New evidence reveals the earliest use of cinnabar in the western Mediterranean: The Neolithic settlement of La Marmotta (Lazio, Italy)

Quaternary Science Reviews 335 (2024) 108746 Available online 12 June 2024 0277-3791/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/). New evidence reveals the earliest use of cinnabar in the western Mediterranean: The Neolithic settlement of La Marmotta (Lazio, Italy) Cristiana Petrinelli Pannocchia a , * , Alice Vassanelli a , Vincenzo Palleschi b , Stefano Legnaioli c , Mario Mineo d , Gerard Remolins Zamora e , Niccol` o Mazzucco a , Juan F. Gibaja f a Department of Civilizations and Forms of Knowledge - University of Pisa, Via dei Mille 19, 56126, Pisa, Italy b Istituto di Chimica dei Composti Organo-Metallici, CNR, U.O.S. di Pisa, Area della Ricerca del CNR, Via G. Moruzzi, 1, 56124, Pisa, Italy c Istituto di Chimica dei Composti Organo-Metallici, CNR, U.O.S. di Pisa, Via G. Moruzzi, 1, 56124, Pisa, Italy d Museo delle Civilt` a, Piazza Guglielmo Marconi 14, 00144, Rome, Italy e Regirarocs, S.L. Plaza de les Homilies 6, 25794, Organy` a, Lleida, Spain f Instituci´ o Mil` a i Fontanals (IMF-CSIC), C/Egipciaques 15, 08001, Barcelone, Spain ARTICLE INFO Handling Editor: Dr Mira Matthews Keywords: Cinnabar Ore Western Mediterranean Central Italy Personal Ornaments La Marmotta (Lazio) Early Neolithic Archaeometric Analysis ABSTRACT Numerous researchers point out the emergence of human symbolism is related to the evolution of the complexity of human cognition. Red mineral pigments have been used extensively, particularly with anatomically modern humans, for various purposes. However, the management and supply of these pigments during prehistoric periods remains poorly investigated. Still today, the limited application of physico-chemical analyses often leads to a simplistic attribution of these pigments as ochre. The studies of data from recent literature presented in our paper show a progressive introduction and exploitation of cinnabar ore, to achieve a red pigment, from the seventh millennium BC. In this panorama, the new data obtained from the analyses of samples of artefacts from La Marmotta (Italy) show a wide use of cinnabar in central Italy from the early Neolithic and attest to the earliest use of this ore in the western Mediterranean area. 1. Introduction Natural pigments, especially red type, were widely used in prehistory. In recent years, several scenarios have been proposed to assess the use of these colours for cultural, medical, and artistic purposes. The emergence of human symbolism and a sense of art is usually related to the evolution of the complexity of human cognition (Watts, 2002, 2010; McBrearty and Brooks, 2000; d’Errico et al., 2003; d’Errico et al., 2010; d’Errico and Henshilwood, 2011; Henshilwood and Marean, 2003; Henshilwood and Dubreuil, 2009). Recent archaeological data testify to the earliest use of red pigments in Africa and the Near East during the Middle Stone Age (Watts, 1999, 2002; Rigaud et al., 2006; Jacobs et al., 2006; Bar-Yosef Mayer et al., 2009; Zilh˜ ao, 2012). At Blombos Cave (South Africa) in a layer dated circa 100.000 years ago, quartzite grinders used to produce powder of ochre, which was then voluntarily stored in shells (Henshilwood et al., 2009, 2011) were found. In Europe, a series of new dates obtained from carbonate crusts suggest attributing the artistic activities discovered in three Iberian caves (La Pasiega in Cantabria; Maltravieso in Extremadura; Ardales in Andalucía) to the last Neanderthals (Hoffmann et al., 2018). Other accounts bear witness to the interest of Neanderthal groups in the use of red colour. At Maastricht-Belv´ ed` er, a layer referable to the middle Palaeolithic revealed the presence of non-local hematite infiltrating the sediment, following voluntary, not well-defined, human activity. Lumps of red pigment and a Glycymeris sp., having traces of ochre on the concave surfaces, were discovered in a burial dated 92.000 years ago at Qafzeh Cave (Israel, Walter, 2003). During the upper Palaeolithic, the presence of red pigments in burials increased. The earliest evidence of this tendency includes the socalled red lady, a male burial from Paviland (South Wales, 26.000 years agoAldhouse-Green, 2000), the grave of a young prince at Arene Candide (Italy, 24.000 years ago Pettitt et al., 2003) and the double child burial at Sunghir (Russia, 24.000 years ago Formicola and Buzhilova, 2004). In all these cases, the pigment was sprinkled on the bodies. It is * Corresponding author. E-mail addresses: [email protected] (C. Petrinelli Pannocchia), [email protected] (A. Vassanelli), [email protected] (V. Palleschi), [email protected] (S. Legnaioli), [email protected] (M. Mineo), [email protected] (G.R. Zamora), [email protected] (N. Mazzucco), [email protected] (J.F. Gibaja). Contents lists available at ScienceDirect Quaternary Science Reviews journal homepage: www.elsevier.com/locate/quascirev https://doi.org/10.1016/j.quascirev.2024.108746 Received 7 March 2024; Received in revised form 28 May 2024; Accepted 28 May 2024 Quaternary Science Reviews 335 (2024) 108746 2 not completely clear if the Palaeolithic human groups used the mineral powder for a preservative purpose or a ceremonial function. The use of colour seems to assume a significant role in the social life of the Palaeolithic groups, as underlined by the considerable evidence of rock paintings from southwestern Europe. Despite our knowledge being restricted to preserved evidence, ethnographic observation on modern hunter-gatherer societies and experimental activities permits us to suppose a wider range of uses of natural pigments related to daily life. Mineral pigments can be used as tanning and preserving agents for leather and food, as substances added to adhesives to reinforce their adherent capacities, as medicinal substances, antiseptics, and insect repellents (Schmandt-Besserat, 1980; Velo, 1984; Rifkin, 2011). It certainly should not be ruled out the application of pigments to produce marks on objects or bodies, as later demonstrated by tattoos discovered on the mummy of ¨ Otzi (Val Senales, Italy, Hedges et al., 1992), dated to the Copper Age (Table 1). During the Holocene, the red colour maintained its importance, although we cannot attest to the same significance or symbolic value attributed to it in the previous period. Traces of red pigment are found in about 50% of the European Mesolithic burials (Grünberg, 2015). From the Pre-Pottery period in the Near East, we recognise a large use of red pigment in architectural and funerary contexts. One of the first attestations dated to the Natufian site of Ain Mallaha (Israel, Perrot et al., 1988; Bocquentin, 2003; Valladas and Kalteneker, 2007; Valla et al., 2017) where the pigment was in the plaster covering the house walls. From the end of ninth millennium BC (Table 1), the habit of decoring the internal walls of dwellings became much more widespread, as testified by numerous sites, such as Ἁin Ghazal (Jordan, Rollefson, 1983; Banning and Byrd, 1987; Rollefson et al., 1992; Regagnon, 2001 Schmandt-Besserat, 2013; Grissom and Griffin, 2013), As¸ikli H¨ oyük (Turkey, Esin et al., 1991; ¨ Ozbas¸aran, 2011; Quade et al., 2014), Çatalh¨ oyük (Turkey, Mellaart, 1967; Hodder, 2001, 2004, 2012, 2015; Carter, 2009; Orton et al., 2018; Bayliss et al., 2015; Schotsmans et al., 2022), and Arpachiyah (Jezireh irakienne, Stuckenrath and Ralph, 1965; Hijara, 1980; Campbell, 2000). At the latter site, the red colour was also used in depositional contexts and for pottery decoration (Anderson et al., 2014). Although colours played a significant role in prehistoric human societies, there is still a lack of systematic studies on their management and supply modalities in different chronological and regional contexts (Up¯ ıte, 1987; Sulgostowska, 1990; Delibes de Castro, 2000; Salomon et al., 2008; Zilh˜ ao et al., 2010; Çamurcuo˘ glu, 2015; Domingo and Chieli, 2021). Data from Çatalh¨ oyük suggests the inhabitants used a wide range of pigments from the end of the eighth millennium BC. Ochre was the most relevant, while yellow ochre, cinnabar, blue azurite, and green malachite were less common (Schotsmans et al., 2022). It turns out to be cinnabar and not ochre also the red colour present in a Halaf burial at Tel Abu Huraira (Syria, Molleson et al., 1992; Moore et al., 2000; Molleson, 2016). In central Europe, the use of cinnabar was first claimed by M.M. Vasi´ c (Vasi´ c, 1932) asserting to have found evidence of cinnabar in every layer of the site of Vinˇ ca (Serbia), ranging from the middle of the sixth to middle of the fifth millennium BC. Recent analyses (Gaji´ c-Kvaˇ sˇ cev et al., 2012) confirm the presence of this pigment in this site, although with a much lower incidence than previously assumed. The earliest evidence of cinnabar in the Balkan region seems to relate to the settlement of Ploˇ cnik (Serbia), where a vessel containing red powder and red-painted figurines, dating back to the second half of the sixth millennium BC, were discovered. The results obtained from the analyses were not exhaustive to indicate the source area of the ore. The mineral could have been collected in the region of Mount Avala in the Table 1 Radiocarbon dates of the principal Neolithic and Copper Age sites mentioned in the text (calibration with OxCal v4.4.2 using IntCal20, Reimer et al., 2020). Ain Mallaha Israel GifA 100400 10540 90 10736-10527 Perrot et al. (1988) GifA 70013 8740 40 7932–7611 Ἁin Ghazal Jordan AA 1164 9100 140 8530–8231 Rollefson et al. (1992) AA 5196 7670 100 6633–6432 Çatalh¨ oyük Türkiye PL 980525A 8340 90 7524–7199 Bayliss et al. (2015) OxA 11764 6707 38 5640–5569 ¨ Ozbas¸aran (2011) As¸ikli H¨ oyük Türkiye GrN 28617 8980 40 8261-8220 AA 87976 8690 20 7735–7603 Quade et al. (2014) Arpachiyah Iraq P 585 8064 78 7129–6839 Stuckenrath and Ralph (1965) BM 1531 6930 60 5887–5732 Palmisano et al. (2022) Tel Abu Huraira Syria OxA 1227 8320 80 7508–7195 Moore et al. (2000) OxA 1931 7890 90 7031–6642 Vinˇ ca-Belo Brdo Serbia Hd 16661 6353 66 5463–5221 Schier (1996) NOSAMS 67700 5890 45 4827–4714 Bori´ c (2009) OxA 24923 5335 35 4247–4056 Bori´ c (2009) Ploˇ cnik Serbia OxA 14684 6354 36 5368–5231 Bori´ c (2009) OxA 14678 4431 36 3263–3015 Bori´ c (2009) Casa Montero Spain Beta 206512 6410 40 5471-5324 Bustillo et al. (2009) Cova de l’Or Spain OxA 10192 6310 70 5362–5214 García-Borja et al. (2006) Montelirio Spain CNA 589 4400 60 3259–2916 Bayliss et al. (2016) OxA 30385 4151 30 2872–2635 Bayliss et al. (2016) Campo de Hockey Spain CNA 833 5665 50 4544–4449 Vijande Vila et al. (2015) CNA 360 5020 50 3942–3711 Pe˜ nacalera cave Spain Beta 491868 4620 30 3491–3364 Gleba et al. (2021) Perdig˜ oes Portugal Beta 327750 4030 40 2579–2475 Emslie et al. (2016) Beta 308789 3840 30 2343–2207 Carpignano Italy LTL 126A 5666 60 4547–4448 Tiberi and Dell’Anna (2013) Grotta dei Cervi of Porto Badisco Italy LTL GCD2 5452 45 4346–4258 Quarta et al. (2018) Catignano Italy Ly 2186 OxA 6455 60 5477–5367 Colombo (2006) LTL 1155A 5587 60 4488–4353 Casale Somaini Italy LTL 3472A 5240 50 4217–3981 Barich et al. (1968) Lunghezzina Italy OxA 80789 4740 45 3629–3382 Manfredini (2012) Torre della Chiesaccia Italy LTL 3486A 4129 45 2869–2583 Anzidei et al. (2011) La Vela VII Italy Utc 10555 5579 45 4452–4354 Dal Rì et al. (2002) Utc 10556 5555 48 4448–4347 Mummy of Similaun (¨ Otzi) Italy OxA 3376 4450 80 3342–2928 Hedges et al. (1992) OxA 3371 4660 55 3519–3367 C. Petrinelli Pannocchia et al. Quaternary Science Reviews 335 (2024) 108746 3 Suplja Stena (Serbia) mine. The latter is close to the Vinˇ ca site, and about 300 km from Ploˇ cnik. Lithic tools and pottery have been found in the mine, testifying to its use since the late Neolithic. However, the earliest traces of mining may have been removed by later activities (Gaji´ c-Kvaˇ sˇ cev et al., 2012). Vessels decorated with cinnabar have also been found in Croatia, at the late Neolithic site of Grapˇ ceva Cave (Kaiser and Forenbaher, 2016). In the western Mediterranean, evidence of extensive exploitation of Neolithic cinnabar has been found in Almaden (Ciudad Real, Spain, Hunt-Ortiz et al., 2011). Other smaller mines in Spain are in Castellon, Valencia, and Murcia provinces, but the extraction activities in these latest areas seem to have started in medieval times (Domingo et al., 2012). Almaden mine was active from the Iberian early Neolithic, as witnessed by the evidence recovered at the flint mine of Casa Montero (Madrid, Spain). On this site, which was also a blade production centre, distributed all over the region (Casta˜ neda, 2016), a blade was found covered with cinnabar. Radiocarbon dating of two charcoal samples collected in the flint shafts suggests its use from the second half of the sixth millennium BC (Bustillo et al., 2009). Cinnabar was also found inside a Glycymeris sp. shell at Cova de l’Or (Alicante, Spain), an early Neolithic settlement, occasionally used as a burial place (García-Borja et al., 2006). In Spain, the presence of this pigment in funerary contexts increased from the late Neolithic to the early Copper Age. Numerous burials, such as the dolmen de Alberite (C´ adiz) and the dolmen of Casas de Don Pedro in the Guadiato Valley (Bueno Ramírez et al., 2020) dated to the end of the fifth millennium BC, show the widespread use of cinnabar to cover the bodies of the buried and parts of the funeral area. In the Cueva de Los Murci´ elagos de Zuheros (Cordoba, Martínez Fern´ andez et al., 1999), in a similar chronological horizon, the pigment was used as a filler for the grooves of pottery decoration and to cover the surface of some lithic tools. Analyses (Emslie et al., 2022) of several burial contexts located in Portugal and Spain, ranging from the middle Neolithic to the Copper Age (i.e.: Campo de Hockey, Spain, Vijande Vila et al., 2015; Emslie et al., 2015; Cova da Moura Cave, Portugal, Silva, 2002; Perdig˜ oes, Portugal, Emslie et al., 2016; Montelirio and Montelirio Tholos, Spain, Emslie et al., 2016; García Sanju´ an et al., 2018) have revealed high levels of total mercury (THg) in human bones from to the end of the fifth millennium BC. According to the researchers who conducted the study, this high presence of mercury is probably due to the use of cinnabar as body paint or as medicine, during rituals and social practices. By the end of the third millennium BC, THg levels in human bones became low, probably because of changes in funeral customs and human groups’ social dynamics (Emslie et al., 2022). The first evidence for the use of cinnabar for dyeing textiles (Pe˜ nacalera cave, Cordoba) is also attributed to the Copper Age (Gleba et al., 2021). This paper aims to present the results obtained from the analysis of artefact samples from the early Neolithic site of La Marmotta. The XRF analyses have underlined the choice of the inhabitants of the site to exploit two different substances to achieve a red colour: cinnabar and ochre. The presence of cinnabar will be discussed in relation to the different artefact types. It will also present an updated picture of the oldest evidence of the use and exploitation of cinnabar in Western Europe, among which, in the light of the new evidence, La Marmotta provide the earliest one. In Italy, chemical analyses conducted on the Square Mouthed Pottery culture (SMP) burial of a child (Tomba 3, first half of the fifth millennium BC, Dal Rì et al., 2002) at La Vela di Trento in Northern Italy, revealed traces of cinnabar on the necklace’s shell beads, bracelet, skull, and on a hemispherical bowl that the child held in his hands. In the late Neolithic grave of Grotta Patrizi di Sasso Furbara (Lazio, Grifoni Cremonesi and Radmilli, 2000), cinnabar was found on the skull of the buried person, on the bed, and in the grooves of the engraved decoration of a vase, placed near the body. Additionally, cinnabar residues were found on the skull of a man buried at Carpignano (Apulia, Tiberi and Dell’Anna, 2013). In a different context, a Serra d’Alto vessel with cinnabar painted decoration (Quarta et al., 2018) was discovered at Grotta dei Cervi of Porto Badisco (Apulia), a cave used for ritual purposes (Graziosi, 1980; Aprile et al., 2017). Traces of this pigment have also been attested on an animal bone handle in the settlement of Catignano (Abruzzi, Colombo, 2006). During the Copper Age, numerous hypogeum burials, related to the Rinaldone facies (from the fourth to the beginning of the second millennium BC, Anzidei and Carboni, 2020), including Sgurgola, Ponte San Pietro, Bandita San Pantaleo (Barich et al., 1968), Casale Somaini, Lucrezia Romana, Ponte delle Sette Miglia, Lunghezzina (grave 3), Romanina (Anzidei and Carboni, 2020) and some Laterza facies’ sites, as Torre della Chiesaccia 2 (grave 4, Anzidei et al., 2011), attest the use of this pigment to cover parts of the body, mainly the skull, and lithic artefacts, usually arrowheads and dagger. Lumps of the ore were also occasionally deposited near the buried. 2. The site of La Marmotta The early Neolithic site of La Marmotta (Fig. 1) is located under the waters of Lake Bracciano (Anguillara Sabazia, Lazio). It was excavated from 1992 to 2006. In 2009 a small archaeological intervention was carried out under the supervision of the Soprintendenza speciale al Museo nazionale preistorico etnografico Luigi Pigorini (today the Museo delle Civilt` a) (Fugazzola, 2002). The archaeological site lies approximately 300 m away from the modern shoreline, submerged at a depth of 11 m (8 m of water and 3 m of sediment), which has permitted exceptionally good conservation. The numerous and varied objects and implements made of wood, basketry and textiles reflect their importance for the Neolithic communities and the technical skill they reached to manufacture them. This technological know-how can undoubtedly be explained by a socioeconomic organisation based on the specialisation of particular artisans. Only people with specialised knowledge would be capable of building the large dwellings, but also many of the artefacts described below: bows, sickles, spindles, wooden recipients, baskets, fabrics and canoes (Fig. 2). That is why La Marmotta can be considered a Pompeii-like Neolithic site. The radiocarbon dating of the village of La Marmotta was mainly carried out on wooden samples taken from poles deeply embedded in the lake. The site has been dated again more recently using short-lived samples (charred seeds of T. dicoccum) from Layers 1 and 2 (Fugazzola and Tinazzi, 2010; Mazzucco et al., 2022; Mineo et al., 2023). The radiocarbon calibrated dates obtained so far are distributed over a time interval between 5690 BC (Pole 214, Square A2) and 5260 BC (Pole 21, Square B29). The stratigraphic sequence is composed of three main layers (Layer II, Layer I, Layer ‘Chiocciolaio’), without any real interruption or sterile layer between them. Layers have been defined on the basis of the pedological characteristics, the radiocarbon dates and the type and style of the pottery assemblage associated. Layer II corresponds to the beginning of the settlement. The pottery assemblage belongs to the Tyrrhenian facies of the Impressed Ware Culture, and is characterized by the high-quality production of a diversity of vessels, including some unique shapes, like the pirogue-shaped vases (Fugazzola et al., 1993). Layer I is characterized by a greater number of findings of all types and represents the time of the most intense occupation of the village. This pottery is decorated with paintings and incised motifs (Sasso-Fiorano style). The last phase of Layer I, also called Layer ‘Chiocciolaio’, represents the abandonment of the settlement. The settlement was formed by rectangular houses, 8–10 m long by 6 m wide, with internal compartments and a central hearth. Near some houses up to 5 canoes were documented (Mineo et al., 2023). The distribution of the houses seems to indicate that there was an organization of space. La Marmotta was based on a consolidated domestic economy in which several animal and plant species were consumed: goats, sheep, cattle, pigs, wheat, barley, legumes (lentils, broad beans and peas), and C. Petrinelli Pannocchia et al. Quaternary Science Reviews 335 (2024) 108746 4 possibly also grapes and opium poppies (Rottoli, 1993; Tagliacozzo, 2005; Salavert et al., 2020). The analysis of the lithic raw materials shows that the population of La Marmotta had a well-established contact network with communities from several parts of the central Mediterranean. This is demonstrated by the presence of different varieties of flint probably acquired locally or regionally, for example in the Apennines, whereas others came from distant locations, such as the Defensola mines in the Foggia region on the east coast of Italy (Radi and Danese, 2003; Pessina and Tin´ e, 2010; Muntoni et al., 2021), the obsidian from the Lipari and Palmarola islands (De Francesco et al., 2012), and some of the polished green axes and adzes probably originating from the Alpine territories (D’Amico, 2000). 3. Materials and method The study was performed on eighteen artefacts showing red pigment traces (Figs. 3, 7 and 8, Table 2): ten ornaments, two stone tools, one pottery sherd, and five raw material samples. The more extensive analysed set involves adornment items. Residual evidence of a thin, red-coloured substance was discovered on ten beads, mostly on lithic and shell beads and in traces on clay and seed beads. The seed beads are part of a necklace on display at the Museo delle Civilt` a in Rome, which has been assembled for exhibiting reasons with heterogeneous pieces coming from different parts of the excavation. Two other clay beads from the same ensemble (unpainted and not listed in Table 2) were used for comparison with the XRF spectra obtained on red-painted clay bead 14504. Further to these objects, similar traces were found on two stones, probably employed as querns. It was also decided to analyse samples of reddish raw materials recovered during the excavation, and a pottery fragment, where a white and a red substance to fill the incised decoration was clearly visible. Ornaments and macro lithic tools were measured, weighed, and classified, according to their shape and size, based on typologies proposed in the literature (Ornaments: Barge, 1987; Bains, 2012; Micheli, 2009; Macro-lithic tools: Hamon, 2006; Adams et al., 2009; Lunardi and Starnini, 2013). The surfaces of all the items were observed with the naked eye and with an optical binocular microscope (BMS 7458, with 10x to 50× magnification), connected to a digital camera (Canon EOS 550D). Finally, all the data and observations were recorded in a database. In order to define the nature of raw materials of artefacts analyses and residues, archaeometric analyses were conducted. To the ED-XRF analysis (Mantler and Schreiner, 2000) the Elio XRF spectrometer produced by Bruker was used. The experimental conditions were set to 90 s acquisition time, voltage equal to 40 keV and current equal to 80 mA. A confocal micro-Raman system linked to a Leica DLML microscope with a 0.75 numerical aperture 50×NPLAN objective was also used. The measurement was performed using the emission of a CW He–Ne laser at 633 nm. The spectral detection is realized using a single grating monochromator (1200 lines mm −1 ), coupled with a Peltier-cooled CCD detector (578 ×400 pixels of 22 μ m ×22 μ m). The spectral resolution of the spectrometer is 2.0 cm −1 . Before the analysis, the instrument was Fig. 1. Location of the settlement of La Marmotta (Anguillara Sabazia, Lazio, Italy). Fig. 2. Plan of the La Marmotta site with structures and canoes. C. Petrinelli Pannocchia et al. Quaternary Science Reviews 335 (2024) 108746 5 wavelength-calibrated using the 520.0 cm −1 Raman band of a pure silicon crystal. 4. Archaeometric results In Fig. 4A, the spectra of the raw materials 36303 and 47121 are compared. The first sample is dominated by the lines of mercury (L a = 10.0 keV, L b =11.8 keV, L g =13.8 keV and the corresponding escape peaks at these energies minus 1.7 keV, resulting by the interaction with the silicon detector). The fluorescence lines of sulphur are also visible, very close to each other, at low energy (K a =2.3 keV, K b =2.5 keV) (P´ erez-Diez et al., 2023). The presence of mercury and sulphur in a mineral indicates the presence of cinnabar (HgS). The spectrum of the raw material 47121, on the other hand, does not show any fluorescence line at the energies of sulphur and mercury; however, the fluorescence lines of iron are prominent at K a =6.4 keV and K b =7.1 keV), hinting to its identification as red ochre. The raw sample 36303 was also analysed using a laboratory confocal micro-Raman. The resulting Raman spectrum is shown in Fig. 4B. The two main Raman bands of cinnabar around 253 and 343 cm −1 are clearly visible (P´ erez-Diez et al., 2023), thus confirming the ED-XRF results of the previous analysis. The spectra of the raw materials 36169 and 39041 (not shown here) are very similar to 36303, while the one of item 47270 confirms that it is similar to 47121, thus suggesting that the first three are cinnabar (dominant lines sulphur and mercury), while the latter two are red ochre (dominant lines iron). The ED-XRF spectra obtained on the lithic beads 41428a, 41428b, 41428d and 41010 are shown in Fig. 5A. The four beads were found in proximity (see Fig. 8). The differences between the ED-XRF spectrum of the cinnabar layer and the one of the bead core are evident when comparing two zones on the 41010 sample, one covered in red and the other with the core exposed (see Fig. 5B). Mercury and sulphur lines are practically absent in the exposed region, the signal observed in the red region also shows the fluorescence lines of the iron present in the substrate. This evidence seems to exclude the possible mixing of cinnabar and ochre in the red pigment; in fact, the iron signal is mainly coming from the steatite bead core, while the sulphur and mercury signals come from the red cinnabar layer on the bead surface. Being the ED-XRF a volume technique (Angeli et al., 2019), the intensity of the iron lines observed is essentially the same in the two regions (covered and not covered in red). The lithic bead 37482 gives an ED-XRF spectrum (not shown here) similar to the ones in Fig. 4A, although the colour of the cinnabar layer is darker than the others. In this case, the signal of mercury is slightly lower than for the 41010 sample (probably due to the lower thickness of the painted layer) but the iron signal is practically identical in both cases (Fig. 6). This evidence strengthens the hypothesis that cinnabar and ochre are not used together, at least in the samples that we have analysed. The clay bead 14504 also evidences dark surface residues, but its fluorescence spectrum is much richer than the others (see Fig. 5C). Besides sulphur, potassium, calcium, titanium, mercury, and iron, the EDXRF spectrum also shows clear fluorescence lines of rubidium, zirconium, and niobium. The fluorescence lines of these elements are typically found in volcanic sand and rocks (Tykot, 2017a, 2017b). In Fig. 5D we show the comparison of the fluorescence spectra of the red-painted clay bead 14504 with two other clay beads from the exposed necklace (unpainted). Besides the sulphur and mercury signals, which are peculiar to sample 14504, the ED-XRF signals of Rb, Zr and Nb are practically identical in the three spectra. This might indicate the use of the addition of coarse sand of volcanic origin, collected along the shores of the lake, to prepare the clay mixture. The spectrum of the decorated pottery sherd 50628 confirms that the red surface layer is made of cinnabar; the body of the shred shows the expected elements that are typically present in the clay (K, Ca, Ti, Mn, Fig. 3. Map of the distribution of artefacts with cinnabar residues. C. Petrinelli Pannocchia et al. Quaternary Science Reviews 335 (2024) 108746 6 Fe, Cu and Sr). Also in this case, the fluorescence lines of the volcanic sand elements Rb, Zr and Nb are visible, although weaker than the ones observed in the clay beads Fig. 7A). The spectrum of the seed bead of the exposed necklace evidences the presence of lines of calcium, titanium, manganese, iron, copper, and strontium, on top of which we observe the ED-XRF fluorescence lines of sulphur and mercury from the surface residuals (Fig. 7B). The ED-XRF spectra of the three Spondylus beads (36300, 36301, 36302) are dominated by the calcium lines since this element is the main component of the shell in the form of calcium carbonate. The lines of strontium are also very evident because Sr has the same chemistry of Ca and often substitutes it in the form of strontium carbonate. Also in this case, the characteristic lines of Hg, associated with the presence of cinnabar on the surface of the shell, are evident in the XRF spectrum (Fig. 7C). The two querns (15895 and 19661) were used for processing the raw cinnabar, as evidenced by the red cinnabar residuals on their surfaces (Fig. 7D). However, the two spectra have significant differences. Besides the usual elements (K, Ca, Ti, Fe, Sr) and the lines of sulphur and mercury, the spectrum of item 19661 also shows the fluorescence lines of Rb, Zr, and Nb (see Fig. 7D). The reason for this difference is not clear; the two querns (15895 and 19661) seem visually similar, but only in one we observed the presence of Rb, Zr, and Nb, which would hint to a possible volcanic origin of the stone. Further mineralogical analysis will clarify the nature of the two stones. 5. Discussion In most cases, the analyses have established that the brick-red substance detected in three of the five raw materials analysed (36303, 36169 and 39041 in Table 2) is mercury sulphide (HgS), known as cinnabar. The other two raw materials were identified as ochre (47121 and 47270). As previously indicated, traces of red pigment were found on numerous ornamental items (Fig. 7), almost exclusively on lithic (5) and shell beads (3). The substance retains the original red pigment colour, although it has taken on a darker hue in some areas. Cinnabar is an unstable pigment in the long term and can turn black when exposed to light (Çamurcuo˘ glu, 2015; McCormack, 2000) or due to its long stay under water (B´ earat et al., 2013). The latter seems the most likely reason given the particularity of the context of discovery. Analyses have shown that a darker shade of red pigment, observed in some cases at Çatalh¨ oyük (wall paintings and depositions) and in wall paintings from the Classical period, is caused by the use of a mixture of ochre and cinnabar, probably because the latter was in short supply. The preparation of the mixture seems to have been aimed at a desire to enhance the brilliance of the ochre and to accentuate its symbolic value (Çamurcuo˘ glu, 2015; Busacca, 2020). The hypothesis that this was the case at La Marmotta is less probable. The collected data show a deliberate choice to use cinnabar to decorate the ornaments. There is no evidence, at least in the samples analysed, of the possible use of a mixture of cinnabar and ochre since, as the XRF spectra show, the Fe detected in the red samples is compatible with the composition of the substrate material (i.e. Fig. 5B). The five lithic beads analysed were obtained by processing steatite. One of these items (41010, Fig. 8A) can be considered a short cylinder Table 2 List of the artefacts analysed showing red pigment traces. SAMPLE NO. OBJECT MATERIAL LAYER SQUARE RESIDUAL TRACES 41428a Bead Steatite I A416 Cinnabar 41428b Bead Steatite I A416 Cinnabar 41428d Bead Steatite I A416 Cinnabar 41010 Bead Steatite I A414 Cinnabar 37482 Bead Steatite I A420/ 469 Cinnabar 36300 Bead Spondylus Shell I A41 Cinnabar 36301 Bead Spondylus Shell I A41 Cinnabar 36302 Bead Spondylus Shell I A41 Cinnabar 14504 Bead Clay I D191 Cinnabar Exposed necklace Bead Seed Und. Und. Cinnabar 36303 Raw Material Sample Pigment I A41 Cinnabar 36169 Raw Material Sample Pigment I A41 Cinnabar 39041 Raw Material Sample Pigment I A488 Cinnabar 47121 Raw Material Sample Pigment I A509 Ochre 47270 Raw Material Sample Pigment Postholes A464 Ochre 50628 Incised pottery sherd Clay I D300 BIS Cinnabar 15895 Quern stone I D237 Cinnabar 19661 Quern stone I A22 Cinnabar Fig. 4. ED-XRF spectra: A. Comparison between the ED-XRF spectra of two of the raw materials analysed (36303 – cinnabar and 47121 – red ochre); B. Raman spectrum of sample 36303. C. Petrinelli Pannocchia et al. Quaternary Science Reviews 335 (2024) 108746 7 based on the dimensions. The other four items (41428a, b, d, Fig. 9, and 37482) have a discoidal shape. Three of these last were part of a set of seven beads recovered together, probably belonging to the same jewel. One of these discoidal beads (41428d, Fig. 9d), in particular, shows a thick stratum of pigment covering its surface completely. This highlights a specific habit of the inhabitants of the village to cover the darkcoloured raw material of the lithic beads with a reddish pigment. Conspicuous traces of cinnabar were also found on the surfaces of three ovoidal beads (36300, 36301, Figs. 8C, and 36302), with rather large dimensions. Based on the macroscopic inspection of the surface characteristics, these items are made in Spondylus sp. The use of these shells for making ornaments and their associated prestige value is known for the early Neolithic in the Mediterranean basin (Borrello and Micheli, 2004; Micheli, 2014; Windler, 2018). Some scholars claim that the Neolithic communities valued the Spondylus shell for its cultural significance and its white colour (Windler, 2019; Kurzawska and Sobkowiak-Tabaka, 2020). At La Marmotta, on the contrary, the use of the red colour to cover their surfaces may reflect the intention to recall the intense purple colour, characterising the outer layer of certain Spondylus sp. (Dimitrijevic’ and Tripkovic’ 2006). It should be remembered that the red colour was used not only to cover the surface of these white-looking products but also for the dark lithic beads described above. Furthermore, the data highlights that only a small percentage of the total assemblage was dyed. This may indicate the intent to achieve an aesthetic effect by enhancing the appearance of only a few beads within a composition. This habit probably did not concern only lithic and shell beads as evidenced by the coloured clay bead analysed (Fig. 8B). This last small bead with cinnabar traces on the surface is similar in dimension to the cylindrical lithic bead. The seed bead analysed (exposed necklace, Fig. 8D:1; Petrinelli Pannocchia and Vassanelli, 2023) is small in dimension and preserves its Fig. 5. A. Spectra obtained on the lithic beads 41428a, 41428b, 41428d and 41010; B. Comparison of spectra of two areas from the 41010 sample; C. Spectrum obtained from the clay bead 14504; D. Comparison of the fluorescence spectra of two clay beads (14504 and two beads from the exposed necklace). Fig. 6. Iron and mercury signals in samples 37482 (black curve) and 41010 (red curve). C. Petrinelli Pannocchia et al. Quaternary Science Reviews 335 (2024) 108746 8 automorphic ovoidal shape. Compared to the beads described up to now, the limited amount of cinnabar residue on this item makes it difficult to determine if it was originally coated with cinnabar or if it came into contact with pigment-coated materials. In a few beads, traces of red pigment were found even inside the perforations. This was observed in two items, a Spondylus shell bead (36301, Fig. 8C) and a clay bead (14504, Fig. 8B). The presence of the pigment inside the perforations suggests possible dyeing of the thread with the red colour. As mentioned above, the earliest use of cinnabar to treat textiles in the Mediterranean region is testified by the discovery at Pe˜ nacalera cave in the Sierra Morena hills, Southern Spain. The colouring of the thread used to tie the beads at La Marmotta could be considered indirect evidence of dyeing fabrics that could chronologically anticipate the appearance of this practice in the Mediterranean area. Due to its nature, the ore must be ground into a fine red powder before being mixed with other components to make a spreadable paste or applicable liquid. The presence of three cinnabar samples of raw materials (36303, 36169, 39041) and two querns (15895 and 19661) indicate that the cinnabar was prepared at the site and was extensively used by this Neolithic community. Indeed, the analysis of the residues in the incisions of a ceramic sample (56628, Fig. 10) testifies to the use of cinnabar to heighten the chromatic effect of the pottery decoration. As previously mentioned, similar use has been found in the late Neolithic levels at the Grapˇ ceva site in Croatia. This is one of the few sites where mercury sulphide was used as a pigment to paint the ceramics after firing (Kaiser and Forenbaher, 2016). For the Italian peninsula, the use of cinnabar by the first agro-pastoral communities as a filler for impression and engravings has been only supposed until now (Grifoni Cremonesi, 2004). According to the data reached from the XRF analysis, the La Marmotta site represents the oldest analytically proven use of cinnabar for filling pottery decorations in the Italian peninsula. The results of the analyses revealed that in addition to cinnabar, the craftsmen used ochre as a pigment. The use of both pigments was observed in other sites as well, where they were employed for different purposes (Gaji´ c-Kvaˇ sˇ cev et al., 2012). At La Marmotta, among the items analysed red ochre was detected only as raw material, therefore its use remains to be clarified. Several vessels discovered at the site were described as having internal surfaces coated with ochre (Fugazzola, 2002). This could testify to the use of pottery to prepare or store red pigment, as testified by the vessels discovered in the Vinˇ ca sites (Mioˇ c et al., 2004). Further analyses on a wider spectrum of archaeological materials could clarify how the villagers processed and used the different pigments, providing additional information on the possible role of the red colour in the site’s life. However, the absence of ochre on the surfaces of the ornaments here analysed may suggest different uses of the two substances. As previously mentioned, the lack of extensive and systematic archaeometric analysis in the Italian peninsula might have led to a misclassification of red pigments, which are often generically labelled as ochre. Consequently, information regarding the utilisation of cinnabar during the Neolithic is far from being complete. The Neolithic context accounts reveal similarities with the use of cinnabar at La Marmotta. Among the farming communities, this red pigment seems to have had a specific symbolic value tied to the body (La Vela and Grotta Patrizi) and to certain products: pottery (Grotta Patrizi, Grotta dei Cervi and La Fig. 7. A. Comparison between the ED-XRF spectra of the painted layer (red) and ceramic body (black) for sample 50628; B. ED-XRF spectrum of the seed bead in the exposed necklace; C. ED-XRF spectrum of the surface of Spondylus bead 36301. The other two (36300 and 36302) are very similar; D. Comparison of the ED-XRF spectra of the surface of the two querns 15895 (black) and 19661 (red). C. Petrinelli Pannocchia et al. Quaternary Science Reviews 335 (2024) 108746 9 Marmotta) and ornaments (La Vela and La Marmotta). During the Copper Age, the exploitation became organised, as showed by the evidence of southern Tuscany (see next paragraph), and the intensive mining ensured a constant and secure supply of this red pigment. The practice of sprinkling the buried bodies with red pigment seems to have continued, often limited to the skull as if it was believed the representative part of the essence of an individual, as attested in numerous hypogeum burials in Latium. Among the grave goods, the lithic weapons seem to replace the previous connection between personal ornaments and red colour. Conversely, in Tuscany, the number of sites showing the use of cinnabar declines progressively, especially in the northern part, where a preference for the white colour seems privileged, as testified by the numerous marble beads recovered in the Copper Age burials (Vassanelli et al., 2023). 5.1. The presence of cinnabar in the Italian Peninsula: supplying areas In Italy, several deposits (88) reveal the presence of cinnabar ore, mainly found in volcanic environments and hot spring deposits (mindat. org). These are almost exclusively along the central-northern Tyrrhenian coast, roughly from Lazio to Liguria region. Only two localities attest to the presence of cinnabar in the southern part of the peninsula: Monte La Mula, in the province of Cosenza, and Palinuro, in the Aeolian islands (Fig. 11). The south of Tuscany is the closest region where cinnabar prehistoric deposits can be found. This mineral is present on Monte Amiata, in the inner part of the region, and on the Monti dell’Uccellina, along the coast. The distance between La Marmotta and both areas is about 100 km as the crow flies. Fig. 8. A. Steatite short cylinder bead 41010 with cinnabar residue; B. Clay bead 14504; C. Spondylus shell bead (36301); D. Exposed necklace with detail of a seed bead (1) and a clay bead (2). Fig. 9. The seven steatite beads recovered together (on top), and various images of bead number 41428d, where the cinnabar residues that almost entirely cover the surface are clearly visible. C. Petrinelli Pannocchia et al.