Stone and Osseous Adornments in the Mesolithic and Early Neolithic of the Iron Gates

3.1 Shell

The Mesolithic communities of the Iron Gates (Figure 1b and Table 1) used locally available aquatic resources especially Lithoglyphus naticoides shells, which were turned into adornments through the perforation of the last whorl. According to Păunescu’s field notes, at Cuina Turcului all perforated Lithoglyphus naticoides shells were discovered during the 1965 excavation season and came from the same (“Epipalaeolithic II”) level, suggesting that these shells were part of a composite adornment. Perforated shells of Lithoglyphus sp. are also represented in Early Neolithic (Starčevo-Criş) contexts at Cuina Turcului. Common to all examples are the perforation, which appears to have been made by the same technique and the development of the use-wear. The characteristic elements of the perforations are the sub-circular hole which sometimes has slightly irregular edges with a “chipped” appearance and cracks at the points of impact.

Holes with irregular edges in gastropod shells can result from natural impacts, especially among species that inhabit the gravelly beds of fast-flowing rivers. Lithoglyphus naticoides (the gravel snail) inhabits slow-flowing rivers, but can be found on soft or stony ground. Also, the convexities of gastropod shells such as the whorls are worn by friction on sand (e.g. Langley & O’Connor, 2015; Smith & Nelson, 2003), which can lead to the development of a perforation. Natural erosion results in an overall polishing of the shells (Dupont, Laporte, Courtaud, Duday, & Gruet, 2014). All these factors were taken into account and a systematic evaluation of the items was performed, starting from the taphonomic alterations that might occur. In the case of the archaeological specimens analysed, we could detect a consistent pattern of perforation morphology, which would tend to exclude natural processes. Moreover, among a natural death assemblage of L. naticoides shells, we observed variable patterns of breakage around the margin of the aperture, similar to those noted in some other studies (e.g. Rigaud et al., 2014b).

Experimental studies (Lazăr, Mărgărit, & Radu, 2018; Mărgărit, 2016; Mărgărit, Radu, Boroneanţ, & Bonsall, 2018b) suggest that indirect percussion was used to perforate the shells. We identified two distinct areas of use-wear. The first occurs between the perforation and the aperture edge and was caused by friction against a thread. The perforation became strongly deformed toward the aperture, developing a concavity with smoothed walls (Figure 1c). On the aperture, the wall has various morphologies dictated by the intensity of the use-wear (Figure 1d). The second area with use-wear traces occurs on the body of the shell between the perforation and the apex. The surface became smooth, with a macroscopic polish. In several cases, a small hole was noted below the apex (Figure 1e). The use-wear developed as a consequence of the way the thread was attached, as the bead came into contact with the adjacent beads (hitting and rubbing against one another) at the level of the apex.

The Lithoglyphus apertus shells from Cuina Turcului (Figure 1f) were perforated by another technique, namely abrasion (Figure 1g). Around the orifice was observed a flat surface covered by parallel fine scratches (Figure 1h and i). The wear was not so well-developed – the abrasion area was not associated with deformation of the perforation. In contrast, the two shells of Lithoglyphus apertus from Ostrovul Banului have perforations with irregular sub-rectangular holes. The edges of the perforations had a faceted aspect illustrating the use of indirect percussion. The perforation is deformed toward the aperture, having smoothed walls and macroscopic polish. Also, the morphology of the aperture was distorted by thread pressure, becoming concave.

The technological observations on the L. naticoides shells also apply to the Theodoxus danubialis shells (e.g. irregular holes and points of impact) (Figure 1j). This species is present in both the Mesolithic and Neolithic levels at Cuina Turcului, and it seems that indirect percussion was the perforation technique used in both periods. The perforations and apertures were similarly heavily deformed by thread pressure reflecting long-term use of the shells (Figure 1k–m).

The Tritia neritea shells (only recorded from Mesolithic contexts) were also perforated by indirect percussion at Cuina Turcului (Figure 2a) and Schela Cladovei. The perforations have similar characteristics to those in the L. naticoides shells. Accentuated use-wear, strongly deforming the perforation toward the aperture, was noted (Figure 2b–d). Some T. neritea shells from Schela Cladovei (Figure 2e) and Ostrovul Banului were modified for use as appliqués (cf. Cristiani & Borić, 2012) that were probably sewn onto clothing. Experimental studies have shown that to obtain this special form, pressure was employed to remove the walls of the whorls and the apex (see also Cristiani & Borić, 2012). Microchipping (Figure 2f) is visible on the outer surface of the perforation. The pieces are heavily worn, resulting in thinning of the aperture wall and the development of a concavity (Figure 2g). The surfaces of the apex and spiral walls are flattened and exhibit a macroscopic polish (Figure 2h), possibly resulting from friction against clothing.

Figure 2

(a) Mesolithic beads made of Tritia neritea shell (Cuina Turcului) (scale = 1 cm); (b and c) details of the perforations; (d) perforation deformation; (e) Mesolithic appliqués made of Tritia neritea shell (Schela Cladovei) (scale = 1 cm); (f) breaking by pressure; (g) aperture deformation; (h) macroscopic polish; (i) Mesolithic bead made of Columbella rustica (Ostrovul Corbului) (scale = 1 cm); (j) perforation detail; (k) perforation deformation; (l) flattened surface; (m) Neolithic bead made of Columbella rustica (Cuina Turcului) (scale = 1 cm); (n) perforation detail; (o) perforation deformation; (p) use-wear at the apex; (q) Neolithic tubular bead made of Antalis sp. shell (Cuina Turcului) (scale = 1 cm); (r) extremity detail; (s) flattened facet. (a and m–s – after Mărgărit et al., 2020).

A perforated Columbella rustica shell was found with a Late Mesolithic burial at Ostrovul Corbului (Figure 2i). The only technological intervention is the perforation for suspension. The perforation has a rounded morphology (Figure 2j) without any manufacturing marks, suggesting a long period of use. The deformation of the perforation and the smoothed appearance of the walls prevented identification of the perforation technique. The associated wear (Figure 2k) and an area of the shell body with a flattened morphology (Figure 2l) suggest that the piece was sewn onto clothing. This species was also found in the Early Neolithic of Cuina Turcului (Figure 2m) where the shell is similarly heavily worn. The perforation is sub-circular with “flanged” walls and fine striations, which suggests it was achieved by abrasion (Figure 2n). There is also a deformation of the perforation wall in the form of a small concavity (Figure 2o), resulting from thread pressure. The shell apex is fractured and shows intense use-wear polish (Figure 2p).

The Zebrina detrita (land snail) shell from the Early Mesolithic of Cuina Turcului was perforated through the last whorl, starting from the inside. The main characteristics are the sub-circular morphology and irregular edges of the perforation, suggesting the use of pressure technique. No use-wear is visible on this specimen.

Tusk shells of Antalis spp. have a natural conical shape and a curved profile. The two Mesolithic specimens from Climente II and Cuina Turcului do not have a pronounced conical form, suggesting they were obtained by a segmentation procedure. Two techniques used for the segmentation of tusk shells are known: sawing and bending (Vanhaeren & d’Errico, 2001). The extremities are worn away, obscuring the technique used to produce the beads. The Early Neolithic specimen from Cuina Turcului (Figure 2q) provides more use-wear information. It exhibits extremely advanced use-wear (Figure 2r) on the narrower end, which precludes identification of the segmentation procedure. The tip shows a marked concavity that corresponds with a flattened area (Figure 2s) on the body of the shell exhibiting macroscopic polish, which resulted from use as an adornment and suggests the piece was worn for a long period. The anterior end of the shell appears to have been fractured post-depositionally.

Bivalves (Unio sp.) appear for the first time in the Starčevo-Criș levels of Cuina Turcului (Figure 3a). Two valves of Unio sp. have a perforation located approximately in the same position, below the umbo. The perforation is sub-circular with an irregular outline (Figure 3c). In places, the perforation edge has a faceted aspect indicating impact points; there are also cracks starting from some of the impact points, suggesting the application of indirect percussion. On one of the valves, a second perforation appears to have been initiated but not finished (Figure 3b), which seems to confirm the use of indirect percussion. Experiments on modern specimens (Sztancs, Beldiman, Barbu, & Barbu, 2016) suggest indirect percussion was applied bilaterally and repeatedly to create a perforation with the required dimensions. We could not identify use-wear on the perforation walls of the specimens from Cuina Turcului; hence, their use as pendants is hypothetical.

Figure 3

(a) Neolithic perforated Unio sp. valve (Cuina Turcului) (scale = 1 cm); (b) unfinished perforation; (c) perforation detail; (d) Mesolithic bone pendant (Cuina Turcului) (scale = 1 cm); (e) sawing marks; (f) scraping marks; (g) perforation details; (h) red ochre specks; (i) Mesolithic antler pendant (Cuina Turcului) (scale = 1 cm); (j) scraping marks; (k) perforation detail; (l and m) decoration details; (n) Mesolithic (left) and Neolithic (right) perforated fish vertebrae (Cuina Turcului) (scale = 1 cm); (o) bending technique; (p) perforation detail (a, c, i and j – after Mărgărit et al., 2020).

3.2 Bone and Antler

The bone pendant (Figure 3d) from the Early Mesolithic of Cuina Turcului was made on a flat blank, seemingly obtained by sawing followed by bending, with the segmentation marks (Figure 3e) still visible at both extremities. The external face preserves the original bone morphology. The inner face was regularised by longitudinal scraping (Figure 3f). The perforation was obtained through rotation (Figure 3g), starting from the inner face, resulting in a hole with a conical profile. Specks of red ochre (Figure 3h) are visible on the edges of the perforation and toward the distal extremity.

The debitage procedure used in the case of the flat antler blank from the same (“Epipalaeolithic I”) cultural horizon as the bone pendant (Figure 3i) could not be determined, as all manufacturing traces were erased when the inner surface was scraped (Figure 3j) to make it more regular. At one end, the piece has a perforation made by rotation only (Figure 3k), which started from the outer surface. On this surface is a series of nine parallel oblique lines produced by sawing (Figure 3l). The piece is broken at the opposite end, but four transverse saw marks (Figure 3m) are still visible.

In the case of the catfish (Silurus glanis) vertebrae from the Mesolithic (Figure 3n, left) and Starčevo-Criș (Figure 3n, right) horizons at Cuina Turcului, the vertebral spines were detached (Figure 3o). The central perforation was produced by bifacial rotation. The beginning of use-wear development, in the form of macroscopic polish and fine scratches, is visible on the periphery of the perforation (Figure 3p).

Two bone beads are present in the Early Neolithic assemblage from Cuina Turcului. One is a discoidal bead made on a flat blank (Figure 4a, left), but we were unable to determine the debitage procedures since the rim has been heavily abraded (Figure 4b) to obtain the circular morphology. The perforation was achieved by bifacial rotation. The second piece (Figure 4a, right) was made on a volume blank and has a sub-oval outline reflecting the bone morphology. Segmentation was performed at both ends by sawing (Figure 4c) followed by abrasion of the sawn edges. The medullary channel was used as the perforation.

Figure 4

(a) Neolithic bone beads (Cuina Turcului) (scale = 1 cm); (b) abrasion marks; (c) sawing marks; (d) Mesolithic cyprinid teeth (Schela Cladovei) (scale = 1 cm); (e and h) use-wear traces on the fracture edge; (f and i) use-wear traces on the globular body; (g and j) use-wear traces on the neck.

3.3 Tooth

Pharyngeal teeth of Rutilus sp. (probably R. frisii) were found in burials 38 and 40 at Schela Cladovei (Boroneanț, 1990) (Figure 4d) and in burial 24 at Ostrovul Corbului. For fastening, the natural shape of the pharyngeal tooth was used opportunistically. No technological interventions are evident, other than those related to the extraction of the tooth from the pharyngeal bone. Tooth extraction was quite easy and was performed either through percussion around the tooth or by bending of the jawbone as experimental studies have shown (Mărgărit et al., 2018b). The teeth show intensive use-wear, reflecting their mode of attachment and use before deposition in the grave. The intensity of the use-wear is variable among the teeth, ranging from “fresh” surfaces, to surfaces with a strong macroscopic polish and the development of a groove with use-wear traces. Analysis of the archaeological finds identified three areas where use-wear developed, which are indicative of the fixing of the teeth (sewn onto cloth/leather) individually. Two of the wear areas occur on only one of the tooth facets, related to friction between the tooth and the garment. The first appears as a fractured edge (Figure 4e and h), gradually acquiring a round shape and displaying macroscopic polish and fine striations. The second occurs on the crown of the tooth (Figure 4f and i) adjacent to the neck and shows the same polish, traces of compression, and fine striations. The third use-wear area is on the neck itself (Figure 4g and j), caused by the fastening of the tooth onto cloth or leather. This area displays striations with differing morphologies covering the entire surface, showing that the thread was first wrapped around the entire circumference of the neck, then sewn. The striations range from fine to deep incisions, likely depending on the duration of the tooth’s life as an ornament or on the characteristics of the fixing thread.

The pharyngeal teeth from Ostrovul Banului are in different stages of extraction from the pharyngeal bone (Mărgărit, Boroneanț, & Bonsall, 2017). In most cases, to remove the bone material, a percussion procedure was applied around the tooth, such that the blank still retains fragments of it. In one case, the use of bending is attested, indicated by a V-shaped fracture at the level of the bending plane. The first method is undoubtedly safer, allowing intact blanks to be obtained. Microscopic analysis identified no wear traces, indicative of the use of the teeth as appliqués like those from Vlasac (Cristiani & Borić, 2012) or Schela Cladovei (Mărgărit et al., 2018b).

In the case of the Mesolithic red deer canines from Cuina Turcului and Climente II (Figure 5a), the perforation was made through the root of the tooth. Both unifacial rotation (one example) and bifacial rotation (12 examples) were employed. In eight cases, the perforated area was prepared by longitudinal scraping, the marks of this preparation being visible on the periphery of the perforations, while for a ninth specimen the surface was prepared by removing small splinters. On one item, bifacial longitudinal scraping was used to thin the piece, and that operation was continued until perforation was achieved, the perforation having an elongated form (Figure 5b and c). Two areas with use-wear were observed on the teeth. The first occurs between the side of the tooth and the perforation, which tends to be deformed in this area, while the wall of the perforation was affected by friction with a thread, becoming flattened or even with a slight depression (suggesting more prolonged use) and exhibiting macroscopic polish (Figure 5d–f). The location of this use-wear indicates the canines were suspended in such a way as to produce the most intense wear along the lateral edges, suggesting they were sewn onto clothing. The second area of use-wear occurs on the lobe of the tooth and consists of flattening of the surface (Figure 5g) and the development of a macroscopic polish associated with irregular scratches that are visible under magnification – again, likely the result of friction with clothing.

Figure 5

(a) Mesolithic perforated Cervus elaphus teeth (Cuina Turcului) (scale = 1 cm); (b) scraping marks; (c and d) perforation details; (e and f) use-wear marks at the perforation level; (g) use-wear marks at the crown level; (h) Mesolithic perforated Vulpes vulpes tooth (Climente II) (scale = 1 cm); (i) perforation detail; (j and k) use-wear along the perforation; (l) Neolithic perforated Vulpes vulpes tooth (Cuina Turcului) (scale = 1 cm); (m–o) perforation details (c, l, m and o – after Mărgărit et al., 2020).

In the case of the fox (Vulpes vulpes) canine from Climente II cave (Figure 5h), the perforation was performed from both sides with preparatory thinning of the perforation area. In the first stage, the area was prepared with small cuts. Then, sawing was applied, creating a groove that is visible at the top of the perforation, on both faces. Finally, the perforation was created by bifacial rotation (Figure 5i). Use-wear developed laterally, identical to the red deer teeth (Figure 5j–k). Interestingly, the only example of a perforated fox tooth identified at Cuina Turcului (Figure 5l) is reportedly from a Starčevo-Criș context. It was perforated in the middle by bifacial rotation (Figure 5m and n). The manufacturing marks are still visible (Figure 5o), suggesting the piece received little use.

In the Mesolithic levels of Cuina Turcului, other pendants made of perforated teeth were discovered. A wild boar lower incisor (Figure 6a) was perforated by bifacial rotation (Figure 6b). The piece had been worn for a long time, so the rotation marks were erased and, at the periphery of the perforation, the surface became flat with fine scratches (Figure 6c). For the wolf incisor (Figure 6d), a more complex procedure was applied: first thinning of the surface by slightly oblique scraping, thus creating a depression with a small oval perforation (Figure 6e). No use-wear was identified. The herbivore incisor (Figure 6f) was processed in a unique manner: the root was removed by sawing followed by bending (Figure 6g), and the perforation made by bifacial rotation (Figure 6h). For the beaver incisor (Figure 6i), preparation of the perforation was initiated by longitudinal scraping, followed by perforation by rotation. A transverse break across the perforation indicates the latter operation was not finished. The last (indeterminate) tooth is heavily fractured, preserving only a part of a perforation, accomplished most likely by rotation.

Figure 6

(a) Mesolithic wild boar incisor (Cuina Turcului) (scale = 1 cm); (b and e) perforation detail; (c) wear along the perforation; (d) Mesolithic wolf incisor (Cuina Turcului) (scale = 1 cm); (f) Mesolithic herbivore incisor (Cuina Turcului) (scale = 1 cm); (g) segmentation side; (h) perforation detail; (i) Mesolithic beaver incisor (Cuina Turcului) (scale = 1 cm); (j) Mesolithic flat “beads” made from Sus scrofa canines (Icoana) (scale = 1 cm); (k) abrasion of the sides; (l) scraping marks; (m) perforation detail; (n) use-wear adjacent to the perforation (f, g and i – after Mărgărit et al. 2020; l and m – after Mărgărit et al., 2018b).

The two beads from Icoana (Figure 6j) made from boar canines are unusual for the Mesolithic in that the blanks were obtained by longitudinal splitting of the tooth. The sides were regularised by abrasion (Figure 6k) and thus the precise debitage procedures could not be identified. The beads exhibit a central perforation, performed by bifacial rotation (Figure 6m). The area to be perforated had first been thinned by scraping (Figure 6l). One bead exhibits heavy use-wear, resulting in the obliteration of the rotation striations and a smooth perforation wall (Figure 6n). In the case of the second bead, the striations are still visible, suggesting less intense use.

3.4 Stone

The stone ornaments in our series are probably all from Starčevo-Criș contexts, except for the two beads from Alibeg (Figure 7a) where both Final Mesolithic and Early Neolithic occupations were recognised (Boroneanț, Mărgărit, & Bonsall, 2019). An unpublished AMS ¹⁴C measurement on a herbivore bone from the “pithouse” where the beads were found returned a Final Mesolithic date (Boroneanț, Bălășescu, Sava, & Bonsall, forthcoming). The rim and both faces of the beads were regularised by abrasion (Figure 7b and d), while the perforations were produced by bifacial rotation (Figure 7c and e).

Figure 7

(a) Neolithic stone beads (Alibeg) (scale = 1 cm); (b and d) abrasion marks; (c and e) perforation details; (f) Neolithic stone disc (Cuina Turcului) (scale = 1 cm); (g) edge detail; (h) perforation detail; (i) Neolithic belt element (Cuina Turcului) (scale = 1 cm); (j) abrasion marks; (k) grooving detail; (l) perforation detail; (m) Neolithic malachite preforms (Schela Cladovei) (scale = 1 cm); (n–p) unfinished perforations (f–l – after Mărgărit et al., 2020).

Early Neolithic stone ornaments have been published from Cuina Turcului (Păunescu, 1978). Only two were available for analysis. One qualifies as a disc bead or ring (Figure 7f). Both faces of the piece were intensely abraded (Figure 7g), and the central perforation was made by bifacial rotation (Figure 7h). The second item was possibly a belt element (cf. Bonnardin, 2009) made of greenish stone (Figure 7i). The distinctive shape was created by cutting and abrasion (Figure 7j). At one end, two grooves (Figure 7k) were incised on both faces. The perforation has a biconvex profile and was created by bifacial rotation, then enlarged by scraping (Figure 7l).

At Schela Cladovei, numerous malachite and greenschist cylindrical beads were recovered (Figure 7m) together with preforms and debitage waste, indicating in situ processing of the beads (Figure 7n and o) (Boroneanț et al., 2019).