INTRODUCTION

Artifacts made from organic materials carry important historical information. Their study is often complicated both by the state of preservation of the artifacts themselves and by destruction of their constituent organic compounds. The data obtained as a result of the studies of such artifacts can be of great value for archaeological research and reconstruction of the life and traditions of representatives of different cultures.

In 1988, the Tuva archaeological expedition of the Institute of the History of Material Culture, Russian Academy of Sciences, under the leadership of M.E. Kilunovskaya and Vl.A. Semenov, found a partially mummified burial in a wooden block in the mound 2 at the Saryg-Bulun burial ground (Central Tuva, Russia) that belonged to the Aldy-Bel culture and dated from the end of the 7th—beginning of the 6th centuries BC. A girl aged 6–7 years was buried in the grave 5 ([1], p. 11; [2]), despite the grave goods found here that are typical for most male burials, a wooden bow, a coin with a groove on a wooden handle, a quiver with arrows, a knife, and an inlaid belt were also found. The organic components of these objects, as well as the headdress and fur coat, were well preserved due to the lack of air access: the lid of the larch hollowed-out log was tightly closed at the time the find was discovered; the grave itself was covered with several layers of slabs closely adjacent to each other. It is noteworthy that along with the segmented bow, fragments of its winding were preserved: a dried golden substance, on which natural notches are visible (Fig. 1) and a large bundle of tendons, probably used to strengthen the arms of the bow upon firing.

Fig. 1.
figure 1

Bow from burial 5 of the Saryg-Bulun burial ground: 1 is the wooden base, 2 is a drawing of the wooden bow [3], 3 is the bow winding, 4 is a bundle of tendons (outer side) from the arms of the bow; 1, 3, and 4 are photos by A.A. Makeeva.

Full size image

Since the discovery of the finds, many researchers have made various assumptions about the nature of the bow winding. Even during the excavations, paleozoologist A.V. Gal’chenko (Altai State University, Barnaul) hypothesized that the found bow was covered with fish skin. Therefore, in all subsequent publications, this find was described as a tight-fitting bow made from burbot skin ([1], p. 394; [3]; [4], p. 17). In 2019 A.Ya. Mazina (The Grabar Art Conservation Center, Moscow) carried out the first technical and technological examination of this object in the laboratory of physical and chemical research, which showed that the winding contains degraded protein compounds ([1], p. 394).

At the same time, the appearance of the winding is presented as a dark and at the same time golden, fragile, and overdried substance with natural “notches,” close to the golden bark of a caragan (or caragana)—a plant that grows quite widely in the forest-steppe and mountainous regions of Tuva (lat. Caragana) and could be used for decorative purposes (Fig. 1). For example, in barrow 1 of the Olon-Kurin-Gol 10 burial ground (Mongolia, Pazyryk culture), a Scythian-era bow was found with decorative ornamentation with thin “boiled” birch bark ([5], p. 466, Fig. 36). In the Arzhan-2 mound (Uyuk Basin, Tuva, 7th century BC), the “king’s” bow, judging by the preserved fragments, was a composite, i.e., made of wooden planks wrapped in birch bark, under which an additional fibrous organic layer was recorded ([6], p. 50). According to the reconstruction, this was precisely a “Scythian” bow, i.e., short recursive with a reflex handle ([7], p. 220). It was longer, 117 cm, but the length of the bowstring was 95 cm.

The remains of a bow were also discovered during excavations in Mongun-Taiga at the Kholash burial ground, mound 83. A wooden bow of ~127–130 cm long was found there; its body was covered with fish skin, which was best preserved at the ends (up to 4 cm wide). The Tuva expedition also found fragments of strips similar to those described above at the Suglug-Khem 1 burial ground, dating back to the 2nd century BC ([4], p. 17).

To identify the nature of the components that make up archaeological finds, various physical and chemical methods have been widely used recently: infrared (IR) spectroscopy, atomic spectroscopy, mass spectrometry (MS), gas and high-performance liquid chromatography with mass-spectrometric detectors (GC/MS and HPLC/MS) [8–16].

The purpose of the presented study was to determine the nature of the wrapping material for the shoulders of an ancient bow (late 7th–early 6th centuries BC) ([1], p. 404) from burial 5 of the Saryg-Bulun burial ground in Central Tuva, as well as to confirm or refute the hypothesis of the bioorganic origin of rare facing material.

EXPERIMENTAL

Research on the material of the winding of the arms of an ancient bow from burial 5 of the Saryg-Bulun burial ground in Central Tuva took place in several stages and was based on two hypotheses: plant or animal origin of the winding material. The first stage of the study was a comparative study of the IR spectra of a sample of ancient bow winding and samples of the bark of modern caragana from the steppe regions of Tuva. To test the hypothesis of animal (bio-organic) origin of the winding material of the ancient bow, investigation of the solubility of the test sample in aqueous alkali, analysis of the protein content by the Bredsted method, and analysis of the amino-acid composition (AAC) of collagen proteins by the method of thin layer chromatography (TLC) were carried out. To test the hypothesis of using fish skin, the composition of lipid substances in the extract of the bow-winding material was analyzed using GC/MS.

The IR spectra were recorded using a Nicolet iS50 Fourier-transform IR spectrometer (Thermo Fischer Scientific, USA) in the attenuated total internal reflection mode (diamond crystal, KBr beam splitter) in the range of 4000–400 cm–1, number of scans 64, resolution 4 cm–1. The study was carried out without any additional sample preparation.

Determination of the AAC composition of the ancient bow-arm-winding material was carried out by TLC on aluminum plates coated with silica gel (TLC Silicagel 60 plates (10 × 20 mm)). For this, the sample solution obtained by studying the solubility in aqueous alkali was diluted twice with water and acidified to pH 7.5 with dilute hydrochloric acid. To separate the AAC zones, a solvent system was used: n-butanol–glacial acetic acid–water (4 : 1 : 1). A sample of the test solution was applied to the plates using a 25-μL microsyringe so that the diameter of the spot did not exceed 4–5 mm, and the center of the spot was on the starting line. In parallel, 5 μL of a mixture of 0.1% aqueous solutions of standard samples of glycine and valine was applied to the starting line of the plate. The development of AAC chromatographic zones on the plates was carried out by treatment with a 1% solution of ninhydrin in ethanol and subsequent heating in an oven at a temperature of 105°C for 3–5 min. On the chromatogram of the test solution, spots were marked by position corresponding to the spots on the chromatogram of the solution of standard AAC samples, and their color intensities were compared.

Analysis of the composition of lipid substances in the extract of the ancient-bow-winding material samples was carried out using GC/MS. For this, 2 mL of a 3% solution of sulfuric acid in methanol was added to ~10 mg of the winding sample material under study and treated in an ultrasonic bath (80°C for 3 h). After cooling to room temperature, 2 mL of water and 5 mL of ether were added to the reaction mixture and shaken vigorously for 10 min. The layers were separated, the upper ether layer was separated and evaporated at room temperature to dry residue. The residue was dissolved in 50 μL of n-hexane and examined by GC/MS.

GC/MS analysis was carried out using an HP-6890 chromatograph with an MSD 5975 MS detector from Agilent Technologies. The chromatography conditions were: capillary column HP-5ms with a length of 30 m and internal diameter of 0.25 mm, and a stationary-phase film thickness of 0.25 µm. The initial column temperature is 80°C (holding time 4 min); the temperature increases from 80 to 280°C at a rate of 5°C/min. Holding at the final temperature is carried out for 10 min. The carrier gas is helium at 1 mL/min with a split ratio of 1 : 10. The evaporator temperature is 280°C and the detector interface temperature is 280°C. The sample volume is 1 µL. Detection was carried out by electron ionization in the scanning mode using the total ion current in the range of 50–900 m/z. The scanning rate is 1.76 scan/s, the ionization energy is 70 eV, and the temperature of the quadrupole and the ion source is 150 and 230°C.

RESULTS AND DISCUSSION

The IR spectra of the samples are shown in Fig. 2a. In the IR spectrum of the winding of an ancient bow (Fig. 2a, curve 1), bands corresponding to stretching vibrations of N–H bonds (3280 cm–1), Amide I bands (C=O stretching vibration, 1633 cm–1), and Amide II bands are observed (N–H bending vibrations and C‒N stretching vibrations, 1542 cm–1) and its overtone at 3070 cm–1, Amide III (N–H bending vibrations + C=O bending vibrations + C–C stretching vibrations, 1230 cm–1). Such absorption bands are characteristic of nitrogen-containing compounds, including collagen, a protein that is a key component of animal and fish skins [8, 9], and are absent in the components (such as cellulose, lignin, etc.) that make up plant cells, including caragana (Fig. 2a, curves 2, 3). We note that the spectra of the outer and inner layers of the modern caragana sample differ significantly, which indicates the difference in their chemical composition. In addition, absorption bands typical of most organic compounds are observed, corresponding to stretching and bending vibrations of C–H bonds: (2926, 2852 and 1447, 1409 cm–1, respectively). On the other hand, the sample contains an extremely intense absorption band with a maximum at 1030 cm–1, which is related to the stretching vibrations of C–O bonds and typical of plant raw materials [10]. The white inclusions present in the sample (Fig. 2a, curve 4) are gypsum (CaSO4⋅2H2O) and have characteristic bands at 3528, 3399, 1683, 1620 (vibrations of bound water), 1109, 672, and 599 cm–1 (vibrations of the sulfate anion) [11], in the spectrum of the winding, they are presented in the form of shoulders in more intense bands.

Fig. 2.
figure 2

IR spectra of the samples: 1 is the ancient bow winding, 2 and 3 are the bark of a modern caragana bush (outer and inner layer, respectively), 4 is white inclusions on the sample (gypsum) (a); GC/MS chromatogram of the winding extract (b).

Full size image

We note that the specifics of the sample, i.e., the winding of the bow, which is separated from the base, does not allow one to unambiguously determine from the IR spectra data, whether it is a plant (Caragana bush) with some kind of treatment with animal glue or fish skin, on which fragments of wood from the bow itself have stuck. Therefore, at the next stage of research, the GC/MS method was used.

To analyze the possible animal origin of the sample under study, its solubility in aqueous alkali was studied. 0.5 mL of 4.5% solution of KOH in water was added to the winding sample (~10 mg), and the solution immediately began to turn brown. After treatment in an ultrasonic bath (60°C, 20 min), the bow-winding sample dissolved completely. A sample of Caragana bush bark, treated in a similar way, did not dissolve, which confirmed the protein nature of the ancient material being studied.

The alkaline solution of the winding sample gave a positive qualitative Biuret reaction to the peptide bond with a copper-sulfate solution. The quantitative content of total protein was determined using the Bredsted method, which was ~13% (in terms of the weight of the sample). Consequently, the winding material of the ancient bow was of protein origin [17].

It is known [17] that the AAC composition of fish proteins has some specific features compared to the proteins in the meat of warm-blooded animals and birds. Thus, in [18], the results of a comparative study of the compositions of AAC collagen proteins of fish and animal origin are presented. The AAC compositions of 2% alkaline solutions of animal (cattle) and fish (silver carp) skin were studied. It was found that the main differences between animal and fish skin are observed in the relative content of two AACs (glycine and valine): in fish skin, the glycine/valine content ratio is 1, and in animal skin, it is 12.5.

Therefore, the results of a comparison of chromatograms from the winding and standard AAC samples, which showed that the glycine/valine content ratio is ~1, allow one to conclude that the winding material under study is fish skin.

According to GC/MS data, saturated fatty acids (C8–C18) and unsaturated oleic acid (C18:1) were found in the bow-winding extracts (Table 1). In addition, dicarboxylic acids (DC-4–DC-9), with a predominance of azelaic acid (DC-9), as well as hydroxy acids (C18:2-OH and C18:1-OH), which are products of the degradation of unsaturated fatty acids, were found. The chromatogram of the winding sample extract is shown in Fig. 2b.

Table 1. Identification of the main compounds by GC-MS in the studied extract of the bow-winding sample
Full size table

The results of identifying the peaks of the main compounds are presented in Table 1.

Interestingly, succinic acid was found in the extract under study. As was shown in [19], succinic acid is formed during the decomposition of only one fatty acid, i.e., docosahexaenoic acid, over time. This acid is present in the lipids of fish oil, is polyunsaturated, and decomposes over time. The decomposition of docosahexaenoic acid produces succinic acid, which is more stable and therefore, can serve as a biomarker for fish-product residues [19].

Thus, the research carried out gives every reason to assert that the material for winding the arms of the ancient bow was fish skin.

It is known that in China, composite bows were strengthened with fish skin. The Jurchen people in the 10th–15th centuries used fish glue, and the Mongols, when creating compound bows, attached to the outer side of the kibiti arc “a material capable of withstanding greater tension than wood. This layer took on the load and reduced the deformation of wood fibers” ([20], p. 229). They used tendons that run along the back of large ungulates, which could withstand tension up to 20 kg/mm ([20], p. 230).

Let us note that archaeological objects made of fish skin are practically not found in the burials of Tuva, but it is possible that this is due to the fact that this kind of research of organic materials on the archaeological finds of this region has not been carried out.

CONCLUSIONS

A step-by-step test of two hypotheses about the origin of the bow-winding material from the Saryg-Bulun burial ground made it possible to determine that the winding was made of fish skin, which confirmed the assumption of paleozoologist A.V. Galchenko, made during excavations in 1988.

The bow from burial 5 of the Saryg-Bulun burial ground can be classified as simple one. It has a segment-shaped cross section. Its total length is 100 cm and width is 4 cm. There are cutouts at the ends for attaching a bowstring. Its shaft consists of several layers, but the base itself is solid, consisting of one piece of wood (birch?). A wide strip of tendons from a large ungulate, preserved along with the winding, could have been used to strengthen the bow. However, an extremely dry microclimate is needed for this kind of strengthening. Perhaps this is why the additional use of fish skin made it possible to waterproof the wooden base and tendons.

The study of complex weapons is of great importance for understanding the culture of the nomads of the Sayano-Altai region nomads of the Paleometal era. Well-preserved bows are very rare, so a detailed study of each specimen allows one to better reconstruct ancient production technologies and pay attention to the technological techniques used by ancient craftsmen. As one can see, birch bark and fish skin could be used to wrap bows. The choice of material was probably determined not only by production needs or natural factors, but primarily by cultural symbols that had connections with everyday phenomena, historical events, folklore, and beliefs of the nomads of the Sayano-Altai.