Radiocarbon dates for pictographs in Ignatievskaya Cave, Russia: Holocene age for supposed Pleistocene fauna. (Notes & News).

Steelman, K.L. ; Rowe, M.W. ; Shirokov, V.N. 等

Key-words: pictographs, AMS radiocarbon dating, Russia, Bronze/Iron Age rock art, Holocene, Palaeolithic rock art, Pleistocene fauna

Ignatievskaya Cave is located in the northwestern foothills of the southern Ural Mountains. It is on the right bank of the Sim River, a tributary of the Belaya River (FIGURE 1). Foothills there are ~200-700 m high. Average temperature is 0.1 [+ or -] 1[degrees]C with 500-700 mm of precipitation per year. The area is a coniferous forest, mixed with deciduous trees (FIGURE 2). Pine, fir, birch, alder and oak are most typical. Large grasses, ferns and cereal plants are present. There is rich, black topsoil over a carbonate (limestone) base.

[FIGURES 1-2 OMITTED]

The largest cave in this karst region, with 600 m of Devonian limestone passages, Ignatievskaya Cave is ~3 m wide and ~2 m high with a flat horizontal ceiling, with 4 main chambers (FIGURE 3): Entrance Hall, Main Corridor, Large and Far Halls--the last two connected by two narrow tubes. Cave temperature is nearly constant at 5[degrees]C.

[FIGURE 3 OMITTED]

Although Ignatievskaya Cave has long been known by local residents, ancient red ochre and black charcoal prehistoric images were discovered by archaeologists V.T. Petrin, S.E. Chairkin & V.N. Shirokov only in 1980. The site was studied on field trips from 1980 to 1986 by Petrin et al. (1992) and in 1995 by Scelinskij & Shirokov (1999). Over 50 pictographs were found and recorded during those studies.

Paintings are located only in the Large and Far Halls at ~120 m from the entrance. There are no prehistoric drawings in the Entrance Hall. This site is a good example of image integration into naturally occurring rock morphology. In Large Hall, pictographs were placed on vertical and sloping walls and ceiling, as well as in niches and depressions. Images are also located on a large pillar. In contrast, most pictographs in Far Hall are painted on the ceiling, with only a few on the walls.

Description of pictographic images

Drawings vary from 1.5 cm for symbols to as large as 2.3 m for animals and anthropomorphs. Lines are 1 to 5 cm wide. Paintings are two colours: shades of red (iron ochre) and black (charcoal). In Far Hall, black figures dominate; in Large Hall, red paintings are more prominent. There are only three black paintings in Large Hall.

Images are animals, symbols, anthropomorphs and indefinite figurative motifs. Representations of `mammoths' (6 or 7) and horses (4) constitute a majority of animal images. There are, however, paintings of an ox, a rhinoceros-like animal, a composite animal with a camel-like body and a mammoth-like fanciful creature. All animals (20-30 cm long) in Large Hall were sketched as shaded silhouettes, with minimal features depicted. Some `mammoth' drawings do not show a neck and only one has tusks. A horse image is sketchily drawn without a mane or ears (FIGURE 4). Animals have straight legs, and at times only two are depicted.

[FIGURE 4 OMITTED]

Unlike Palaeolithic rock art of animals in France and Spain, images here are ambiguous; identification is problematic. Straight legs make images of black animals rather static. Motion is only suggested in one `mammoth' image that has bent knee joints. Another `mammoth' (FIGURE 5) apparently has a characteristic high-domed head, a trunk and possible tusk(s). But these assignments are not definitive. The `camel' body of a composite animal is drawn with only three or four lines. Black animals (0.3 to 1.3 m) are mostly outline drawings. A red rhinoceros has four legs, roughly sketched, as on other animals. Its body was drawn in outline with partial shading, and is ~2.3 m long.

[FIGURE 5 OMITTED]

Anthropomorpic images in Far Hall are represented by a black male figure ~32 cm long and a red female figure (FIGURE 6) ~1.2 m long. These anthropomorphs form a complex composition with a large figure of a rhinoceros. There is a black drawing of a `phantom'. A red motif (20 cm tall) in Large Hall is characterized as a combined human and aviform. All these figures are outlines.

[FIGURE 6 OMITTED]

Numerous signs include: single lines and groups, dots, meanders, arrow-shaped forms, tridents, triangles, extended parallelograms, cross- and anchor- shaped motifs, as well as patches of colour. One ring-shaped motif looks like a rhinoceros footprint. Signs outnumber human and animal figures threefold, varying from 1.5 cm to >1 m.

Previous investigations at Ignatievskaya Cave

Archaeological excavation undertaken in the cave uncovered an Upper Palaeolithic level. Its sediments contained fragments of torches, charcoal, stone and bone artefacts, as well as pieces of ochre, ornaments and animal bones. More than 1300 stone tools were unearthed, mostly near the cave entrance. They were made from different types of stone: jasper, flint, quartzite, sandstone, limestone, etc. Scrapers, retouched flakes, burins, scaled pieces and notched denticulate forms dominate artefacts. There are blades with blunted edges, including some worked at the ends. These could have formed part of a composite tool with a bone, horn or wooden shaft. Two pendants were made from the canine of a polar fox and a tooth of an ox or bison. Two tiny round flat beads were made from bone or tusk.

Three radiocarbon dates obtained from charcoal from Large Hall place human activity in Ignatievskaya Cave at Late Pleistocene: 14,240 [+ or -] 150 BP (Laboratory of Siberian Branch of Russian Academy of Sciences (LSBRAS), 2209); 13,335 [+ or -] 195 BP (Institute of Evolutional Morphology and Ecology of Animals, 365); 10,400 [+ or -] 465 BP (LSBRAS, 2468). One of us (VNS) assumed that some images were of Pleistocene age. Aside from depiction of `mammoths' and `camels', there is no direct evidence to connect painted images with Pleistocene occupation of the site. Paintings could be contemporary with more recent archaeological finds within or outside the cave. Although Upper Palaeolithic finds occur near the cave entrance, there were few artefacts from the Bronze (early 2nd to mid 1st millennia BC) and Iron (mid 1st millennium BC to late 1st millennium AD) Ages here, and no ancient drawings.

Experimental procedure

Sample collection

During August 1998, we removed small samples of charcoal pigment from three pictographs and a red iron oxide pigment from a fourth with scalpels while wearing rubber gloves. Samples, including underlying rock and accretionary mineral matter, were placed on aluminium foil, wrapped and stored in plastic bags. Backgrounds, adjacent rock with no pigment, were collected and analysed. These went to Texas A&M University (hereafter TAMU). Each sample was examined with an optical microscope to ensure that no extraneous material was included in the sample. None was found.

Chemical pretreatment

Again, rubber gloves were worn to avoid contamination during all sample handling. Procedures for chemically pretreating archaeological charcoal vary slightly from laboratory to laboratory; all involve treatment with acid-base-acid (e.g. Bowman 1990; Taylor 1987). We routinely eliminate both acid washes as we have shown them to be unnecessary with plasma-chemical extraction (Pace et al. 2000). Carbonate and oxalate carbon are not extracted by plasma; only organic material was removed to be analysed by radiocarbon measurements (e.g. Russ et al. 1992; Chaffee et al. 1994).

To remove potential humic acid contamination, we immersed each sample in ~5 ml of 1 M NaOH and placed them in an ultrasound bath for an hour at 50 [+ or -] 5[degrees] C. After NaOH wash/ ultrasonication treatments, when the resulting supernatant was coloured, subsequent NaOH washes were necessary because humic acids appear brownish-yellow in NaOH. After supernatant was colourless, one additional treatment with NaOH was done to ensure humic acid removal. Adsorbed C[O.sub.2] and water from air are removed by vacuum and plasma treatments. Plasma extraction was performed on dried filtrate.

Plasma-chemical treatment

The plasma-chemical method used to extract organic carbon from these pictographs was described elsewhere (e.g. Russ et al. 1990; Hyman & Rowe 1997). Ultra-high purity bottled argon and oxygen (99.999%) were used for plasmas. Rotary pumps are sufficient to maintain vacuum ([~10.sup.-4] torr). First, a low-temperature ([less than or equal to] 150[degrees]C), low-pressure (~1 torr) oxygen plasma was used to preclean the reaction chamber; this was repeated until [less than or equal to] 0.001 mg carbon, as C[O.sub.2], was released. Once the chamber was clean, samples were introduced via a copper-gasketed stainless steel flange port under a flow of argon to prevent atmospheric C[O.sub.2], aerosols or organic particles from entering the system. After the chamber was resealed and the sample degassed under vacuum and heat, low-temperature argon plasmas were used to desorb C[O.sub.2] molecules from the sample and chamber walls by inelastic collisions of the nonreactive, but energetic, argon species. Adsorbed C[O.sub.2] on the samples was reduced to [less than or equal to] 0.001 mg of carbon by vacuum pumping.

Next, an oxygen plasma oxidized charcoal in black pigment samples and unknown organic material in the iron oxide pigment to C[O.sub.2]. Decomposition of inorganic carbon present (limestone and calcium oxalate) was prevented by plasma conditions. Oxidising plasmas react only with organic carbon in the samples, leaving substrate rock and accretionary carbonates and oxalates intact (Russ et al. 1992; Chaffee et al. 1994). Carbon dioxide from samples was flame-sealed into glass tubes cooled to liquid nitrogen temperature (-194[degrees]C) after water was frozen out with a dry-ice/ethanol slurry (-58[degrees]C), and finally sent for radiocarbon analysis to the Center for Accelerator Mass Spectrometry at the Lawrence Livermore National Laboratory (CAMS).

Results

TABLE 1 shows results of plasma extraction for four Ignatievskaya Cave samples and the three radiocarbon dates obtained. Carbon extracted from the red woman pictograph was insufficient to obtain a radiocarbon date. Since the relationship between radiocarbon years and calendar years is not linear, calibration is necessary in order to assess the actual calendar date of any archaeological sample. Therefore, the 2[sigma] (95.4% confidence level) calibrated range using the OxCal Program, version 3.5, calibration is included with uncalibrated ages (Ramsey 2000; Stuiver et al. 1998). The calibration curve for sample 4RU292 is shown in FIGURE 7.

[FIGURE 7 OMITTED]

Discussion

These [sup.14]C dates for pictographs at Ignatievskaya Cave are more recent than was generally supposed by Russian archaeologists (Petrin & Shirokov 1990; Petrin 1992; Scelinskij & Shirokov 1999). Excavations and depictions of extinct animals such as those interpreted as mammoths suggested Palaeolithic antiquity. Although other evidence at this site must be considered, these pictograph dates cannot be ignored, especially because of previous doubts about Palaeolithic antiquity for drawings at Ignatievskaya Cave (Formozov 1999; 2000).

Absence of contamination

Dates obtained for the drawings are more recent than was expected for `mammoth' images. Could soot, containing modern carbon, have contaminated the samples? Ignatievskaya Cave was named after Ignatij, a hermit, who lived there in the late 18th and early 19th centuries. After his death, religious processions took place there every year until the 1930s. Inside the cave, people used torches and burned other objects such as tyres. Due to its accessibility, numerous tourists, also using torches, visited it. Some walls and the ceiling appear to be covered by soot. In our (KLS & MWR) experience, what appears to be soot is often a black mineral deposit (sometimes calcium oxalate). Similarly, potential presence of lichen, fungi, bacteria and general wind-blown organic matter, if present in significant amounts, could cause a falsely young age. We saw no indication of lichen or fungi contamination with microscopic examination.

To assess potential contaminations, especially bacterial and wind-blown materials, we routinely take background samples from unpainted rock adjacent to pigment samples. That process takes on an important role here because resultant dates were much younger than expected, at least for the representation of a `mammoth'. The amount of carbon extracted from a background is used to estimate contamination in unpainted rock, and hence of sample containing pigment. Relative sizes of a pigment-containing sample and its background are considered when necessary. Without a background, potentially devastating contamination in a sample may not be taken into account, rendering a radiocarbon date untrustworthy.

Less than 0.002 mg of carbon was extracted from Ignatievskaya Cave background samples for the dated images, indicating that contamination is negligible. We calculate here the percentage modern (1950) contamination that would be necessary to skew expected ages (according to an assumed mammoth extinction time for the area of 12,000 years ago) at Ignatievskaya Cave to the radiocarbon ages measured. We used activity half-life and contamination equations from Bowman (1990:27-28): A = [A.sub.0][e.sup.-t/8033] and [A.sub.m] = f[A.sub.x] + (1-f)[A.sub.s], where A = [sup.14]C activity at time t, [A.sub.0] = modern [sup.14]C activity, f = fraction of contamination, [A.sub.m] = measured [sup.14]C activity, [A.sub.x] = contaminant [sup.14]C activity, and [A.sub.s] = `true' sample [sup.14]C activity.

Results of these calculations are summarized in TABLE 2. For the `mammoth' pictograph, 23% contamination with modern (1950) carbon is required to obtain a `true' age of 12,000 years BP. This level of contamination was not observed in backgrounds of these samples. In addition, this level of contamination from lichen or fungi would have been visible during microscopic examination. Thus, the potential problems listed above are not the cause of the unexpectedly recent dates obtained, unless bacterial growth occurred preferentially in the paint. Bacterial growth associated with organic material added during paint manufacture would be expected to approximate the age of the painting itself.

Evidence for Palaeolithic antiquity

The strongest evidence for a Palaeolithic age of Ignatievskaya images is the depiction of supposed Pleistocene animals. `Mammoths' and other Palaeolithic fauna predominate over depictions of other animals in both halls. Other Ice Age creatures, namely horses, a rhinoceros-like animal and ancient oxen, are also depicted in Ignatievskaya Cave. The image of a composite animal with a `camel' body is interesting, especially when one realizes that bones of camels are known from Pleistocene deposits in the Ural Mountain region. This set of extinct species differs from animals represented at open-air rock-art sites in this region. Such sites contain reproductions of elk, deer, roe, water-birds, occasional drawings of bear and undetermined figures of small animals. One of us (VNS) surveyed and studied all ~70 known rock and cave-wall art sites of the Ural Mountains.

In support of Palaeolithic ages, mammoths apparently disappeared in the southern Ural Mountains by the Late Pleistocene. Radiocarbon dates from remains of mammoths indicate these animals were extinct 12-14,000 years BP (Sulergitsky 1995). In Large Hall, in the area of drawings, only Upper Palaeolithic artefacts have been found. Therefore, the repertoire of images from Ignatievskaya Cave was thought to represent an Upper Palaeolithic age (Kosintcev 1990: 171-8). However, there is little stratigraphic evidence of relative contemporaneity of mammoth remains with Upper Palaeolithic drawings of Ignatievskaya Cave. During excavation of friable deposits close to a red horse panel (group 23 in Petrin 1992), there is a tiny seam of a white earthy incrustation of unknown mineralogy. Numerous pieces of charcoal and ochre, as well as stone tools and a pendant made from a polar fox canine, were unearthed beneath it. Polar fox, as a species, inhabited the South Ural Mountains only in the Pleistocene (Kosintcev 1999). Since white incrustation covers some art motifs located on the cave walls, it seems plausible that some images in Ignatievskaya Cave are at least 10,000 years old (Kosintcev 1990: 171-8; Petrin & Shirokov 1990; Petrin 1992; Scelinskij & Shirokov 1999).

Evidence for Bronze Age images

Formozov (2000) stated that although some images at Ignatievskaya Cave might be Palaeolithic depictions, others belong to later time periods. Formozov supports a later date for paintings at Ignatievskaya Cave from artefacts of the Bronze Age (level B of Formozov 1998; 2000) and Iron Age (level C) that were found at the site in addition to those of the Upper Palaeolithic (level A).

The repertoire of images and symbols in Ignatievskaya Cave and open-air rock-art sites of the post-Palaeolithic period are very different. Among known anthropomorphic drawings of rock art in the open air, there are no images with female features like the red female figure in Ignatievskaya Cave. In general, most open-air rock-art sites were created from the Late Neolithic to Bronze Age periods, with some even dating to the Iron Age (Chernetsov 1964; Formozov 1969; Shirokov et al. 2000). However, direct dating methods should be applied to have reliable dates for these open-air sites.

Conclusions

Although there was evidence that some paintings in Ignetievskaya Cave were of Palaeolithic age, in particular depictions of megafauna such as mammoths, three radiocarbon dates (~6000 to ~8000 BP) place images into the early Holocene. We have no reason to suspect the pictograph radiocarbon dates presented here. We have shown that modern backgrounds are far too low to cause movement of a true age of 12,000 BP to appear as young as 7370 BP. To bring our ages from the 6000-8000 range up to 12,000 BP would require almost 25% contamination by modern (1950) carbon. That is larger by two orders of magnitude than we have seen here. Analysis of unpainted rock as background eliminates soot from torches as significant contributing factors, as well as other potential contaminants discussed above. Humic acids were removed with NaOH washes as is routine in radiocarbon dating of charcoal. It is nonetheless surprising that the `mammoth' image produced such a recent date (7370 [+ or -] 50 BP). Two commonplace effects that must always be considered for radiocarbon dates on charcoal, `old wood' and `fossil charcoal', if important here, would move the dates in the opposite direction to that observed in this work (Schiffer 1986; Bednarik 1994). That is, the age determined would appear older than expected, not younger as observed.

These radiocarbon dates indicate that either: 1 the `mammoth' image (FIGURE 5) may not actually be of a mammoth;

2 mammoth existence extended into the Holocene in the area, although there is no dating evidence for that;

3 the depiction was not based on a live mammoth; or

4 there was repainting of an older image, such that younger overpainting and original painting charcoal combines to give a meaningless radiocarbon age, more recent than the initial painting event.

Acknowledgements. This research was funded in part by the Office of the Vice-President for Research and Associate Provost for Graduate Studies with a Regents Fellowship and with a travel grant from the International Research Travel Assistance Grant, both from Texas A&M University. Lawrence Livermore National Laboratory is funded by the US Department of Energy under contract W-7405-Eng-48. VNS' text was translated from Russian into English by S. Sharapova. Elena Miklashevich and Paul Bahn also helped with translation and information of various kinds; this manuscript was significantly improved through their efforts. Jason Ott and Ruth Ann Armitage assisted in plasma measurements. Photographs of pictographs were enhanced with Adobe Photoshop by Robert Mark, Rupestrian Cyber Services.

TABLE 1. Results of plasma oxidation and the radiocarbon dates
obtained for Ignetievskaya Cave samples. The `mammoth' dated is
shown in FIGURE 5.

motif colour TAMU# carbon CAMS #

`Mammoth' black 4RU292 0.10 mg 56586
radiating lines black 4RU291 0.21 mg 56271
lines black 4RU318 0.07 mg 67688
woman red 4RU330 <0.01 mg --

motif [sup.14]C age 2[sigma] age
 (BP) range (cal BC)

`Mammoth' 7370 [+ or -] 50 6390-6080
radiating lines 7920 [+ or -] 60 7040-6640
lines 6030 [+ or -] 100 5300-4650
woman -- --
TABLE 2. Calculations showing the percent contamination of modern
carbon required to alter these measured radiocarbon dates to the
archaeologically expected ages.

observed expected [A.sub.m] [A.sub.s] [A.sub.x] %
date date contamination

7370 12,000 0.3995 0.2245 1 23
7920 12,000 0.3731 0.2245 1 19
6030 12,000 0.4721 0.2245 1 32

References

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

BOWMAN, S. 1990. Interpreting the past: radiocarbon dating. Los Angeles (CA): University of California Press.

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

CHERNETSOV, V.N. 1964. Naskalnye Izobrajenia Urala. Chast' 1. M.: On je, chast 2, 1971, Moscow.

FORMOZOV, A.A. 1969. Ocherki po pervobytnomu iskusstvu. Moscow.

1998. O svyatilistche v Ignatievskoy pestchere na Urale // Politropon. K cemidesyatyletiu Vladimira Nickolaenicha Toporova. M., pp. 871-874.

2000. O datirovke rospicey v Ignatievskoy pestchere na Ugnom Urale // Rossiyskaya arheologia, No. 1: .215-217.

HYMAN, M. & M.W. ROWE. 1997. Plasma-chemical extraction and AMS radiocarbon dating of rock paintings, in S.M. Freers (ed.), American Indian Rock Art 23: 1-9. San Miguel (CA): American Rock Art Research Association.

KOSINTCEV, P.A. 1990. Istoria razvitia fauny krupnyh mlekopitaustchih // Istroricheskaya ekologia givotnyh gor Ugnogo Urala. Cverdlovsk, Russia.

1999. Formation and evolution of the Holocene fauna of megamammals in the Urals and West Siberia, in N. Benecke (ed.), Archaologie in Eurasia 6: 133-9.

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 24: 95-102. San Miguel (CA): American Rock Art Research Association.

PETRIN, V.T. 1992. Paleoliticheskoe svyatilistche v Ignatievskoy pestchere na Ujnom Urale. Novosibirsk.

PETRIN, V.T. & V.N. SHIROKOV. 1990. Naskalnye izobragenia Ugnouralskogo regiona (itogy i perspektivy issledovany). Problemy izuchenia naskalnyh izobrageny v SSSR. M.: 73-8.

RAMSEY, C.B. 2000. Manual for OxCal Program Version 3.5, University of Oxford Radiocarbon Accelerator Unit. www.rlaha.ox.ac.uk/oxcal/

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

Russ, J., M. HYMAN, H.J. SHAFER & M.W. ROWE. 1990. Radiocarbon dating of prehistoric rock paintings by selective oxidation of organic carbon, Nature 348: 710-11.

SCELINSKIJ, V.E. & V.N. SHIROKOV. 1999. Hohlenmalerei im Ural. Kapova und Ignatievka. Die altsteineitlichen Bilderhohlen im sudlichen Ural. Thorbecke, Germany.

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

SHIROKOV, V.N., S.E. CHAIRKIN & U.P. CHEMIAKIN. 2000. Uralskie pisanitsy. Reka Nejva. Ekaterinburg, Russia.

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.

SULERGITSKY, L.D. 1995. Cherty radiouglerodnoy hronologii mamontov (mammuthus primigenius) Cybiri i Cevera Vostochnoy Evropy // Issledovania po Pleistotsenovym i Sovremennym Mlekopitaustchim. SPb.,1995, T. 263; pp.163-183.

TAYLOR, R.E. 1987. Radiocarbon dating: an archaeological perspective. New York (NY): Academic Press.

K.L. STEELMAN, M.W. ROWE, V.N. SHIROKOV & J.R. SOUTHON *

* Steelman & Rowe, Department of Chemistry, Texas A&M University, PO Box 300012, College Station TX 77842-3012, USA. steelman@mail.chem.tamu.edu rowe@mail.chem.tamu.edu Shirokov, Institute of History & Archaeology, Urals Branch of Russian Academy of Sciences, 56 Luxemburg Street, Ekaterinburg 620026, Russia. istor@uran.ru Southon, Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory, Livermore CA 94551-9900, USA. Southon1@llnl.gov

Received 22 March 2001, accepted 5 November 2001, revised 6 November 2001