Inst. of Archaeol., Hebrew Univ., Jerusalem, Israel.

OxA-5053charcoal, 1823,Co2, δ13C = −14.6‰8550 ± 110
OxA-5054charcoal, 1824,CO2, δ13C = −14.6‰1760 ± 50

Comment (N. G-I.): these samples originated from the waterlogged Acheulian site of Gesher Benot Ya'aqov. The Middle Pleistocene (and evidently the Lower) sediments are overlaid by a flood plain sediment deposited by the River Jordan. It was of interest to find out the age of these deposits. One hypothesis was that all the deposits are the result of a single depositional event (described in the literature), a mega-flood that took place at the end of the 1960s. Indeed in the last field season deposits of modern refuse were found in the flood plain. Nevertheless, resulting from these analyses, it seems that the deposit is not homogeneous, and in it are some more ancient components. Some geographers or sedimentologists will find these results interesting.


Uan Muhuggiag

Samples of desiccated seeds from the Neolithic rock shelter of Uan Muhuggiag located in the Tadrart Acacus in south-west Libya (25:10N 10:35E). Excavated by B. Barich between 1978 and 1983. Previous radiocarbon determinations indicate two occupation horizons separated by a rocky landslide: the older horizon dates between 7438 ± 120 BP and 5350 ± 200 BP, the younger one dates between 4730 ± 120 BP and 2220 ± 220 BP (Barich 1987, 1992). Submitted 1993 by M. van der Veen, Sch. Archaeol. Stud., Univ. Leicester. Comments by K. Wasylikowa, Institute of Botany, Polish Academy of Sciences, Krakow, Poland, and M. van der Veen.

OxA-4389Phoenix sp. plant remains. MV93, 05, δ13C = −23.3‰2130 ± 70
OxA-4390Citrullus lanatus plant remains, MV93, 06, δ13C = −26.1‰5400 ± 80

Comment (K. W. and M. V. der V.): the plant assemblage from this site consists entirely of wild plants (wild grasses, wild herbaceous plants and a few tree taxa; Wasylikowa 1992), even though 90% of the faunal remains belonged to domestic animals. The assemblage did not contain any indicators of plant cultivation and the presence of remains of the date palm (Phoenix sp.) and watermelon (Citrullus lanatus (Thunb.) Mansf.), which do not belong to the native flora of Libya, was of special interest. For this reason, these samples were submitted to test whether they represent secondary intrusions. The determinations obtained show that they belong to the archaeological contexts from which they were recovered.

Desiccated fruit-stones of the date palm (Phoenix sp.) were found in the top layer in sector B from which three other determinations were obtained: desiccated seeds of Citrullus colocynthis (Gd-4290: 2220 ± 220 BP); desiccated coprolites (Gd-4288: 2770 ± 80 BP): and desiccated fruits of Balanites aegyptiaca (Gd-2854: 3810 ± 80 BP). These results identify the mixed nature of plant material preserved in this layer, but exclude modern contamination, and indicate that the shelter was still used at c. 2100 BP. Date stones of comparable age were found at the agricultural settlement of Zinchecra in Fezzan, southern Libya (Van der Veen 1995).

Desiccated seeds of Citrullus lanatus (Thunb.) Mansf. have been found in layer 2a which contained abundant remains of grasses and stones of Balanites aegyptiaca. Desiccated Balanites stones were dated to 5420 ± 100 BP (Gd-2960). According to various authors (Jeffrey 1967; Zeven and Zhukovsky 1975) watermelon is native to tropical and southern Africa. C. Jeffrey (Herbarium, Royal Botanic Garden, Kew) has kindly examined the anatomy of the Citrullus lanatus seeds from Uan Muhuggiag, and he concluded that they are more or less identical to modern perennial specimens of this species collected on sandy waste places in Niger (Jeffrey, pers. comm.). In his opinion “… it is likely that the archaeological material represents this perennial plant, rather than the typical cultivated annual C. lanatus in strict sense.” However, he has also indicated that the Uan Muhuggiag specimens “…and the Niger perennial species seeds fall within the range of variation of both C. colocynthis and C. lanatus, according to published data, so are not specifically diagnostic.” Our seeds may represent hybrids of the two species or a subspecies of C. lanatus. Clearly more studies are needed to elucidate the exact status of these seeds and to determine the exact area of origin of the domestication of watermelon. The presence of watermelon seeds at Uan Muhuggiag may indicate that about 5000 years ago the range of the wild C. lanatus extended further north than at present. We cannot say how much further north, but the plant must have been available for gathering within the area exploited by the nomadic pastoralists who brought the seeds to the Uan Muhuggiag rock shelter.



Sample of plant remains from the site of Nikoni (46:12N 30:26E), Ukraine, submitted by A. Dodonov, Geol. Inst., Russian Academy of Sciences, Moscow, Russia.

OxA-7970plant remains, humus, BR 1, δ13C = −24.7‰26 760 ± 240

Comment (A. D.): this determination provides evidence of the stratigraphical position of the Bryansk palaeosol in the northern Black Sea shore region. It is necessary to note that the chronology of loess-palaeosol formation even at the top of sections is rather confused due to many misunderstandings and different thermoluminescence dates previously produced. That is why 14C determinations are very important to clarify the situation if we have acceptable material for dating. The reduced sample sizes that are required make AMS very useful for stratigraphical research. See also Dodonov et al. (2000).


Basalt Lake

Samples of plant remains from Basalt Lake. Greenland (72:43:04N 22:27:09W) and Lake S of Basalt Lake, Greenland (72:42:8N, 22:29:7W), submitted by B. Wagner and W.-D. Hermichen, Alfred Wegener Institute for Polar and Marine Research, Potsdam, Germany.

Basalt Lake  
OxA-7253plant remains, 1/1205–2, 32–34cm, δ13C = −27.9‰845 ± 40
OxA-7254plant remains, 3/1205–2, 124cm, δ13C = −27.9‰4175 ± 50
OxA-7286plant remains, 2/1205–2, 40–42cm, δ13C = −27.6‰985 ± 50
OxA-7287plant remains, 5/1205–2, 180–182cm, δ13C = −27.7‰6455 ± 70
Lake S of Basalt Lake  
OxA-7264plant remains, 13/1212–3, 160–162cm, δ13C = −28.8‰8580 ± 160
OxA-7265plant remains, 6/1212–3, 4–6cm, δ13C = −28.6‰215 ± 50
OxA-7266plant remains, 7/1212–3, 16–38cm, δ13C = −29.4‰1590 ± 50
OxA-7267plant remains, 8/1212–3, 68–70cm, δ13C= −27.8‰2970 ± 55
OxA-7268plant remains, 9/1212–3, 90–92cm, δ13C = −29.1‰4025 ± 55
OxA-7269plant remains, 11/1212–3, 125cm, δ13C = −23.8‰6515 ± 65
OxA-7270plant remains. 12/1212–3, 136–138cm, δ13C = −27.7‰6950 ± 65
OxA-7559plant remains, 10/1212–3, 118–120cm, δ13C = −28 8‰6005 ± 80

Comment (B. W. and W.-D. H.): these determinations were from two lake sediment cores from East Greenland. The lake sediments were used to get some new information about the time of deglaciation of the investigated area, to reconstruct the Holocene environmental history of the region, and to compare this information with existing records from earlier expeditions. The near-surface sample from the smaller lake to the south of the Basalt Lake indicated that the surface sediments were not disturbed by the piston corer taken for the recovery of the sediment.


Withdrawn OxA numbers

The following OxA numbers have been withdrawn for other reasons:

OxA-8332 and OxA-8758 from Thorneybank (replaced with OxA-10160 and OxA-10161) because of anomalous or variable CN ratios. OxA-9624 from Eye Kettleby was withdrawn for chemical pre-treatment reasons.