Introduction

The region between the Danube and Carpathians is a key area for prehistoric archaeological research, particularly during the 5th millennium BC. This period was characterized primarily by the evolution of the Kodjadermen-Gumelnița-Karanovo VI technocomplex (c. 4600/4500-3900 cal BC) with significant territorial expansion. This technocomplex spread from the Carpathian Mountains (in the north) to the Dniester (in the east) and to the Aegean Sea (in the south). Numerous radiometric data have been published (e.g., Bem 2001; Bréhard & Bălășescu 2012; Reingruber 2015; Popescu et al. 2023): at Hârșova, radiocarbon dates indicated the interval of 4336-4057 cal BC and 4369-4235 cal BC (Bréhard & Bălășescu 2012), at Bordușani-Popină between 4490-4263 cal BC (Gillis et al. 2013), at Căscioarele between 4450-4264 cal BC (Lazăr et al. 2018), at Sultana-Malu Roșu between 4539-4365 cal BC and 4174-3961 cal BC (Lazăr et al. 2016), at Gumelnița between 4331-4060 cal BC (Lazăr et al. 2020), at Cunești between 4596-4439 cal BC (Mărgărit et al. 2013) or at Vitănești, between 4354-4231 cal BC (Ludwig et al. 2009; Bréhard & Bălășescu 2012). North of the Danube, the Gumelnița culture is divided into phases A and B, each with two sub-phases, the last one –B2– not being clearly defined. Based on the series of published radiocarbon dates, it is estimated that the Gumelnița A1 phase evolved between 4600/4550-4350 BC, with the A2 phase falling within the interval 4500-3950 BC, and the B1 phase between 4250-3950/3900 BC. Thus, we see that there are overlaps between the phases of the same culture, the transition phenomenon did not unfold uniformly north of the Danube (Bem 2000).

This period was marked by significant changes in human subsistence strategies, social organization, and technological innovations. North of the Danube, stratified tell settlements are characteristic with villages being rebuilt for generations on the same sites. Some tell settlements were occasionally surrounded by defensive structures such as ditches and embankments, while others were enclosed by palisades with dwellings disposed in parallel rows (e.g., Petrescu Dîmbovița 2001; Lazarovici & Lazarovici 2007; Popovici 2010; Ștefan 2010). Larger, multi-room houses were sometimes constructed with a single story and stone foundations that were aligned in symmetrical arrangements (Lazăr et al. 2018). For the Gumelnița culture, houses with simple clay floors are the most common dwelling structure, while those with floors made of logs covered by a layer of plastered clay are rare (Popovici 2010). The social, technological and economic transformations led to changes in the meat component of the diet (e.g., Bréhard & Bălășescu 2012; García Vázquez et al. 2023) with a significant share of game or aquatic resources; funerary inventories that seem to reflect social hierarchies (e.g., Slavchev 2010; Windler et al. 2013; Lazăr et al. 2018; Chapman 2020); the use of copper and gold as raw materials (e.g., Whittle 1996; Thomas 1999; Bailey 2000; Anthony 2010; Lazăr et al. 2018; Chapman 2020; Radivojević et al. 2021), female figurines made of various materials (clay, bone, gold) (Andreescu 2002, 2009), Spondylus ornaments from the Mediterranean world (Mărgărit 2024a) contrasted with numerous beads made of local Unio shells (Mărgărit 2019; Mărgărit & Radu 2021).

For the osseous industry, the intense exploitation of red deer (Cervus ‌elaphus ‌Linnaeus, 1758) antlers is evident, with some sites showing with an almost equal proportion between bone and antler tools (e.g., Vitănești, B1 level). For domestic animals, the raw materials derive mainly from the ox (Bos ‌taurus ‌Linnaeus, 1758) and sheep (Ovis ‌aries ‌Linnaeus, 1758) / goat (Capra ‌hircus ‌Linnaeus, 1758) species, while other species –pig (Sus domesticus Erxleben, 1777) or wild boar (Sus scrofa Linnaeus, 1758), dog (Canis ‌familiaris ‌Linnaeus, 1758), horse (Equus ‌ferus ‌Linnaeus, 1758)– appear sporadically. A special case seems to be that of the Sus species, which have significant bone remains in terms of number, but whose canines, especially the lower ones, are very little exploited. Within the archaeological assemblages, the predominant tools are those used in domestic activities, made from bones, antlers, teeth, or shells, followed at a significant distance by various types of hunting or fishing weapons, mainly crafted from antler, and a limited number of ornaments or feminine figurines, mostly made from bone. Among the types of tools identified, bevelled tools and pointed tools predominate, followed by spatulas, flattened astragali, spindle whorls, to which are added the various massive antlers tools such as bevelled tools, hoes and hammers, along with pressure flakers, arrowheads or harpoons. Although rare, Sus sp. canines and Unio shells were used as scrapers, while tortoise carapaces served as containers. Also, some typological categories, that were present in the previous stages (Early and Middle Neolithic) at the Lower Danube, such as spoons, rings on blank in volume, bone belt elements or Spondylus bi/trilobed beads, disappear, while new typological categories appear, such as flat, prismatic or en violon bone figurines, triangular bone pendants or dog mandibles used as tools (e.g., Mărgărit & Popovici 2012; Averbouh & Zidarov 2014; Mărgărit et al. 2014, 2022; Mihail & Provenzano 2014; Lazăr et al. 2016; Mărgărit & Toderaș 2019; Mărgărit 2020, 2023).

Starting from archaeozoological studies and technological analyses of osseous industries, we aim to investigate the importance of the Cervus ‌elaphus species in the economy of the Gumelnița communities. Our study explores this pheno­menon along two dimensions: 1) Economic value: the analysis of cervid remains provides insights into hunting strategies, seasonality, methods of collecting antlers from the skull or the butchering patterns of animal carcasses. Zooarchaeological data also contribute to understanding how cervids were integrated into the diet and, more generally, into the economy; 2) Technical use: the role of cervids in artifact production. We will investigate which anatomical elements of the skeleton were used, what operational schemes were applied, and what types of tools/weapons were made. We will emphasize less on the share of shed antlers within the osseous industries because it involves a different method of acquisition than hunting, namely expeditions for gathering. Comparisons will be made between phases A2 and B1 of the Gumelnița culture, and case studies will be a series of important tell settlements (Fig. 1) from the Lower Danube.

Cervus ‌elaphus as food source

With the appearance of the first farmers north of the Danube (c. 6200 BC) during the Early and Middle Neolithic, the percentage of domestic taxa (mainly ovicaprids and cattle) in the faunal assemblages surpasses 70 % (Bălășescu et al. 2005a; Bălășescu & Mărgărit 2014). Later on, during the Gumelnița culture (Table 1; Fig. 2), we note the direct and tight connection between the interest of the populations for hunting, fishing, gathering, and the existence of a rich wild faunal spectrum. For example, at the tell settlement of Gumelnița, stable isotope analyses point to the existence of an agrarian community, with half of the diet consisting of cereals and legumes, supplemented by freshwater resources and hunting. Domestic animals were exploited mostly for by-products and less for meat as shown by García Vázquez et al. (2023).

During the development of the Gumelnița culture, the type of economy changes and hunting reaches unexpectedly high frequencies in certain places. The animals hunted were mainly red deer, wild boar, aurochs (Bos ‌primigenius ‌Bojanus, 1827) and to a lesser extent wild horse (Equus ‌ferus ‌Linnaeus, 1758) or carnivores: wolf (Canis ‌lupus ‌Linnaeus, 1758), fox (Vulpes ‌vulpes ‌Linnaeus, 1758), lynx (Lynx ‌lynx ‌Linnaeus, 1758), wild cat (Felis ‌silvestris ‌Schreber, 1777), etc. (Bălășescu et al. 2005a). There are sites where the remains of wild animals reach and even exceed 50 %. An example is the Vitănești tell settlement, where wild mammals accounted for 68.3 % in Gumelniţa A2 level and 48 % in the Gumelniţa B1 level (Bălășescu & Radu 2003; Bălășescu et al. 2005a, b; Bălășescu 2014). At Căscioarele, B1 level, hunting is attested by the high number of wild mammal remains (Bolomey 1964), reaching 84.1 % (Bălășescu et al. 2005a, b) from the faunal assemblage. A. Bolomey (1964, 1968) notes that deer remains make up more than half of the total osteological material, mentioning 1193 fragments from the Gumelnița B1 level. However, the situation seems radically different in the Gumelnița A2 level from this tell settlement. The predominance of domestic animals is attested, the red deer being present with only 84 fragments (Bolomey 1968). There are tell settlements, such as those at Hârşova and Borduşani (Gumelnița A2) where wild mammals account for only a quarter of the total mammal remains. At Tangâru, the percentage of wild mammal remains was only 2.7 % (Bălășescu et al. 2005b). These examples reflect varied adaptations of hunting strategies. We can probably see a correlation between the development of deer hunting in certain sites (e.g., Căscioarele, Vitănești) and the diversity of weapons made from bone and antlers (e.g., Mărgărit et al. 2022, 2023; Mărgărit 2024b), and, conversely, the limited variety of weapons (e.g., Bordușani, Hârșova) in sites with a marginal role of this type of hunting.

The red deer is a species that inhabits large ecosystems, often moving over medium to long distances in search of food and water. It prefers forests, as well as the neighbouring biogeocenoses, especially steppes, incursions into such regions being frequent, particularly in seasons when food is scarce. Between 7000 and 3000 BC, studies illustrate the impact of average annual temperatures and increased humidity, which allowed for the widespread development of red deer, highlighted by the expansion of the ecosystems they were part of (Popescu 2008). Even if the restrictive climatic factors of the analyzed period essentially influenced each community to identify its own subsistence strategies that exploited mainly local resources, the literature has shown that high hunting rates could have resulted from the interaction of several factors: the appearance of new husbandry techniques (Bovinae and Suinae), the development of complex socio-economic relations between the sites, or an increased social value attributed to wild mammals (Bréhard & Bălășescu 2012).

The red deer is well represented as number of remains in most of the Gumelniţa tell settlements. At Borduşani, Gumelniţa and Hârșova, red deer are outnumbered by the wild boar, and by the aurochs at Vlădiceasca and Însurăţei. The situation is largely the same with regard to NMI (minimum numbers of individuals) (Bălășescu et al. 2005b). At the Teleorman Valley (southern Romania), cervids remains account for 20.7 % in the Gumelnița A2 level and 16.7 % in the Gumelnița B1 level, the red deer being the most hunted animal (Bălășescu & Radu 2003).

The analysis of the metapodial and calcaneal remains of red deer shows that more males than females were hunted which could indicate selective hunting due to their larger size (larger amounts of meat) and the antlers being used in the manufacture of tools and weapons (Bălășescu et al. 2005a, b). Hunting was carried out throughout the year, as proven by the discoveries from the Vitănești tell settlement (Teleorman Valley), where the red deer is represented by all anatomical elements, including cranial elements with antlers on the head (at least 10 individuals) (Bălășescu & Mărgărit 2014). Not coincidentally, in this settlement the processing of antlers cut from the skull is attested, even if not in a significant proportion compared to those made from shed antlers (Mărgărit et al. 2022). Skulls with antlers on their heads were also found at the tell settlements from Hârşova, Mǎriuţa or Luncaviţa (Bălășescu et al. 2005a).

Cervus ‌elaphus supplier as raw material for the osseous industry

Discussion

The study was conducted on 89 artifacts made of bone (n = 70) and antler from hunted animals (n = 19). Regarding the two phases compared, in A2 levels, 57 pieces could be identified, and in B1 levels, 32 artifacts. However, it should be noted that the number of Gumelnița B1 assemblages published is smaller and new studies are needed, so we cannot generalize the data.

For A2 levels (n = 47), the analysis of the assemblages showed, in terms of bone processing, a predominance of flat blanks (n = 39), in comparison with those preserving the anatomical shape (n = 8). The principle of segmentation debitage was retained to produce these blanks in volume with two practical options: one implementing a sectioning procedure by direct percussion, the other by sawing. Partioning débitage method was applied with variants that allowed the production of a variable number of blanks: bipartition (two blanks), quadripartition (four blanks), and successive partitions (a variable number of blanks depending on the way the bone was broken). This blank production by partioning was achieved in most cases by double grooving, much less often by percussion or by a combination of percussion and grooving. A transformation scheme by direct shaping can be added, identified in the case of the processed astragalus and phalanx. For surface modification, abrasion was the most used technique. It is characterized by a flattened area on which fine, dense striations develop, parallel to each other, with a longitudinal or circular arrangement, depending on the direction of the friction process. It is seconded by scraping, sometimes both techniques being used. In this case, the specific traces are long, deep striations, parallel to each other, arranged parallel to the longitudinal axis of the artifact.

Another arrangement technique was perforation by rotation or percussion. In the first case, the perforation has a circular morphology, with concentric striations resulting from the rotation procedure; in the second situation, at the periphery of the perforation, small overlapping material removals resulting from the impact with the hammer are evident. In the case of the B1 levels, the information is extremely similar. Of the 23 bone pieces, only two preserve the anatomical volume. There are bones under processing that preserve technical stigmatas of so that these would also have ended up, eventually, as flat blanks. Also, the segmentation transformation scheme is identical with the same predominance of double grooving. Only in the case of transversal debitage, we did not identify the use of sawing. In the case of perforations, the only technical option was rotation. We can affirm that the productivity of the assemblages was high because the transformation schemes based on quadripartition or bipartition allow obtaining two/four similar blanks that can be transformed into finished objects.

In the case of pieces made of antler, there are fewer variables at the technical level. Regardless of the level from which they come, all blanks preserve the anatomical volume. In the context of a transverse exploitation of the bone block relative to its longitudinal axis, the only debitage method used was segmentation, which, moreover, has only one practical variant: sectioning by percussion. For surface modification, percussion was applied, especially to eliminate the pearling and, to a lesser extent, abrasion. Volume modification was achieved only through perforations, recording the following variants: percussion and the combination of percussion and rotation, at Gumelnița A2 contexts and rotation or percussion, at Gumelnița B1 contexts. As in the case of bone, we can say that the differences, at a technological or typological level, between the two phases of the Gumelnița culture are minimal. Antler blanks in volume were preferred because the tools were to be used for percussion actions (hammers and bevelled tools) and there was a need for robust, hard blanks that could withstand shocks. We can also invoke mechanical values more suitable for these types of tools, because antler is more elastic than bone, and according to this it is better adapted for the manufacture of various tools specific to prehistoric periods (e.g., Billamboz 1977; Michels & Zurbruchen 1991; Averbouh 2000; Riedel et al. 2004; Vercoutère et al. 2007). It absorbs shock and impact, due to the important proportion of organic matter in its structure (MacGregor 1985).

We must say that this analysis cannot give us a complete picture of the technological exploitation of red deer bones. This is primarily because in each archaeological assemblage, a large part of the pieces cannot be determined as raw material. For example, it can be seen that the Cervus ‌elaphus ribs are either missing or that they were technologically transformed because they offer a special, wide and flat blank (Table 2). However, in all cases, it was only possible to establish that they come from species of the Bos ‌taurus / Cervus ‌elaphus, without a precise attribution. A second example is that of the metapodials. The radical transformation of the bone through technological interventions, again, makes it impossible to determine the species. We can take as an example level B1 from the Vitănești tell settlement. Here, 307 bone artifacts were inventoried, of which only 16 items could be attributed with certainty to the Cervus ‌elaphus species. To these are added 20 bones belonging either to the species Bos ‌taurus or to Cervus ‌elaphus ‌(Mărgărit et al. 2023). We see that, even if we take into account the bones that we can say come from large mammals, their number remains low within the assemblages. We would have expected a much higher number of pieces made from metapodials because we are talking about bones with a straight morphology, from which a high number of blanks with regular shapes could have been obtained through the quadripartition method present at all of the analyzed sites. Another example would be the site from Căscioarele, where the difference between the two levels (A2 and B1) is not signi­ficant in terms of number of pieces, although, as we have shown above, there are significant differences in the weight of osteological remains belonging to the Cervus ‌elaphus species. We can continue with a comparison between the sites of Hârșova and Bordușani, where the osteological remains belonging to wild animals do not exceed a quarter of the total remains, but where the artifacts that are the subject of this study are in comparable proportion or exceed the number of those from sites where hunting has a greater proportion (Tables 2, 3). For the site of Căscioarele, more interesting are the statistics related to the antlers from hunted red deer. Starting from the archaeozoological studies, we would have expected that such antlers would be much less numerous in the A2 level. Or, out of 14 antlers from hunted animals, eight come from the B1 level and six come from the A2 level. Furthermore, all items from the A2 level are hammers, while these hammers are absent in the B1 level.

The observation regarding the reduced weight within the assemblages remains valid also for the antlers from a hunted animal (bois de massacre) (Table. 2). There is a significant discrepancy between the number of antlers from a hunted animal and those that were shed. For example, at the Vitănești site, B1 level, 258 pieces made of antler were counted, of which only one piece was made of antler from a hunted animal, preserving a fragment of the skull next to the basilar area of the antler (Mărgărit et al. 2022). The shed antler is more suitable for processing because it no longer requires hunting the animal that bore it, thereby involving less effort. An antler from a hunted animal that has completed its structural and morphometric development possesses the same physical properties as a shed antler, both of which are fully developed. This provides valuable insights into the desired properties (sometimes, partially developed antlers are used, and obviously, the same properties are not functionally sought) as well as indications of the harvest season, in addition to the information provided by taphonomic changes observed in archaeological antlers (Averbouh 2000). In this case, we think it is about a good knowledge of the mechanical properties of the raw materials.

Conclusions

We can state that there are no significant differences between the two phases of the Gumelnița culture, neither at the level of selected bone types, nor at the typological or technological level, a valid observation especially when we have the possibility to compare sites that contain remains from both phases (e.g., Vitănești and Căscioarele tell settlements). What is valid for all the analyzed sites, however, is the fact that the number of artifacts made from Cervus ‌elaphus bones remains constantly reduced within the osseous industries, even if it was the main game in some sites. The selection of blocks also depended on other factors besides the economic ones. And here, we also reach the cultural determinant often mentioned in literature. The two determinants (economic versus cultural) seem to complement one another, meaning that the knowledge concerning which parts of the animal skeleton may be used, materialized in the collective memory (Averbouh 2003; Choyke & Daróczi-Szabó 2010). It was observed that the changes in the selection of raw materials suffered insignificant modifications during long periods of time, although modifications of the species availability took place (Choyke et al. 2004; Luik 2009, 2011). To support this observation, we can mention the case of Cervus ‌elaphus canines. At the settlement of Cuneşti, a Bos ‌taurus molar were processed and used as pendant. The surface was, thus, modifying the natural shape, in order to imitate a red deer canine. At the Gumelnița tell settlement, a red deer canine was imitated in limestone. To date, there were no perforated red deer canines from this period, while teeth turned into pendants were quasi-absent (Mărgărit 2019). This was probably a cultural choice, although red deer was abundantly hunted by the Gumelnița communities as already detailed. Instead, these teeth were remembered through imitations so that the symbolism of this tooth became special. This option reflects the ways in which the animal materialized as a symbol in the context of human behavior. We can conclude that although our approach was initiated to identify the economic significance of the red deer within the Gumelnița communities, it was ultimately proven that its exploitation was also subject to cultural determinants. This suggests that the cultural values, symbolic meanings, and social distinctions associated with red deer and other animal remains played a decisive role in shaping their selection and treatment, highlighting the importance of symbolic and ritual considerations over mere economic or practical factors in these communities.

Acknowledgments

Thank you to Francesca Știrbu for helping with a revision of the English of this article. Thanks also to the two anonymous reviewers of Anthropozoologica for their insightful comments.

Fig. 1. — Gumelnița tell-settlements mentioned in this study.

Table 1. — Faunal spectrum of some Gumelnița tell-settlements. Abbreviation: NR, number.Bordușani GA2Carcaliu GA2Hârșova GA2Luncavița GA2Căscioarele GB1Vitănești GA2Vitănești GB1Măriuța GB1SpeciesNR%NR%NR%NR%NR%NR%NR%NR%Bos ‌taurus ‌Linnaeus, 1758199621.4211824.5361811.6421823.592017.10105511.684323.0222642.97Ovis ‌aries ‌Linnaeus, 17582192.3540.832554.8080.87170.60830.9571.56285.32Capra ‌hircus ‌Linnaeus, 1758730.7820.42771.4590.9730.11300.3270.74142.66Ovis/Capra131214.0881.66118222.26586.2870.252682.9772.109618.25Sus domesticus Erxleben, 1777232024.906112.68100919.0012213.20572.01119013.180922.097814.83Canis ‌familiaris ‌Linnaeus, 1758134314.41132.7089616.87212.271665.872522.8902.46132.47Total domestic726377.9520642.83403776.0343647.1945115.94287831.7190351.9745586.50Canis ‌lupus ‌Linnaeus, 1758640.69––260.4960.6530.11720.79180.49––Vulpes ‌vulpes ‌(Linnaeus, 1758)1431.53––140.2640.4310.04610.67190.5230.57Lynx ‌lynx ‌(Linnaeus, 1758)80.09––60.1110.11––50.06––––Felis ‌silvestris ‌Schreber, 1777490.53––200.3830.3210.04250.2820.05––Panthera ‌leo ‌(Linnaeus, 1758)––––––––––30.03––––Meles ‌meles ‌(Linnaeus, 1758)160.17––150.2860.6540.142082.29330.90––Martes ‌martes ‌(Linnaeus, 1758)30.03––80.1510.1110.0410.0120.05––Mustela ‌putorius ‌Linnaeus, 175820.02––––––––10.01––––Mustela ‌nivalis ‌Linnaeus, 1766––––10.02––––––––––Lutra ‌lutra ‌(Linnaeus, 1758)380.41––20.0410.11––110.12––––Ursus ‌arctos ‌Linnaeus, 1758––––––––10.04210.210.03––Equus ‌ferus ‌Linnaeus, 1758350.3820.42200.3840.432458.667738.52206.01101.90Equus sp. (little size)––––––––––––20.05––Sus scrofa Linnaeus, 17587578.127214.9789016.7613114.1855019.44132614.62095.71173.23Cervus ‌elaphus ‌Linnaeus, 17585405.8017436.17781.4729131.49119342.17173219.159516.25305.70Capreolus ‌capreolus ‌(Linnaeus, 1758)1601.7251.04661.24232.491214.281481.6180.4950.95Alces ‌alces ‌(Linnaeus, 1758)––––––––––40.04––––Dama ‌dama ‌(Linnaeus, 1758)10.0110.21––––––––––––Bos ‌primigenius ‌Bojanus, 1827750.80173.53551.04121.302398.45136815.151514.0661.14Castor ‌fiber ‌Linnaeus, 1758971.0440.83320.6020.22160.573634.01153.14––Lepus ‌europaeus ‌Pallas, 1778660.71––400.7530.3230.11891.0100.27––Total wild205422.0527557.17127323.9748852.81237884.06621168.3175948.037113.50Total determinated9317100.00481100.005310100.00924100.0028291009089100.03662100.00526100.00

Fig. 2. — Faunal remains of Cervus ‌elaphus ‌Linnaeus, 1758: A, left neurocranium with antlers (lateral and medial view, Hârșova); B, left humerus distally fused (cranial view, Vitănești); C, right metacarpus distally fused (dorsal view, Bordușani); D, right tibia proximal fused (dorsal view, Luncavița); E, left calcaneus fused (medial view, Vitănești). Scale bars: 5 cm. Photos credits: A. Bălășescu.

Fig. 3. — Bone artifacts found at the Gumelnița A2 levels: A, processed mandible (Cunești); B, E, F, H, O, distal end details; C, I, abrasion marks; D, hammer (Bordușani); G, bevelled tool (Cunești); J, pointed tool (Bordușani); K, grooving marks: L, scraping marks; M, incisions detail; N, proximal end detail. Scale bars: A, D, G, J, 1 cm; B, C, E, F, H, I, K-O: 250 μm. Photos credits: M. Mărgărit.

Fig. 4. — Bone artifacts found at the Gumelnița A2 levels: A, F, pointed tools (A, Hârșova, F, Căscioarele); B, G, L, grooving marks; C, H, M, abrasion marks; D, I, N, distal end details; E, J, usewear at the mesial area; K, needle (Vitănești); O, perforation detail. Scale bars: A, F, K, 1 cm; B-E, G-J, L-O: 250 μm. Photos credits: M. Mărgărit.

Fig. 5. — Bone artifacts found at the Gumelnița A2 levels: A, spatula (Vitănești); B, J, usewear at the distal end; C, K, M, abrasion marks; D, I, bevelled tool (D, Hârșova; I, Vitănești); E, blank (Hârșova); F, grooving marks; G, waste (Hârșova); H, sawing marks; L, handle (Hârșova); N, flattened phalanx (Vitănești); O, perforated phalanx (Vitănești); P, perforation detail: Q, usewear at the level perforation. Scale bars: A, D, E, G, I, L, N, O: 1 cm; B, C, F, H, J, K, M, P, Q: 250 μm. Photos credits: M. Mărgărit.

Fig. 6. — Bone artifacts found at the Gumelnița B1 levels: A, flatenned astragalus (Vitănești); B, C, F, M, R, abrasion marks; D, knife (Vitănești); E, L, P, distal end details; G, blank (Vitănești); H, J, N, S, grooving marks; I, waste (Vitănești); K, pointed tool (Căscioarele); o. bevelled tool (Vitănești). Scale bars: A, D, G, I, K, O: 1 cm; B, C, E, F, H, J, L-N, P-R: 250 μm. Photos credits: M. Mărgărit.

Fig. 7. — Bone artifacts found at the Gumelnița B1 levels: A, needle (Măriuța); B, abrasion marks; C, H, K, M, P, grooving marks; D, G, J, distal end details; E, perforation detail; F, I, pointed tools (F, Vitănești; I, Măriuța); L, N, blanks (Vitănești); O, waste (Vitănești). Scale bars: A, F, I, L, N, O: 1 cm; B-E, G, H, J, K, M, P: 250 μm. Photos credits: M. Mărgărit.

Fig. 8. — Antler artifacts found at the Gumelnița A2 levels: A, E, I, hammers (A, Bordușani; E, I, Căscioarele); B, C, F, G, L, M, details of the active area; D, percussion marks; H, K, perforation details; J, abrasion marks; N, O, debitage waste (N, Bordușani; O, Hârșova). Scale bars: A, E, I, N, O: 1 cm; B-D, F-H, J-M: 250 μm. Photos credits: M. Mărgărit.

Fig. 9. — Antler artifacts found at the Gumelnița B1 levels: A, hammer (Vitănești); B, C, details of the active area; D, percussion marks; E, bevelled tool (Căscioarele); F, preform (Căscioarele); G, undeterminated object (Căscioarele). Scale bars: A, E-G: 1 cm; B-D: 250 μm. Photos credits: M. Mărgărit.

Table 2. — Numerical distribution of artifacts according to typological category.Raw materialTypological categoryBordușaniCarcaliuCuneștiHârșovaLuncavițaCăscioareleVităneștiMăriuțaTOTALArchaeological levelA2A2A2A2A2A2B1A2B1B1BonePointed tool46–323432229Bevelled tool––22–––12–7Needle––––––11–13Handle–––2––––––2Processed phalanx–––––––2––2Flattened astragalus–––1––––1–2Knife––––––––1–1Spatula–––––––1––1Hammer1–––––––––1Undeterminated object2––2–––1––5Blank–––2––––6–8Waste4––2––––3–9AntlerHammer1––––6––1–8Bevelled tool––––––2–––2Preform––––––2–––2Undeterminated object––––––4–––4Waste1––2––––––3Table 3. — Numerical distribution of artifacts according to raw material. *, archaeological assemblages directly studied by the authors.Skeleton elementsBordușani*CarcaliuCunești*Hârșova*LuncavițaCăscioarele*Vitănești*Măriuța*Archaeological levelA2A2A2A2A2A2B1A2B1B1Antler2––2–68–1–Astragalus–––1––––1–Mandible––1–––––––Radius1–––––––––Ulna––11––––1–Metacarpus––––––––5–Metapodialus96–9235641Femur–––1––––––Tibia–––1–––1––Metatarsus1––1––––42Phalanx–––––––2––TOTAL13621629139163