Direct radiocarbon dating of megalithic paints from north-west Iberia.

Steelman, K.L. ; Ramirez, F. Carrera ; Valcarce, R. Fabregas 等

On the Iberian Peninsula, post-Palaeolithic paintings--in contrast with Palaeolithic images-have received scant attention from the AMS radiometric technique. In fact, only one radiocarbon date consistent with generally expected values has been previously determined on a painted megalith; charcoal from a black-painted tomb panel in a corridor at Antelas, Viseu, Portugal, was dated to 4655 [+ or -] 65 BP (Cruz 1995a,b). Here we present nine accelerator mass spectrometry (AMS) radiocarbon dates for megalithic paintings in north-west Iberia (Figure 1), placing these paintings as the oldest known examples of prehistoric art in Galicia (northwest Spain). Two of us (FCR & RFV) have actively investigated north-west Iberian megalithic art since 1998, systematically locating and recording pictorial remains (doubling the number of known sites in the Galician area) (Carrera Ramirez & Fabregas Valcarce 2003).

[FIGURE 1 OMITTED]

The progressively widespread use of radiocarbon dating using AMS since the late 1980s has overcome an important barrier to dating rock paintings--the ability to analyse small amounts of material available in a paint sample. AMS has opened new possibilities for dating those paintings that contain an organic component (charcoal, blood, fats/oils). The first AMS radiocarbon date for a rock painting was obtained for a charcoal painting located in South Africa (van der Merwe et al. 1987). Since that time, researchers have used AMS to radiocarbon date rock paintings throughout the world (see Rowe 2001 for a review of published dates).

Paintings on stone, whether on a cave wall or a megalithic monument, pose serious and unique challenges for accurate dating: (1) images are often painted on limestone, a carbon-containing mineral; (2) the amount of carbon from the paint sample available for dating is small-orders of magnitude less than a typical artefact; (3) little is known about binders and/or vehicles used in making ancient paints; (4) physical contamination must be removed; and (5) organic material unassociated with painting activity can occur in unpainted rock (see Rowe 2001; Bednarik 2002; Steelman et al. 2002; Steelman & Rowe 2005, for further discussion concerning pictograph dating, as well as Pettit & Bahn 2003; Valladas & Clottes 2003, both published in Antiquity). Charcoal paintings are most commonly dated, but the plasma-chemical extraction method also allows the dating of pictographs with inorganic pigments (if an organic material was added to the paint). Our (MWR & KLS) laboratory has repeatedly attempted to verify our results by dating radiocarbon standards with previously measured ages and rock paintings for which an archaeologist had some inferred age range (Rowe & Steelman 2001).

Using plasma-chemical extraction and AMS, we have obtained nine radiocarbon dates for paint samples taken from six Galician megalithic monuments (Figure 2). These results stand out by their general agreement with one another.

[FIGURE 2 OMITTED]

Description of monuments studied

We sampled black paints from gneiss, granite and schist dolmen stones at Pedra da Moura, Casota do Paramo, Pedra Cuberta, Maimoa de Monte dos Marxos, Forno do Mouros, Anta de Serramo, Coto dos Mouros and Dolmen de Dombate, all located in Galicia (north-west Spain). These monuments are all passage graves, consisting of a polygonal chamber and a short corridor (Figure 3). Typically, the orthostats (vertical upright wall stones) forming the corridor are shorter and smaller than those of the chamber. Unfortunately, these monuments have suffered from vandalism in which some stones were removed as construction material. Inspection of the accessible orthostats reveals visible remnants of paint; however, the graphic images are often highly degraded and unrecognisable.

[FIGURE 3 OMITTED]

Red and black geometric paintings were typically applied over a white plaster coating the interior of the stone monument. We collected only black paint samples, except for one red paint sample from Dolmen de Dombate. Pigment of black paint was identified as charcoal by the characteristic structure of burnt wood observed under an optical microscope, as well as by the disappearance of black colour noted during plasma oxidation. In the limited attempts to characterise megalithic paint manufacture, dark pigments have consistently been identified as wood charcoal (Shee Twohig 1981; Bello Dieguez 1995).

Sample collection

Samples were removed from the stones using a surgical scalpel, and then placed directly into a sterilised glass vial. The decision of how much sample to remove was subjective, and varied depending on whether the material removed was primarily paint or, as more commonly occurred, included a portion of the plaster layer. The quantity of sample collected was kept to a minimum, typically less than 30mg and often much less, of the order of 10mg, when the sample included only the surface layer rich in paint. We attempted to collect samples from areas with little visual significance and which appeared to be free from biological activity (algae, lichens, etc.). In some cases, we first cleaned the sampled area, always using mechanical means (small brushes) and no water.

Pedra da Moura (Vimianzo, A Coruna)

Paint sample M1 (~26mg) was collected from the second orthostat on the left side of the corridor (L2). It contains none of the white plaster and is composed primarily of black paint and accretion minerals.

Casota do Paramo (Boiro, A Coruna)

Paint sample M2 (~33mg) was taken from the base of the first orthostat on the left side of the chamber (C 1) and consists of as much preparative white plaster as black pigment.

Pedra Cuberta (Vimianzo, A Coruna)

We collected paint sample M3 (~13mg) from Pedra Cuberta. The well-preserved black paint was collected from orthostat L1, the first on the left side of the corridor.

Mamoa do Monte dos Marxos (Rodeiro, Pontevedra)

No longer in existence as an intact monument due to severe vandalism, the orthostats have been moved to the Museum of Pontevedra. Of significant interest is the presence of two distinct layers of paint applied on seven of the twelve orthostats examined. Two paint samples were removed from orthostat 7: sample M4 (~6mg) was from the underlying layer of paint (between plaster and surface paint) and sample M5 (~16mg) was taken from the overlying surface layer.

Forno dos Mouros (Toques, A Coruna)

The preservation of the red and black horizontal zigzag paintings is good on one of the uprights (C1, Figure 4), but very fragmentary on the rest. Sample M6 (~22mg) was collected for dating from the C1 upright in 1998 and consists of black paint made of charcoal.

[FIGURE 4 OMITTED]

Anta de Serramo (Vimianzo, A Coruna)

Two small paint and plaster samples (each ~20mg) were taken from the lower part of upright C1, close to the soil, making it very difficult to obtain a clean extraction. This fact might explain the outlying radiocarbon result.

Coto dos Mouros (Rodeiro, Pontevedra)

There is evidence of superimposition of paint, as at Monte dos Marxos. Sample M10 (129rag) contained multiple layers and sample M11 (114mg) was from the underlying paint layer. During sample collection, we noted a black coating covering the interior of Coto dos Mouros, which seemed to be due to the burning of at least one tyre inside the monument. While both samples M10 and M11 were removed from areas away from the black coating and did not contain any visible soot, we must be cautious about radiocarbon dates from this monument.

Dolmen de Dombate (Cabana, A Coruna)

Preserved paintings only occur on those surfaces covered by the soil prior to excavation. Two paint samples were collected from Dolmen de Dombate, but unfortunately we were unable to determine their age. The first sample consisted of black paint (M8) taken from orthostat L2, and was accidentally lost during analysis. A second sample consisting of kaolin plaster plus some red pigment (M9) did not have sufficient organic material for an AMS radiocarbon date. Finally, we radiocarbon dated an organic residue from a pot recently excavated, assigned to the oldest occupation phase and, thus, providing us with a post quem date for the building of the passage grave and, presumably, its interior decoration.

Experimental procedure

Each paint sample was viewed under an optical microscope to detect any incorporated extraneous materials. Microscopic fibres were observed in the majority of the paint samples (M2, M3, M6, M7, M8, M11) and removed with tweezers. We believe the fibres are from plant rootlets that are actively damaging the buried stones. Sodium hydroxide washes in an ultrasonic water bath were performed following the procedure outlined in Steelman et al. (2002). While we rarely encounter humic acids in rock art samples, we did observe a colour change of the base wash indicating the presence of humic acids in these megalithic paint samples as might be expected for buried materials. Multiple base washes were performed until no colour change was observed. In our procedure, decomposition of carbon-containing minerals such as carbonates and oxalates into carbon dioxide is prevented by running the plasma at low-temperature (< 150[degrees]C). With plasma-chemical extraction, only organic material is removed for radiocarbon measurements.

Megalithic paint samples were subjected to plasma-chemical extraction and AMS [sup.14]C measurement. With our procedure, a low-temperature oxygen plasma (partially ionised low-pressure oxygen) selectively oxidises organic material in a paint sample, while any incorporated carbon-containing minerals remain unreacted. The copious amounts of carbonate and oxalate minerals typically included in rock paintings have no effect on radiocarbon dates when this technique is employed. With plasma-chemical extraction, only organic material is removed for AMS radiocarbon measurement. We routinely omit acid washes used by other laboratories, as they are unnecessary with plasma-chemical extraction (Russ et al. 1992; Chaffee et al. 1994; Pace et al. 2000). In contrast to our method, acid treatments used in conjunction with combustion methods may not completely remove carbon-containing minerals associated with rock paintings and their inclusion will influence radiocarbon results (Hedges et al. 1998; Armitage et al. 2001). Organic carbon in a paint sample is converted to carbon dioxide during plasma-chemical extraction. This carbon dioxide is reduced over a metal catalyst to form a graphite target for AMS 14C measurement at Lawrence Livermore National Laboratory's Center for Accelerator Mass Spectrometry (CAMS). For Anta de Serramo sample M7, carbon dioxide used for the AMS target was combined from two separate paint samples that were processed individually. Taken individually the samples did not produce enough carbon to date, but when combined gave 53[micro]g of carbon, sufficient for an AMS measurement.

Results

The radiocarbon results from this study of megalithic paintings are listed in Table 1. The AMS laboratory identification number and calibrated (20 [sigma]) age ranges are also listed. Calibration was performed using the OxCal computer program version 3.5 (Bronk Ramsey 2000) with atmospheric data from Stuiver et al. (1998). Two radiocarbon dates on an organic residue from an excavated pot at Dolmen de Dombate are 4900 [+ or -] 40 BP (CAMS 101903) and 6890 [+ or -] 40 BP (CAMS 101904), with a weighted average of 4895 [+ or -] 30 BP and calibrated to 3715-3635 cal BC at 95 per cent (20 [sigma]) using OxCal.

Discussion

Radiocarbon dating of charcoal paints corresponds to a maximum age for the creation of the image. While charcoal is the most commonly dated archaeological material, it dates the death of the plant from which the charcoal was made. It does not directly date the event of interest, in our case the production of a painting. It is plausible that wood was collected from a live plant and then burned to make charcoal paint; in this instance, a radiocarbon date would be contemporaneous with painting. However, two other situations can occur: old wood and old charcoal. The old wood effect occurs when wood may have died long before it was burned to make charcoal (Schiffer 1986). Alternately, for the case of old charcoal, wood could have been burned at an earlier unknown time and the resultant charcoal used much later to execute a painting (Bednarik 1994).

To a first approximation, our radiocarbon results for charcoal megalithic paintings are contemporary with published dates for monument construction phases and/or early activity at decorated sepulchral megaliths (Table 2). They are also consistent with a previously determined age of a black painting at Antelas, Viseu, Portugal. An average of primary monument construction and/or utilisation dates from Table 2 is 4840 BP with a sample standard deviation of 190 BP (approximately 4050-3000 cal BC). This average overlaps with the Antelas date (Cruz 1995 a,b) and six of our radiocarbon results (for Pedra Moura, Casota do Paramo, Pedra Cuberta, Forno dos Mouros and Mamoa do Monte dos Marxos M4 & M5) at 95 per cent probability (2 [sigma]). Our results suggest megalithic monument decoration occurred at approximately the same time as primary construction. An average of the Antelas date and our six overlapping radiocarbon dates on megalithic paints is 4930 BP with a sample standard deviation of 220 BP (approximately 4300-3100 cal BC). Our values are therefore consistent with the first occurrence of passage-graves at approximately 4000-3600 cal BC (Alonso & Bello Dieguez 1997: 514) and only slightly later than the oldest directly dated single chamber megalith at 5210 [+ or -] 50 BP (4230-3940 cal BC) (Cruz 1995a: 92).

From Table 1, the temporal agreement between Pedra da Moura and Pedra Cuberta paintings is significant because these two monuments are only 1 km from each other. They are similar in architectural design, decorative technique, and location in the landscape, though clearly different in dimensions. Dates of both monuments also overlap with construction of the passage grave of Dombate in Coruna (Alonso & Bello Dieguez 1997), which is about 11 km away.

Two monuments showed signs of multiple painting events. Radiocarbon ages of paint layers were consistent with the observed superposition. At Mamoa do Monte dos Marxos, the radiocarbon data establish this temporal stratification with an age of 4920 [+ or -] 60 BP for the top surface layer (M5) and 5330 [+ or -] 80 BP for the underlying paint layer (M4). Calibrated age ranges for both dates do not overlap at 2 [sigma]. The age differences in the two layers of paint may be a result of continuous, prolonged use of the monument, without significant transformations in its structural design. Or, alternatively, it may represent modification of a primary construction to create a larger or more complex structure, with repainting of the images during this later modification (Fabregas Valcarce 1995: 104). The painted designs of both layers appear to be notably similar, as if ancient restoration of a degraded painting had been intended.

At Coto dos Mouros, sample M10 contains two layers of paint and has an age of 3830 [+ or -] 60 BP, representing a weighted average related to the amount of each layer in the paint sample. Paint sample M11 with only the underlying layer yielded an age of 5540 [+ or -] 70 BE Interpretation of these radiocarbon results from Coto dos Mouros is complex, with results for M10 being unexpectedly young and M11 being quite old. This suggests that M10 has been subject to modern contamination, perhaps by carbon from vehicle tyres. Incorporation of contamination from tyre soot would result in a younger age if natural rubber was used to make the tyre. Alternatively, if the tyre was produced from synthetic rubber, the measured radiocarbon age would be older than the age of the paint due to the inclusion of [sup.14]C-free carbon from petroleum raw materials. For the underlying layer (M11), the 5540 BP result is older than expected, and like M4 from Monte dos Marxos could be explained by the use of old charcoal, perhaps coming from an earlier construction, for both dates are roughly consistent with those obtained from the earliest single chambers in north-west Iberia. In this respect, we must not forget that hidden under the tumulus of Dombate's passage grave there were the remains of another, much smaller, funerary chamber from an earlier building phase at the site.

From Cruz (1995a,b), Antelas is the only case where there is radiometric data for both the construction of a structure (5070 [+ or -] 65 BP) and the execution of its paintings (4655 [+ or -] 65 BP). When calibrated, these age ranges do not overlap at 2 [sigma] suggesting a temporal lag between construction and decoration with a 95 per cent probability that the ages represent different events. It is unknown whether the radiocarbon date for the painting at Antelas is from a single layer of paint or successive layers as we observed at Mamoa do Monte dos Marxos, Coto dos Mouros, and perhaps even Dombate and Pedra Cuberta.

Statistically, the two paint layers at Mamoa do Monte Dos Marxos and the Antelas construction and paint decoration dates indicate an interval between activities. However, there is danger in making such conclusions with only one radiocarbon date for each material studied. There is no way to determine if any of the measurements are rogue due to contamination or laboratory error.

Conclusions

These radiocarbon ages for megalithic paintings agree with the proposed time period for the north-west Iberia megalithic phenomenon (Cruz 1995a; Fabregas Valcarce 1995; Alonso & Bello Dieguez 1997; Bueno Ramirez & Balbln Behrmann 1997), and by doing so, reinforce this cultural chronology. Megalithic painting may have occurred over several centuries or may have taken place over a shorter phase. Because of vagaries in the radiocarbon calibration curve during this time period, the calibrated age range intervals are quite large and it is difficult to resolve differences in calendar ages for these monuments. It is noteworthy that paint samples from eight different monuments produce radiocarbon ages that are within 1000 [sup.14]C years of each other centring around 5000 years BP (3800 cal BC). Though meaningful, these radiocarbon results remain orphans if they are not integrated with other data from future scientific excavations of decorated megalithic monuments.

Table 1. Radiocarbon results for megalithic paint from Galicia and
North Portugal

Sample Site CAMS no. [micro] g C

M1 Pedra Moura 77761 50
M2 Casota do Paramo 77427 50
M3 Pedra Cuberta 77923 120
M4 Mamoa do Monte: 77924 90
 dos Marxos (under)
M5 Mamoa do Monte: 77925 210
 dos Marxos (surface)
M6 Forno dos Mouros 80501 110
M7 Anta de Serramo 88195 53
M8 (a) Dolmen de Dombate Lost --
M9 (b) Dolmen de Dombate Too small 8
 (red)
M10 Coto dos Mouros 83116 165
M11 Coto dos Mouros 83631 100
Cruz 1995a,b Antelas OxA-5433

 Cal BC Probability
Sample Years BP (2[sigma]) (%)

M1 4980 [+ or -] 70 3950-3640 95.4
M2 4740 [+ or -] 120 3800-3100 95.4
M3 5010 [+ or -] 60 3960-3690 92.8
 3680-3660 2.6
M4 5330 [+ or -] 80 4340-3980 95.4
M5 4920 [+ or -] 60 3940-3870 5.3
 3810-3630 88.2
 3560-3530 1.9
M6 4900 [+ or -] 60 3910-3870 1.8
 3800-3620 87.0
 3590-3520 6.6
M7 6050 [+ or -] 110 5300-4700 95.4
M8 (a) -- -- --
M9 (b) -- -- --
M10 3830 [+ or -] 60 2490-2130 94.3
 2080-2060 1.1
M11 5540 [+ or -] 70 4540-4240 95.4
Cruz 1995a,b 4655 [+ or -] 65 3650-3300 91.1
 3250-3100 4.3

(a) No radiocarbon result for this sample because the glass tube
containing plasma-extracted C[O.sub.2] cracked during sealing.

(b) No radiocarbon result for this sample, as there was
insufficient carbon for AMS dating.

Table 2. Radiocarbon dates for strata corresponding to primary
construction and utilisation of megaliths with pictorial decoration.
All dates are from Alonso & Bello Dieguez (1997), except the Antelas
date from Cruz (1995a) and the organic residue from a pot excavated
at Dombate reported here. Calibration was performed using the OxCal
program, version 3.5 (Bronk Ramsey 2000) with atmospheric data from
Stuiver et al. (1998)

Site Sample no. Years BP

Cha de Parada 3 Gif-8289 5070 [+ or -] 100
Antelas OxA-5498 5070 [+ or -] 65
Dombate Avg. of 2 dates 4918 [+ or -] 46
Dombate Avg. of 2 dates organic residue 4895 [+ or -] 30
Madorras 1 GrN-21066 4790 [+ or -] 60
Madorras 1 OxA-5199 4540 [+ or -] 65
Cha de Parada 1 CSIC-954 4820 [+ or -] 40
Cha de Parada 1 ICEN-173 4610 [+ or -] 45

Site Cal BC (2[sigma]) Probability (%)

Cha de Parada 3 4050-3600 95.4
Antelas 3980-3700 95.4
Dombate 3790-3630 95.4
Dombate 3715-3635 95.4
Madorras 1 3700-3490 79.4
 3460-3370 16.0
Madorras 1 3500-3430 6.0
 3380-3020 89.4
Cha de Parada 1 3700-3510 95.4
Cha de Parada 1 3550-3300 81.0
 3250-3100 14.4

Acknowledgements

This research was funded in part by National Science Foundation Archaeometry Grant No. 0209312 and the University of California Office of the President/CAMS minigrant program. Additional funding was from a FEDER-CICYT Research Project (Ref. 1FD97-0805-C02-01), co-financed by the Spanish Ministerio de Ciencia y Tecnologla and European Union funds: Procedimiento Interdisciplinar de caracterizacion, diagnosisy conservacion depintura megalitica. Radiocarbon analyses were performed under the auspices of the U.S. Department of Energy by the University of California Lawrence Livermore National Laboratory (contract W-7405-Eng-48).

References

ALONSO, F. & J.M. BELLO DIEGUEZ. 1997. Cronologia y periodizacion del fen6meno megalkico en Galicia a la luz de las dataciones por C-14, in A. Rodriguez Casal (ed.). O Neolitico Atlatico e as orixes do Megalitismo: 507-20. Santiago de Compostela: Actas do Coloquio Internacional, Abril 1996.

ARMITAGE, R.A., J.E. BRADY, A. COBB, J.R. SOUTHON & M.W. ROWE. 2001. Mass spectrometric radiocarbon dates from three rock paintings of known age. American Antiquity 66: 471-80.

BEDNARIK, R.G. 1994. Conceptual pitfalls in dating of Palaeolithic rock art. Prehistoire et Anthropologie Mediterraneennes 3: 95-102.

--2002. The dating of rock art: a critique. Journal of Archaeological Science 29:1213-33.

BELLO DIEGUEZ, J.M. 1995. Arquitectura, arte parietal y manifestaciones escultoricas en el megalitismo noroccidental, in F. Perez & L. Castro (ed.). Arqueoloxia earte na Galicia prehistorica e romana: 31-98. Museu Arqueoloxico e Historico de A Coruna.

BRONK RAMSEY, C.B. 2000. OxCalprogram, 3.5 edn. University of Oxford Radiocarbon Accelerator Unit.

BUENO RAMIREZ, P. & R. BALBIN BEHRMANN. 1997. Ambiente Funerario en la Sociedad Megalitica Iberica: Arte Megalitico Peninsular, in A. Rodriguez Casal (ed.). O Neolitico Atlantico e as orixes do Megalitismo: 693-718. Santiago de Compostela: Actas do Coloquio International, Abril 1996.

CARRERA RAMIREZ, F. & R. FABREGAS VALCARCE. 2003. The protection and management of the megalithic art of Galicia, Spain. Conservation and Management of Archaeological Sites 6: 23-37.

CHAFFEE, S.D., M. HYMAN & M.W. ROWE. 1994. Radiocarbon dating of rock paintings, in D.S. Whitley & L.L. Loendorf (ed.). New light on old art: recent advances in hunter-gatherer rock art research: 9-12. Berkeley/Los Angeles, CA: University of California Press.

CRUZ, D.J. 1995a. Cronologia dos monumentos con tumulus do Noroeste e da Beira Alta. Estudos Pre-historicos 3:81-119.

--1995b. Dolmen de Antelas. Un sepulcro-templo do Neolitico Final. Estudos Pre-historicos 3: 263-4.

FABREGAS VALCARCE, R. 1995. La realidad funeraria en el Noroeste del Neolitico a la Edad del Bronce, in C. Fernandez (ed.). Arqueoloxia da morte na Peninsula Iberica desde as Orixes ata o Medievo: 95-125. Excmo: Concello Xinzo de Limia.

HEDGES, R.E.M., C.B. RAMSEY, G.J. VAN KLINKEN, P.B. PETTITT, C. NIELSEN-MARCH, A. ETCHEGOYEN, J.O.F. NIELLO, M.T. BOSCHIN & A.M. LLAMAZARES. 1998. Methodological issues in the [sup.14]C dating of rock paintings. Radiocarbon 40: 35-44.

PACE, M.F.N., M. HYMAN, M.W. ROWE & J.R. SOUTHON. 2000. Chemical pretreatment on plasma-chemical extraction for [sup.14]C dating of Pecos river genre rock paintings, in F.G. Bock (ed.). American Indian Rock Art, vol. 24: 95-102. Tucson, AZ: American Rock Art Research Association.

PETTIT, P. & P. BAHN. 2003. Current problems in dating Palaeolithic cave art: Candamo and Chauvet. Antiquity 77: 134-41.

ROWE, M.W. 2001. Dating by AMS radiocarbon analysis, in D.S. Whitley (ed.). Handbook of rock art research: 139-66. Walnut Creek, CA: AltaMira Press.

ROWE, M.W. & K.L. STEELMAN. 2001. Radiocarbon dating using plasma-chemical extraction. American Laboratory 34: 15-9.

Russ, J., M. HYMAN & M.W. ROWE. 1992. Direct radiocarbon dating of rock art. Radiocarbon 34: 867-72.

SCHIFFER, M.B. 1986. Radiocarbon dating and the old-wood problem--the case of the Hohokam chronology. Journal of Archaeological Science 13: 13-30.

SHEE TWOHIG, E. 1981. The megalithic art of western Europe. Oxford: Clarendon Press.

STEELMAN, K.L. & M.W. ROWE. 2005. Dating pictographs: independent dates and their implications for rock art, in J.K.K. Huang & E.V. Culley (ed.) Making Marks: Graduate Studies in Rock Art Research at the New Millenium: 17-26. Tucson, AZ: American Rock Art Research Association.

STEELMAN, K.L., M.W. ROWE, V.N. SHIROKOV & J.R. SOUTHON. 2002. Radiocarbon dates for rock paintings in Ignatievskaya Cave, Russia: Holocene age for supposed Pleistocene fauna. Antiquity 76: 341-8.

STUIVER, M., P.J. REIMER, E. BARD, J.W. BECK, G.S. BURR, K.A. HUGHEN, B. KROMER, G. MCCORMAC, J. VAN DER PLICHT & M. SPURK. 1998. INTCAL98 radiocarbon age calibration, 24000-0 cal BP. Radiocarbon 40: 1041-83.

VALLADAS, H. & J. CLOTTES. 2003. Style, chauvet and radiocarbon. Antiquity 77: 142-5.

VAN DER MERWE, N.J., J. SEALY & R. YATES. 1987. First accelerator carbon-14 date for pigment from a rock painting. South African Journal of Science 83: 56-7.

Received: 24 November 2003; Accepted: 18 May 2004; Revised: 1 June 2004

K.L. Steelman (1), F. Carrera Ramirez (2), R. Fabregas Valcarce (3), T. Guilderson (4) & M.W. Rowe (5)

(1) Department of Chemistry, 205 Laney Hall, University of Central Arkansas, 201 Donaghey Avenue, Conway, AR 72035, USA (Email: ksteel@uca.edu)

(2) Escola Superior de Conservacion e Restauracion de Bens Culturais de Galicia, Rua Xeneral Martitegui s/n 36002 Pontevedra, Spain (Email: fernandocarrera@wanadoo.es)

(3)Universidade de Santiago de Compostela, Departamento de Historia 1, Facultade de Xeografia e Historia, 15782 Santiago de Compostela, Spain (Email: phfaoreg@usc.es)

(4) Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory, Livermore, CA 94551-9900, USA (Email: guilderson1@popeye.llnl.gov)

(5)Department of Chemistry, Texas A&M University, P.O. Box 30012, College Station, TX 77842-3012, USA (Email: rowe@mail.chem.tamu.edu)