The Middle Paleolithic quartz assemblage from Gruta da Figueira Brava (Portugal)

palevol comptes rendus 2025  24  28 Comptes Rendus Palevol est une revue en flux continu publiée par les Publications scientifiques du Muséum, Paris et l’Académie des sciences, Paris Comptes Rendus Palevol is a fast track journal published by the Museum Science Press, Paris and the Académie des sciences, Paris Les Publications scientifiques du Muséum publient aussi / The Museum Science Press also publish: Adansonia, Geodiversitas, Zoosystema, Anthropozoologica, European Journal of Taxonomy, Naturae, Cryptogamie sous-sections Algologie, Bryologie, Mycologie. L’Académie des sciences publie aussi / The Académie des sciences also publishes: Comptes Rendus Mathématique, Comptes Rendus Physique, Comptes Rendus Mécanique, Comptes Rendus Chimie, Comptes Rendus Géoscience, Comptes Rendus Biologies. Diffusion – Publications scientifiques Muséum national d’Histoire naturelle CP 41 – 57 rue Cuvier F-75231 Paris cedex 05 (France) Tél. : 33 (0)1 40 79 48 05 / Fax : 33 (0)1 40 79 38 40 [email protected] / https://sciencepress.mnhn.fr Académie des sciences, Institut de France, 23 quai de Conti, 75006 Paris. © This article is licensed under the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/) ISSN (imprimé / print) : 1631-0683/ ISSN (électronique / electronic) : 1777-571X Directeurs De la publication / Publication directors : Gilles Bloch, Président du Muséum national d’Histoire naturelle Étienne Ghys, Secrétaire perpétuel de l’Académie des sciences réDacteurs en chef / editors-in-chief : Michel Laurin (CNRS), Philippe Taquet (Académie des sciences) assistante De réDaction / assistant editor : Adenise Lopes (Académie des sciences ; [email protected]) Mise en page / Page layout : Audrina Neveu (Muséum national d’Histoire naturelle ; [email protected]) révisions linguistiques Des textes anglais / english language revisions : Kevin Padian (University of California at Berkeley) réDacteurs associés / associate editors (*, took charge of the editorial process of the article/a pris en charge le suivi éditorial de l’article) : Micropaléontologie/Micropalaeontology Lorenzo Consorti (Institute of Marine Sciences, Italian National Research Council, Trieste) Paléobotanique/Palaeobotany Cyrille Prestianni (Royal Belgian Institute of Natural Sciences, Brussels) Anaïs Boura (Sorbonne Université, Paris) Métazoaires/Metazoa Annalisa Ferretti (Università di Modena e Reggio Emilia, Modena) Paléoichthyologie/Palaeoichthyology Philippe Janvier (Muséum national d’Histoire naturelle, Académie des sciences, Paris) Amniotes du Mésozoïque/Mesozoic amniotes Hans-Dieter Sues (Smithsonian National Museum of Natural History, Washington) Tortues/Turtles Walter Joyce (Universität Freiburg, Switzerland) Lépidosauromorphes/Lepidosauromorphs Hussam Zaher (Universidade de São Paulo) Oiseaux/Birds Jingmai O’Connor (Field Museum, Chicago) Paléomammalogie (mammifères de moyenne et grande taille)/Palaeomammalogy (large and mid-sized mammals) Grégoire Métais (CNRS, Muséum national d’Histoire naturelle, Sorbonne Université, Paris) Paléomammalogie (petits mammifères sauf Euarchontoglires)/Palaeomammalogy (small mammals except for Euarchontoglires) Robert Asher (Cambridge University, Cambridge) Paléomammalogie (Euarchontoglires)/Palaeomammalogy (Euarchontoglires) K. Christopher Beard (University of Kansas, Lawrence) Paléoanthropologie/Palaeoanthropology Aurélien Mounier (CNRS/Muséum national d’Histoire naturelle, Paris) Archéologie préhistorique (Paléolithique et Mésolithique)/Prehistoric archaeology (Palaeolithic and Mesolithic) Nicolas Teyssandier* (CNRS/Université de Toulouse, Toulouse) Archéologie préhistorique (Néolithique et âge du bronze)/Prehistoric archaeology (Neolithic and Bronze Age) Marc Vander Linden (Bournemouth University, Bournemouth) référés / reviewers : https://sciencepress.mnhn.fr/fr/periodiques/comptes-rendus-palevol/referes-du-journal couverture / cover : A close-up photograph of a quartz scraper surface. Credits: José Paulo Ruas. Comptes Rendus Palevol est indexé dans / Comptes Rendus Palevol is indexed by: – Cambridge Scientific Abstracts – Current Contents® Physical – Chemical, and Earth Sciences® – ISI Alerting Services® – Geoabstracts, Geobase, Georef, Inspec, Pascal – Science Citation Index®, Science Citation Index Expanded® – Scopus®. Les articles ainsi que les nouveautés nomenclaturales publiés dans Comptes Rendus Palevol sont référencés par / Articles and nomenclatural novelties published in Comptes Rendus Palevol are registered on: – ZooBank® (http://zoobank.org) 563 COMPTES RENDUS PALEVOL • 2025 • 24 (28) © Publications scientifiques du Muséum et/and Académie des sciences, Paris. www.cr-palevol.fr Maria N. R. MELO UNIARQ, Centro de Arqueologia da Universidade de Lisboa, Faculdade de Letras, Alameda da Universidade, 1600-214 Lisboa (Portugal) [email protected] (corresponding author) Marianne DESCHAMPS UNIARQ, Centro de Arqueologia da Universidade de Lisboa, Faculdade de Letras, Alameda da Universidade, 1600-214 Lisboa (Portugal) and Laboratoire Travaux et recherches archéologiques sur les cultures, les espaces et les sociétés (TRACES), Université Toulouse Jean Jaurès, Maison de la Recherche, 5, allée Antonio Machado, F-31058 Toulouse, CEDEX 9 (France) [email protected] João ZILHÃO UNIARQ, Centro de Arqueologia da Universidade de Lisboa, Faculdade de Letras, Alameda da Universidade, 1600-214 Lisboa (Portugal) [email protected] Submitted on 12 May 2025 | Accepted on 20 September 2025 | Published on 10 December 2025 The Middle Paleolithic quartz assemblage from Gruta da Figueira Brava (Portugal) urn:lsid:zoobank.org:pub:FDD95F5D-08BB-4FE0-BB78-EC02BAD44555 Melo M. N. R., Deschamps M. & Zilhão J. 2025. — The Middle Paleolithic quartz assemblage from Gruta da Figueira Brava (Portugal). Comptes Rendus Palevol 24 (28): 563-585. https://doi.org/10.5852/cr-palevol2025v24a28 ABSTRACT For the understanding of Middle Paleolithic subsistence behaviors, Gruta da Figueira Brava (Setúbal, Portugal) is a key site, as it represents the earliest evidence of systematic subsistence exploitation of marine resources in Europe known to date. However, the associated lithic assemblage, more than 80% composed of quartz, has so far remained understudied. Although widely used during the Pleistocene and the Holocene, quartz, the most abundant mineral on Earth, has been traditionally considered as a “second-rate” resource. Due to the raw material’s specific fracture mechanics, quartz lithic assemblages have the tendency to be typologically and morphologically less standardized than flint ones. For these reasons, studies of quartz assemblages have been few and far between, especially when flint and quartzite ones, easier to analyze, are found alongside. Following techno-typological and techno-economic approaches, we carried out a study of a sample of quartz cores and blanks from three successive Figueira Brava human occupation phases spanning the 86-106 ka (thousands of years ago) interval. The results show that several reduction methods were successfully applied, displaying good management of the mechanical properties of the raw material. Alongside the Bipolar-on-Anvil technique, methods such as Levallois and Discoid were used, albeit following strategies somewhat different than with flint and quartzite. At Figueira Brava, the use of quartz – abundant locally and skillfully worked through a diverse range of reduction methods – reflects the good adaptation of the cave dwellers to their environmental context. KEY WORDS Neanderthal, quartz, lithic technology, MIS 5. 564 COMPTES RENDUS PALEVOL • 2025 • 24 (28) Melo M. N. R. et al. RÉSUMÉ L’industrie du Paléolithique moyen en quartz de la grotte de Figueira Brava (Portugal). Pour la compréhension des comportements de subsistance au Paléolithique moyen, la grotte de Figueira Brava (Setúbal, Portugal) est un site-clé, représentant la plus ancienne preuve d’exploitation alimentaire systématique des ressources marines connue à ce jour en Europe. Cependant, l’assemblage lithique, composé à plus de 80 % de quartz, est resté peu étudié jusqu’à récemment. Bien qu’il soit le minéral le plus répandu sur Terre, utilisé par différentes technologies lithiques au cours du Pléistocène et de l’Holocène, le quartz a été traditionnellement classé comme une ressource de second rang. En raison d’une forte tendance à la fracturation, les assemblages lithiques en quartz sont typologiquement et morphologiquement moins standardisés que ceux en silex. Pour ces raisons, les études concernent les industries en quartz restent rares, surtout lorsqu’elles sont associées au silex et au quartzite, qui sont plus faciles à analyser. En suivant les approches techno-typologique et techno-économique, nous avons effectué l’étude d’un échantillon de nucléus et de produits en quartz provenant de trois phases successives d’occupation humaine de Figueira Brava couvrant l’intervalle entre 86 et 106 ka (milliers d’années avant le présent). Les résultats montrent que différentes méthodes de taille ont été appliquées avec succès, ce qui implique une bonne gestion des propriétés mécaniques particulières de la matière première. Les méthodes Levallois, Discoïde, et Bipolaire-sur-Enclume ont été utilisées et différentes stratégies techno-économiques peuvent être mises en évidence par rapport à l’industrie en silex et en quartzite. À Figueira Brava, l’utilisation du quartz, abondant localement et taillé grâce à diverses méthodes de débitage, reflète la bonne adaptation des habitants au contexte environnemental du site. INTRODUCTION ReseaRch backgRound Quartz, silicon dioxide (SiO2), is the most common mineral on Earth and a major component of many igneous, sedimentary, and metamorphic rocks (Driscoll 2010). Because of its abundance in the landscape and durable edge properties (e.g. Abrunhosa etal. 2019) or for symbolic reasons (de Lombera-Hermida & Rodríguez-Rellán 2016), quartz was widely used by prehistoric communities (Mourre 1996; Bracco 1997; Jaubert 1997; Knutsson & Lindgren 1999). Despite its widespread representation in the archeological record, the analysis and classification of quartz artifacts have been particularly challenging for archaeologists, due to what Callahan (1987) defined as the “gravel effect” – the perception of most quartz products as amorphous, not easily recognizable as human-modified. This has led to quartz industries becoming out of favor with the scientific community, which has overwhelming focused on flint ones (Mourre 1996; Knutsson & Lindgren 1999). The prevalence of “flint-thinking” in prehistoric archaeology (Knutsson 2014) has fostered the perception that quartz was primarily employed only when higher-quality raw materials, such as flint, were unavailable. This perception gave rise to the Flint-for-Levallois versus Quartz-for-Discoid dichotomy that has been advocated for the Middle Paleolithic of France, a region where flint is naturally abundant (Boëda 1994). It was only from the 1970s onwards, with the emergency of novel analytical approaches, namely the chaîne opératoire approach, that researchers began to fully consider the range of knapping strategies used by Middle Paleolithic people in the context of the diverse environmental constraints that they faced. The observed variation came to be understood as the result of multiple factors: production requirements and site function; raw material availability; the lithology, size, and crystalline properties of the starting volumes (e.g. presence or absence of fissures and cleavage planes); and the technological knowledge of the knappers (Rolland & Dibble 1990; Delagnes & Meignen 2006). This shift in perspective moved the focus away from direct, flint-based comparisons to uncovering the technological logic behind non-flint lithic raw material usage, especially in areas where quartz is more common than flint, such as Scandinavia (Knutsson etal. 2016) or Northwestern and Central Iberia (Pereira & Benedetti 2013). It was with the development of systematic experimental studies (e.g. Tixier 1978; Texier 1981), the pioneering studies in Scandinavia (Knutsson & Lindgren 1999; Saville & Ballin 2008), and the opening up of new fields of research in Africa, where flint played a secondary role (Mourre 1996), that the study of quartz industries gained the role it deserves. More recently, renewed methodological approaches (e.g. de Lombera-Hermida & Rodríguez-Rellán 2016; Knutsson etal. 2016; Manninen 2016; Pargeter & de la Peña 2017; Daffara 2018; Pargeter etal. 2019; Spry etal. 2021; Yeşilova etal. 2024), along with the analysis of previously unstudied assemblages (e.g. Muñoz del Pozo etal. 2023; Ramos etal. 2024; Daffara etal. 2025), have improved our understanding of quartz lithic technology and led to the development of analytical criteria better suited to its petrographic and mineralogical characteristics. In this context, we present Gruta da Figueira Brava (Setúbal, Portugal) as a case study of quartz use in the Middle Paleolithic of the Iberian Peninsula. The lithic assemblage from an earlier phase of fieldwork has been analyzed (Cardoso & Raposo 1993; Raposo & Cardoso 2000), but publication of the stone tools from the most recent excavations (2010-2013) remains partial: it concerns the flint component plus the use-wear MOTS CLÉS Néandertal, quartz, technologie lithique, MIS 5. 565 Figueira Brava Middle Paleolithic quartz assemblage COMPTES RENDUS PALEVOL • 2025 • 24 (28) analysis of a small sample of 50 quartz items (Zilhão etal. 2020). Here, we summarize the results of a techno-typological study conducted on a sample of the quartz assemblage (Melo 2023). Our findings, alongside those made at other Iberian sites, support the idea that quartz is a versatile raw material amenable to exploitation via complex debitage strategies such as the Levallois and Discoid methods. site descRiption and stRatigRaphic infoRmation The Figueira Brava cave (38°28’14”N, 8°59’10”W, WGS84) is located on the Arrábida coast, south of Lisbon (Fig. 1). It was first described in 1945 by H. Breuil and G. Zbyszewski, who mention the presence of lithic artefacts and Quaternary faunal remains (Breuil & Zbyszewski 1945; Cardoso & Raposo 1993, 1998). In its current configuration, the site has three fig. 1 . — A, Location of the cave in the Serra da Arrábida (Setúbal, Portugal); B, overview of its three extant entrances (after Zilhão et al. 2020). AB Porto PORTUGAL Lisbon Entrance 1 Entrance 2 Entrance 3 fig. 2 . — Site plan with position of the excavated trenches (after Zilhão et al. 2020). Elevations are given in meters above sea level. cave wall edge of breccia erosinal scarp excavation trenches collapsed boulders beachrock SEx trench Entrance 3 Entrance 2 Entrance 1 Area E Area F Area D Area C Area B Area A 2.5 7.5 5.0 10.0 7.5 5.0 5.0 5 m N 566 COMPTES RENDUS PALEVOL • 2025 • 24 (28) Melo M. N. R. et al. entrances (1, 2 and 3) and its interior has been subdivided into six distinct areas (A, B, C, D, E, and F; Fig. 2) (Zilhão etal. 2020). The interior area accessed through Entrance 1 (Areas A and B) has been heavily eroded, but the Pleistocene infilling has been preserved in Entrances 2 and 3 and adjacent interior areas (Areas C and F, respectively). The site was first excavated in the 1980s, when Areas B and C were targeted. Almost four thousand lithic artifacts, including mainly quartz and, secondarily, flint and quartzite artefacts, homogeneously attributed to the Middle Paleolithic, were retrieved alongside a Neanderthal tooth and a wide range of faunal remains, including an important marine resource component (Cardoso & Raposo 1993; Antunes & Cardoso 2000; Raposo & Cardoso 2000). A conventional radiocarbon measurement on a bulk sample of limpet shells collected in layer 2 of Area C (lCEN-387) yielded an age of 30 930 ± 700 BP (c.35 ka) for this context (Cardoso & Raposo 1998). As demonstrated by Zilhão etal. (2020), that radiocarbon result is a vast underestimate of the sample’s age, which is due the age of the shells lying beyond the method’s limit of applicability and the attendant impact of even very small amounts of unremoved contaminants. The correct age of the sequence is based on uranium-thorium (U-Th) and optically stimulated luminescence (OSL) dating of stratigraphically associated speleothems and of the sediments themselves, respectively. This work was carried out in the context of the 2010-2013 excavations of Entrance 3 and Area F, shed new light on the site’s stratigraphic layout, and moved the time of human occupation back to the interval of 86-106 ka, in Marine Isotopic Stages (MIS) 5b and 5c (Fig. 3). This work’s wider behavioral significance, however, rested on its demonstration that Neanderthal coastal populations of the Last Interglacial exploited marine resources as much as coeval (e.g. South African) or later (e.g. Atlantic European Mesolithic) “modern human” ones (Zilhão 2012; Zilhão etal. 2020). Our analysis here uses the four-phases framework of Zilhão etal. (2020). These phases correspond to intervals during which the conditions of formation of the stratigraphic units remained similar at all scales (local, regional, topographic, environmental). Phase FB1 corresponds to the LC complex of Entrance 3; it was not represented in the excavated trenches and is only known through exposures observed along the seaward edge of the erosion-truncated infilling of that area. The sample that we report on is comprised only of finds belonging to the FB2, FB3 and FB4 phases. The FB2 and FB3 sediments were brecciated; their excavation required the use of power tools, namely demolition hammers, and the (inevitable) breakage and loss of smaller items that occur in these circumstances must be kept in mind when these phases’ content is compared with FB4’s. fig. 3. — Stratigraphic correlation scheme: equivalence between the stratigraphic units of the different areas, occupation phases, chronology, and positioning in the Marine Isotope Stage (MIS) sequence. After Zilhão et al. (2020), modified. Area C (1989-1989) Entrance 3 (2010-2013) Area F (2010-2013) Human occupation Chrono-stratigraphy 2a 0 1 2 3 4 5 –– – – –– UC1 UC2-UC6 MC0 MC1-MC2 MC3-MC5 LC1-LC3 CO IT0 IT2 IT1 IH1 IH2 IH3 IH4 IH5 IH6 IH7 IH8 IL1 IL2 IL3 IB1 IB2 Phase FB4 Phase FB3 Phase FB2 Phase FB1 Reworked Holocene MIS 2-MIS 5a MIS 5b (86.0 to 90.0 ka) MIS 5c (92.0-94.0 to 104.0-106.0 ka) MIS 5c-MIS 6 MIS 6 or older 567 Figueira Brava Middle Paleolithic quartz assemblage COMPTES RENDUS PALEVOL • 2025 • 24 (28) Coastal zones are ecotones where marine, estuarine, and terrestrial habitats converge to produce some of the ecologically richest places on the planet (Haws etal. 2010, 2011; Brown etal. 2011). They offer plenty of easily exploitable resources, including mollusks, crustaceans, echinoderms, marine mammals, fish, birds, and plant foods (e.g. pine nuts), all of which are abundantly represented in the Figueira Brava deposit. Given its privileged location, it is not surprising that human usage of the cave extended over such a long period of time. The lithic artefacts retrieved from the site are predominantly made of quartz. Flint, quartzite, and limestone are found less frequently (Table 1), while other raw materials, such as jasper, rhyolite, and lydite, are present in very small amounts. RESEARCH CONTEXT AND KEY QUESTIONS Due to its hardness and its mechanical and physical properties, vein quartz was commonly used to produce stone tools, via a range of technological systems, all through the Pleistocene and the Holocene (Mourre 1996; Bracco 1997; Jaubert 1997; Knutsson & Lindgren 1999). However, in the absence of clear definitions of which features specific to quartz are indicative of how reduction proceeded from initial volume to end product, criteria based on the study of flint collections are routinely applied to quartz, generating classificatory confusion and further complicating technological analysis (de Lombera-Hermida 2009; Driscoll 2011; Knutsson etal. 2016; Rodríguez-Rellán 2016). The traditional underrating of quartz as a secondary resource, only used in cases of local unavailability of flint, has mostly been based on the following arguments: – typologically and morphologically, quartz assemblages are less standardized than flint ones due to the raw material’s proclivity to fracture and lower resistance to compression and traction, leading to less well-defined flaking planes (Mourre 1996; Driscoll 2011; de Lombera-Hermida & Rodríguez-Rellán 2016); – quartz often features internal flaws and microfractures due to post-formation processes (de Lombera-Hermida & RodríguezRellán 2016); – being less prone to conchoidal fracturing, quartz tends to break unpredictably along internal fissures, radially and transversely (Spry etal. 2021). Here, based on the analysis of Figueira Brava’s lithic assemblage, we show that these features of the raw material do not stand in the way of its productive and efficient exploitation for stone tool making. In addition, we also address the following questions of more general interest: – does a particular lithological setting condition the organization of the technology of past humans, i.e., is the geological environment the only determinant of the recourse to quartz, or are there other circumstances that also require consideration? – how much and in which way do the mechanical and petrographic properties of quartz affect the organization and management of reduction sequences? – what technological strategies were used to overcome the limitations associated with the knapping of vein quartz, and did such strategies change over time? – was quartz, at least in some cases, preferred over flint and, if so, why? – did raw material selection change over time (e.g. were quartz, quartzite, and flint always used in similar proportions and for similar purposes)? – what kinds of blanks and formal tools were produced, and is there change over time or relative to raw material selection? MATERIAL AND METHODS We used a techno-economic approach (Geneste 1985): every stage in the production of lithic artefacts is situated within a specific temporal and spatial context, from the acquisition of raw materials to their ultimate discard (Leroi-Gourhan 1964; Geneste 1991). The first step is the identification of potential supply areas, specific sources, and modes of introduction. The second step is the technological analysis of the cores and flakes to reconstruct the chaînes opératoires (Leroi-Gourhan 1964; Tixier 1978; Geneste 1985; Inizan etal. 1999). The third step is the typological classification (Inizan etal. 1999; Bordes 2000) of the retouched products, accompanied by an analysis of blank selection. Our study was carried out on a sample comprised of different raw materials (flint, quartz, quartzite, limestone) totaling 1094 artifacts (Table 1) and corresponding to almost 30% of the total (excluding debris and unmodified blocks and unmodified cobbles, N = 3846). The analyzed sample comes from the SEx trench, which illustrates the FB2 phase, and from two squares (T8 and U8) of the Area F trench representative of phases FB3 and FB4. For the technological analysis of the assemblage, an Excel database was created. The different technological features considered (see Raw material: types and alterations and Technological description below) were meant to ensure that comparison between the different phases of Figueira Brava’s occupation was based on the same descriptive and interpretative criteria. Raw mateRial: types and alteRations For each artefact, we determined the raw material properties potentially underpinning the choice of reduction technology or the morphology of extracted blanks: type, volume morphology, and presence/absence of thermoalteration. Traditionally, quartz has been divided into two main subcategories: vein quartz, commonly referred to as milky quartz, and hyaline quartz, widely known as rock crystal (Mourre 1996). In certain instances, especially when dealing with high-quality and diverse samples, further distinctions can be made based on lithological characteristics. In archaeological assemblages, the described variants have included greasy quartz, coarsegrained quartz, and wind-blown quartz ( Ballin 2008; Saville & Ballin 2008). However, this classification does not consider the mineralogical properties and formation processes, which have significant implications for chemical composition and 568 COMPTES RENDUS PALEVOL • 2025 • 24 (28) Melo M. N. R. et al. fracture mechanisms (de Lombera-Hermida & Rodríguez-Rellán 2016). In this regard, the subdivision in morphostructural types is a useful way to understand the technological choices made by prehistoric communities, as it shows how internal fissures and texture play an important role in raw material selection (de Lombera-Hermida 2009; de Lombera-Hermida etal. 2011). table 1. — Composition of the studied stone tool assemblage. “Others” includes lydite (N = 1), rhyolite (N = 1), and jasper (N = 1). Cores Flakes Tools Debris Hammerstones/Anvils Manuports Total FB2 Quartz N 26 54 7 – – – 87 %24.1% 50% 6.5% – – – 80.6% Flint N – 10 – 5 – – 15 %– 9.3% – 4.6% – – 13.9% Quartzite N 2 2 1 – – – 5 %1.9% 1.9% 0.9% – – – 4.6% Limestone N––––– –– %––––– –– Others N – – 1 – – – 1 %– – 0.9% – – – 0.9% Total (N) 28 66 9 5 – – 108 Total (%) 25.9% 61.1% 8.3% 4.6% – – 100% FB3 Quartz N 53 58 12 13 – – 136 %32.5% 35.6% 7.4% 8.0% – – 83.4% Flint N 1 6 4 – – 11 %0.6% 3.7% 2.5% – – 6.7% Quartzite N 4 2 1 – – 1 8 %2.5% 1.2% 0.6% – – 0.6% 4.9% Limestone N – 1 – – – 2 3 %– 0.6% – – – 1.2% 1.8% Others N 1 3 – – – 1 5 %0.6% 1.8% – – – 0.6% 3.1% Total (N) 59 70 13 17 – 4 163 Total (%) 36.2% 42.9% 8.0% 10.4% – 2.5% 1005 FB4 Quartz N 193 354 96 29 9 5 686 %23.6% 43.3% 11.7% 3.5% 1.1% 0.6% 83.9% Flint N 8 51 3 17 – – 79 %1.0% 6.2% 0.4% 2.1% – – 9.7% Quartzite N 1 15 2 2 – – 20 %0.1% 1.8% 0.2% 0.2% – – 2.4% Limestone N 5 9 – – – 4 18 %0.6% 1.1% – – – 0.5% 2.2% Others N – 10 2 3 – – 15 %– 1.2% 0.2% 0.4% – – 1.8% Total (N) 207 439 103 51 9 9 818 Total (%) 25.3% 53.7% 12.6% 6.2% 1.1% 1.1% 100% Uncertain Quartz 2 2 1 – – – 5 All Total (N) 296 577 126 73 9 13 1094 Total (%) 27.1% 52.7% 11.5% 6.7% 0.8% 1.2% 100% 569 Figueira Brava Middle Paleolithic quartz assemblage COMPTES RENDUS PALEVOL • 2025 • 24 (28) Nevertheless, considering that vein quartz presents a high degree of internal variability, its subdivision in only four morphostuctural groups, cannot, alone, fully describe the different subtypes of quartz observed in Figueira Brava’s lithic assemblage. Hence, for a more comprehensive characterization of the quartz varieties found in the studied assemblage we have adapted the classification proposed for the Cardina-Salto do Boi (Côa Valley, Portugal) case-study (Aubry etal. 2016). This classification emphasizes formation processes and composition, allowing a more complete description of quartz’s appearance. The categories that we retained, illustrated in Figure 4 and described in Table 2, are based on texture, presence/absence of internal flaws, brightness, degree of transparency, and color of the observed surface. Criteria for the recognition of burning in quartz artefacts have seldom been reported, described or discussed, but recent experimental studies (Ballin 2008; Driscoll & Menuge 2011) have identified the following: pitting and “peeled-off” surfaces; glossy patina giving the piece an almost polished appearance; alteration of the original color, usually turning it redder or pink, due to the presence of iron oxides; darker, somewhat purplish hue apparent in iron oxide-rich fissures even when the color change is not homogeneous (Fig. 5). technological descRiption For the general description of the attributes of flakes, cores, and tools we followed Inizan etal. (1999) and Bordes (1950), with some adaptations due to the raw material’s specific constraints (de Lombera-Hermida 2009). For a better classification of debitage methods we followed: Boëda (1993, 1994) and González-Molina etal. (2020) for the Levallois and the Discoid methods; de la Peña (2015) and de Lombera-Hermida etal. (2016) for Bipolar-on-Anvil schemes; Tixier (1978) and Tixier & Turq (1999) for the Kombewa method; and Forestier (1993) for the système à surface de débitage alterne (SSDA) method. In tool classification, we followed a very conservative approach, particularly with regard to notches, as they may be an artifact of post-depositional processes, such as trampling (Thiébaut etal. 2010). RESULTS Raw mateRial In the sample studied, quartz corresponds to 83.5% of the total. There is not much variation across the sequence, even though quartz is somewhat less frequent, and flint correspondingly fig. 4 . — Lithological subtypes identified in the Figueira Brava quartz assemblages (representative flakes, and macro views of their surfaces): A, J9.1; B, J9.2; C, J9.3; D, J9.4a; E, J9.4b; F, J9.4c; G, J9.5; H, J9.6a; I, J9.6b; J, J9.6c; K, J9.7a; L, J9.7b; M, J9.7c; N, J9.7d. Scale bar: 5 cm. AB CD EF GH IJ KL MN table 2 . — Figueira Brava: lithological categories of the quartz assemblage (three technologically undetermined specimens were not considered). Subtypes of anhedric milky white quartz (type J9 of Aubry et al. 2016) N % J9.1 Homogeneous 376 41.3% J9.2 More crystallized, cream to pink quartz 42 4.6% J9.3 More crystallized, orange to grayish brown quartz 6 0.7% J9.4a Milky to opaque gray, with some orange veins quartz 26 2.9% J9.4b Milky to opaque, spotted gray quartz 11 1.2% J9.4c Milky, grainy color (gray or pink) spotted beige quartz 29 3.2% J9.5 Milky to translucent white quartz 85 9.3% J9.6a Milky, grainy color (gray or orange), spotted beige quartz 17 1.9% J9.6b Milky to smoky beige or gray quartz 25 2.7% J9.6c Milky to smoky spotted gray quartz 68 7.5% J9.7a Milky “waxy” light gray or beige quartz 28 3.1% J9.7b Milky grainy zoned darker gray quartz 18 2.0% J9.7c Milky grainy white, gray or beige quartz 83 9.1% J9.7d Milky grainy to vitreous with gradient color change beige to white or gray to white quartz 97 10.6% Total 911 100% 576 COMPTES RENDUS PALEVOL • 2025 • 24 (28) Melo M. N. R. et al. platform angles almost always exceeding 100°. In FB2, the two backed flakes identified are linked to the reconfiguration of the debitage surface in a preferential scheme. In FB3, the Discoid method is possibly represented by a single flake with bipolar removals. As with cores, Levallois blanks are more frequent in FB3, while the Discoid method, poorly represented overall, is most frequently found in FB4 (Table 9). The Bipolar-on-Anvil technique resulted in a more varied range of products: cortical flakes, semi-cortical flakes, bipolar flakes, naturally backed, and short or long cortically backed flakes (Fig. 9A). They tend to be longer (Fig. 10) and to present cortical remnants, usually occupying between 5 and 50% of the dorsal surface. Retouched tools The sample studied contains a total of 126 retouched tools, 116 of which are quartz (Table 1). They are small, with average lengths of c.30 mm. For quartz, the predominant tool types are denticulates, followed by notches and sidescrapers (Table 11; Fig. 11A, C). FB4 is typologically more diverse, which, as with the other techno-economic categories, is most likely a function of the larger sample size. Most retouched tool blanks are flakes bearing little or no cortex (which, when present, is usually distal), and a significant percentage are made on perimeter-edged, non-cortical fig. 10. — Quartz flakes: length (A) and width (B) according to the reduction scheme applied (mm). flakes. Cortex is slightly more common in the blanks from FB4, which is likely related to this phase’s somewhat more frequent usage of the Bipolar-on-Anvil technique (Table 9). FB4 also yielded three tools on other types of blanks: a scraper on a small, flat pebble (Fig. 11F); a denticulate obtained by alternating retouch applied to a patinated/rounded flake (Fig. 11D); and another denticulate retouched on the flat surface of a cleaved fragment. There are also two pieces on chunks: a denticulate and a piece with irregular retouch. Almost half of the retouched tools are broken. When the diagnosis is possible, the retouch usually modifies the dorsal faces on their right or distal edges. Among denticulates, micro-notching is the predominant mode of retouch. Among scrapers, retouch is mostly scaled (Fig. 11B, D), featuring wide, short scars closely resembling fish scales (Inizan etal. 1999). In the case of partially retouched flakes, the modification of the edge is irregular and marginal (Fig. 11E). hammeRstones Cobbles used as hammerstones account for only 1.1% of the lithic assemblage and are attested exclusively in FB4 (Table 1). Although frequently broken, their original size can be reconstructed as >5 cm. Regarding raw material subtypes, J9.4a is the most common (N = 3), followed by J9.2 (N = 2) and J9.1 (N = 2). The low number of hammerstones can be explained by the fact that some were later reused as cores, among which a small percentage bear diagnostic stigmata. This reuse did not result from raw material scarcity, but rather from the fact that, once broken, a hammerstone could provide a suitable platform for further flake extraction. This recycling behavior is most often observed in centripetal cores and one-or-twoextraction cores, which also happen to be the most common core types. When the breakage was particularly clean, the resulting platform would also have been especially well suited for centripetal removals. Reduction Most of the time, quartz cores were exploited in an expedient and “opportunistic” manner to produce regular flakes with <50% cortex and backed flakes (Binford 1979; Arzarello 2003; table 11. — Quartz retouched tool types (per occupation phase and excluding one tool of uncertain stratigraphic provenience). Retouched tool types FB2 FB3 FB4 Total % Notch 1 3 24 28 24.3% Denticulate 2 2 39 43 37.4% Double denticulate – – 1 1 0.9% Bec – – 1 1 0.9% Tayac point – – 2 2 1.7% Lateral scraper 1 4 13 18 15.7% Transversal scraper 1 – 3 4 3.5% Oblique transversal scraper – 1 1 2 1.7% Convergent scraper – – 2 2 1.7% Endscraper – – 1 1 0.9% Partially retouched flake 1 2 6 9 7.8% Retouched piece fragment 1 – 3 4 3.5% Total 7 12 96 115 100% A B 70 60 50 40 30 20 10 60 50 40 30 20 10 0 Bipolar-on-Anvil (N = 23) Bipolar-on-Anvil (N = 30) Levallois (N = 20) Levallois (N = 26) Discoid (N = 15) Discoid (N = 15) SSDA (N = 17) SSDA (N = 20) 577 Figueira Brava Middle Paleolithic quartz assemblage COMPTES RENDUS PALEVOL • 2025 • 24 (28) fig. 11. — Quartz tools: A, denticulate from FB4 (with magnesium treatment on the right); B, transversal oblique scraper from FB3 with magnesium treatment; C, transversal scraper from FB4; D, denticulate on patinated/rounded flake from FB4; E, denticulate from FB4; F, scraper on flat pebble from FB4 phase. Scale bar: 5 cm. Credits: A-C, photos by José Paulo Ruas. The modified edge is indicated. A B C D E F 578 COMPTES RENDUS PALEVOL • 2025 • 24 (28) Melo M. N. R. et al. Vaquero & Romagnoli 2017; Carpentieri & Arzarello 2022). Bringing together the results of the analysis of cores, blanks, and tools, the following can be concluded: – despite FB3’s somewhat higher representation of the Levallois concept, there are no substantial differences between the three phases of human occupation; – throughout, the Bipolar-on-Anvil method was used, application of the Discoid method was rare, and raw material subtypes J9.1, J9.7c, and J9.7d were preferentially selected for the application of the more complex Levallois and Discoid methods (Table 12), of which the Kombewa cores may represent but an initial stage; – at discard, Bipolar-on-Anvil cores are larger (Fig. 7B), so it is possible that, among other applications, this technique was used to open quartz pebbles and cobbles to produce blanks that could be used in more controlled, unipolar or centripetal reduction sequences; – there is no correlation between the weight or length of the core and the size of the last removal (Fig. 12), which is consistent with targeting the production of small blanks, no more than c. 3 cm-long on average (perhaps as a deliberate strategy to compensate for the raw material’s structural weaknesses). It is often noted that the Bipolar-on-Anvil technique is particularly effective for working hard, flawed rocks that are otherwise difficult to reduce to small sizes – such as certain varieties of vein quartz (Hiscock 2015). Future studies focusing on small-flake production may offer valuable insights into the full extent that this technique was put to use at Figueira Brava. economy Through the Figueira Brava sequence, quartz is the most used raw material (Table 1). Figure 13 schematizes this raw material’s exploitation economy. All steps of the chaîne opératoire are represented. On-site, reduction of raw pebbles is demonstrated by the number of cortical flakes. Anvil splitting was frequently employed to initiate volume reduction. The ventral surfaces of the resulting flakes were then exploited either with the same technique or through a variety of other methods. A significant proportion of the selected blanks underwent ramification (Bourguignon etal. 2004; Mathias etal. 2021), with Kombewa cores potentially representing the initial stage of Levallois reduction sequences. The primary objective appears to have been the production of small flakes, <3 cm, whose shape is closely linked to the debitage method applied (Fig. 10). When retouched, blanks were transformed into notches, denticulates, and scrapers. Flint was exploited differently (Zilhão etal. 2020). FB2 yielded no cores and only a few <50%-cortical or non-cortical flakes. FB3 yielded a single core, and a few cortical flakes. FB4 accounts for most of the flint artefacts and yielded eight cores (including fragments), mostly reduced by Kombewa or Levallois methods and largely exhausted. The flake assemblage includes large, preferential Levallois blanks that could not have been extracted from the much smaller cores discarded at the site and must represent imported items. This conclusion is further supported by the fact that, among the totality of the flint assemblage, 64% of the retouched tools are made on cortical or partly cortical blanks. This indicates that they were imported as finished implements or as blanks that, prior to discard, were used or re-sharpened at the site (as shown also by a couple of retouch flakes). In the sample studied here, the few tools are denticulates and sidescrapers. Quartzite cores are present throughout but, unlike flint ones, they are far from exhausted. The blanks are mostly cortical or partly cortical (61%). Tools are rare and represented by denticulates and notches made on partly cortical blanks. As far as tool manufacture is concerned, a similar approach was followed, regardless of raw material. The same types of blanks – flakes with <50% of cortex or non-cortical ones – were selected, and the same kinds of edge modifications were applied. DISCUSSION AND CONCLUSIONS Quartz, a raw material with more internal flaws and more prone to fracture unpredictably, has been widely understood as unsuitable for the application of elaborate knapping techniques. The idea that the technical constraints of the Levallois concept necessitate quality raw materials, in contrast to Discoid or other techniques, derives from a wide range of Middle Paleolithic sites of Southern France (Mathias etal. 2020). However, recent work has called into question this proposition, as Levallois productions using quartz and quartzite have been documented at several Middle Paleolithic sites in the Pyrenees and the Iberian Peninsula (e.g. Eixea etal. 2016; Deschamps 2019; Ramos etal. 2024). the Role of quaRtz in the middle paleolithic of ibeRia Multiple sites of the Middle Paleolithic of Iberia, where nonflint raw-materials, such quartz and quartzite, are more common than flint (Pereira & Benedetti 2013), have challenged the above dichotomy (although a few exceptions exist, like the Abric Romaní, where flint from far away was consistently table 12 . — Quartz subtypes used by different debitage methods and techniques. Bipolaron-Anvil Discoid Levallois SSDA Kombewa Total J9.1 1 6 10 4 39 60 J9.2 – 1 – 1 5 7 J9.3 – – – 1 1 2 J9.4a – – – – 2 2 J9.4b – – – – 1 1 J9.4c – – – – 1 1 J9.5 – 3 1 2 15 21 J9.6a – – – 2 1 3 J9.6b 1 – – – 2 3 J9.6c 3 – 1 2 8 14 J9.7a – 1 – – 4 5 J9.7b – – – – 4 4 J9.7c 2 – 5 3 10 20 J9.7d – 4 3 – 11 18 Total 7 15 20 15 104 161 579 Figueira Brava Middle Paleolithic quartz assemblage COMPTES RENDUS PALEVOL • 2025 • 24 (28) preferred over locally availably quartz and quartzite; Vaquero 1999) (Table 13). The long-standing belief that the use of harder raw materials inevitably requires technological approaches distinct from those used for flint has been challenged by Marks etal. (2001), Mourre & Thiébaut (2008), Eixea etal. (2016), Deschamps (2019), and Deschamps & Zilhão (2018). In the case of the Gruta da Oliveira site in Torres Novas, Portugal, Deschamps & Zilhão (2018) demonstrated that, in the stratigraphic units analyzed, the reduction methods employed were the same for quartzite and flint. A similar pattern is observed at Abrigo de la Quebrada (Eixea etal. 2016), where flint is locally abundant and Levallois reduction sequences are represented among quartzite and limestone too. Likewise, at Gruta Nova da Columbeira in Bombarral, Portugal, where quartz is the predominant raw material, followed by quartzite and flint in similar proportions, the reduction methods applied show no substantial differences between raw material types (Cardoso etal. 2002). fig. 12 . — Last flake removals as a function of core weight (A; excluding five outliers from phases FB3 and FB4 weighing >100 g) and core length (B). A B Length of last removal (mm) R² = 0.032 R² = 0.0715 R² = 0.255 5 10 15 20 25 30 35 40 01020304050607 08 0 Core weight (g) FB2FB3 FB4 Linear (FB2)Linear (FB3)Linear (FB4) R² = 0.2148 R² = 0.086 R² = 0.3102 5 10 15 20 25 30 35 40 20 25 30 35 40 45 50 55 60 65 70 FB2FB3 FB4 Linear (FB2)Linear (FB3)Linear (FB4) Length of last removal (mm) Core length (mm) 580 COMPTES RENDUS PALEVOL • 2025 • 24 (28) Melo M. N. R. et al. Many other sites illustrate the use of the Levallois concept on quartzite, but little is known regarding Iberian Middle Paleolithic sites where quartz is the most used raw material (Table 13). Bracco (1997) notes that, at the site of l’Arbreda, in Catalonia, the debitage methods employed for quartz were distinct, even though the blanks selected for tool manufacture were no different from flint ones. Likewise, at Santa Cita in Tomar, Portugal (Cura etal. 2017), no evidence was found that predetermined knapping methods were applied. At Azinhal, located in Vila Velha do Ródão, Portugal, the Discoid method predominates, with Levallois represented by only a few flakes (Almeida 2012). At Cardina-Salto do Boi, in Vila Nova de Foz Côa, Portugal, both Levallois and Discoid methods were applied, especially so in the assemblage from the GFU6-7 stratigraphic unit (Ramos etal. 2024). At Cova Eirós, Lugo, Spain, the Levallois method is employed more frequently than the Discoid method, the latter having been used for quartzite only (Lazuén etal. 2011; Rodriguez-Alvarez etal. 2011). At the Navalmaíllo rockshelter, in Pinilla del Valle, Spain, where the incidence of the Levallois method is low, application was not observed on chert-like rocks and would seem to have been restricted to quartz and quartzite (Márquez etal. 2013). This body of evidence confirms that there is no necessary relationship between the Levallois concept and any specific raw material. While local availability is known to have played an important role in the Middle Paleolithic (Geneste 1991), in sites where multiple raw materials are present the use of quartz is not necessarily a simple byproduct of immediate abundance. The use of non-flint raw materials in Levallois productions can be related to length of occupation and the activities carried out at the site (Eixea etal. 2016). Further studies are needed to determine whether regional patterns exist in quartz exploitation in Iberia and whether its management varied according to site function. figueiRa bRava’s contRibution Figueira Brava adds to and considerably expands current knowledge about the use of quartz in the Middle Paleolithic of Iberia. Based on our analyses, the following evidencebased statements can be made regarding quartz exploitation at Figueira Brava: – quartz was abundantly and consistently used throughout the site’s human occupation, despite the nearby availability of other raw materials. This observation suggests that the use of quartz might have been dictated by choice, not opportunity. Experimental (e.g. Mourre etal. 2011; de la Peña 2015; Pargeter & de la Peña 2017; Spry etal. 2021) and techno-economic studies (Tallavaara etal. 2010; de Lombera-Hermida & RodríguezRellán 2016; Knutsson etal. 2016; Manninen 2016; Márquez etal. 2016) have shown that quartz can play diverse roles in prehistoric technologies depending on its status and abundance, challenging the view of quartz as a mere substitute. As Eixea etal. (2016: 48) pointed out, particular raw materials with appropriate characteristics may have been preferred for the manufacturing of certain kinds of implements. Further work – including targeted surveys and use-wear analysis – is needed to better characterize Figueira Brava’s quartz lithic assemblages and the economic and functional objectives associated with the consumption of this raw material; – our study confirms that the petrographic and mechanical properties of quartz influenced reduction strategies. While most volumes were exploited expediently, more complex debitage methods, such as Levallois and Discoid, were occasionally applied to selected quartz types with finer, more granulated textures and fewer internal flaws; – specific strategies appear to have been employed to mitigate quartz’s tendency to fracture. Previous work (e.g. Tallavaara etal. 2010; Manninen 2016) has identified technological choices that can help in managing this limitation, such as: 1) producing artefacts with thicker butts to yield thicker products less prone to breakage; 2) producing smaller blanks, especially if intent on subsequent modification into retouched tools (Hiscock 2015; Knutsson etal. 2016); 3) preferentially using cortical or neocortical surfaces as striking platforms (de Lombera-Hermida & Rodríguez-Rellán 2016); 4) organizing reduction in longitudinal sequences (de Lombera-Hermida table 13. — Frequency of quartz in the Iberian Middle Paleolithic sites mentioned in the text. *, Counts are for levels 8-9 only; **, quartz is the majority. Site Location Site Age % Levallois Discoid References Oliveira Portugal Cave MIS 5a 10%* ? ? Marks et al. 2001; Zilhão et al. 2020 Columbeira Portugal Cave MIS 5-2 37%** yes yes Cardoso et al. 2002 Azinhal Portugal Open-air MIS 4-3 c. 50% residual yes Almeida 2012 Santa Cita Portugal Open-air MIS 3 58% no no Cura et al. 2017 l’Arbreda Spain Cave MIS 3 73% no no Bracco 1997 Avellaners Spain Open-air ?77% yes no Mora 1984 Navalmaíllo Spain Shelter MIS 5-4 78% residual yes Márquez et al. 2013 Buena Pinta Spain Cave MIS 4-3 c. 80% no no Mielgo et al. 2024 Cueva del Camino Spain Cave MIS 5-4 c. 80% ? ? Arsuaga et al. 2012 Figueira Brava Portugal Cave MIS 5 83% yes yes Melo 2023; this study Cova Eirós Spain Cave MIS 5 88% yes no Lazuén et al. 2011 Cardina/ Salto-Boi Portugal Open-air MIS 6-3 90% yes yes Ramos et al. 2024 Diable Coix Spain Open-air ?91% yes no Mora 1984 Escoural Portugal Cave MIS3-2 99% yes ? Cardoso 2006 Lagoa Funda 2 Portugal Open-air ?100% yes yes Bicho 2004 Vale da Fonte Portugal Open-air ?100% ? ? Bicho 2004 581 Figueira Brava Middle Paleolithic quartz assemblage COMPTES RENDUS PALEVOL • 2025 • 24 (28) etal. 2011, 2016); and 5) employing the Bipolar-on-Anvil technique. The assemblage from Figueira Brava is consistent with application of these strategies; – the methods used to knap quartz blocks remained largely consistent across the three human occupation phases. Given the small size of the study area relative to the original area of occupation, minor variations, such as the somewhat higher frequency of Levallois in FB3, are parsimoniously explained by sampling bias rather than change in technological preference. Figueira Brava’s unique setting – coastal location with abundant sources of food and raw material – enabled Neanderthal groups to maintain long-term usage of the site. Seasonality fig. 13 . — Figueira Brava quartz economics as illustrated by the phase FB4 assemblage. The black-shaded areas indicate cortex cover. Initial volumes HammerstonesCores Recycling Hammerstone to core Core to hammerstone Hammerstone fragment to retouched tool Blank types Core types and debitage concepts Retouched tool types Cores on a flake’s ventral surface Bipolar-on-Anvil Levallois Discoidal Centripetal Kombew aL evallois SSDA UnipolarPreferential Bipolar CentripetalBipolar CBipolar D first – and second – generation flakes 582 COMPTES RENDUS PALEVOL • 2025 • 24 (28) Melo M. N. R. et al. data indicate human presence in every season: aquatic and marine birds, along with mature pinecones, were exploited/ collected during autumn and winter, while brown and spider crabs were gathered in the summer (Zilhão etal. 2020). Quartz, abundantly available in the vicinity, was the raw material of choice for the making of the stone tools involved in the procurement, processing, and consumption of these resources. Although often considered a lower-quality raw material compared to flint in terms of knapping properties, quartz’s local abundance and adequate performance characteristics (e.g. hardness and durable edges) must have made it entirely suitable for the activities carried out at the site. A careful selection of more granular-textured varieties allowed for the application of more complex debitage methods. The reduction sequences employed combined different technological systems (Bourguignon etal. 2004), taking advantage of the natural convexities of collected blocks, and aimed at the production of small flakes. These strategies fully harnessed the potential of quartz, demonstrating technological flexibility and adaptation to available resources. As H. Knutsson (2014) aptly stated, “simple needs not mean archaic.” Acknowledgements We would like to thank the editor-in-chief, Michel Laurin, the associate editor, Nicolas Teyssandier, and the reviewers for their work on the manuscript. REFERENCES abRunhosa a., peReiRa t., máRquez b., baquedano e., aRsuaga J. l. & péRez-gonzález a. 2019. — Understanding Neanderthal technological adaptation at Navalmaíllo rock shelter (Spain) by measuring lithic raw materials performance variability. Archaeological and Anthropological Sciences 11 (11): 5949-5962. https://doi.org/10.1007/s12520-019-00826-3 almeida n. 2012. — O Paleolítico Médio do Complexo Préhistórico do Arneiro – Santana, Nisa. Dez anos de investigação. Açafa On-Line 5: 273-292. https://www.altotejo.org/acafa/docsn5/ paleolitico_medio_arneiro.pdf antunes m. t. & caRdoso J. l. 2000. — Gruta Nova da Columbeira, Gruta das Salemas and Gruta da Figueira Brava: stratigraphy, and chronology of the Pleistocene deposits. Memórias da Academia das Ciências de Lisboa. Classe de Ciências 38: 23-67. aRsuaga J. l., baquedano e., péRez-gonzález a., sala n., q uam R. m., R odRíguez l., g aRcía R., g aRcía n., á lvaRez - l ao d. J., l aplana c. & h uguet R. 2012. — Understanding the ancient habitats of the last-interglacial (late MIS 5) Neanderthals of central Iberia: Paleoenvironmental and taphonomic evidence from the Cueva del Camino (Spain) site. Quaternary International 275: 55-75. https://doi.org/10.1016/j. quaint.2012.04.019 aRzaRello m. 2003. — Contributo allo studio del comportamento tecno-economico dell’uomo di Neandertal: l’industria litica della serie musteriana del riparo Tagliente (Stallavena di Grezzena, VR, Italia). PhD Thesis, Università degli Studi di Ferrara, Italia. a ubRy t., l uís l., m angado l lach J. & m atias h. 2012. — We will be known by the tracks we leave behind: exotic lithic raw materials, mobility and social networking among the Côa Valley foragers (Portugal). Journal of Anthropological Archaeology 31 (4): 528-550. https://doi.org/10.1016/j.jaa.2012.05.003 aubRy t., baRbosa a. f., luís l., santos a. t. & silvestRe m. 2016. — Quartz use in the absence of flint: Middle and Upper Paleolithic raw material economy in the Côa Valley (Northeastern Portugal). Quaternary International 424: 113-129. https://doi. org/10.1016/j.quaint.2015.11.067 ballin t. 2008. — Quartz as a mineral - Its properties, formation and provenance. Scottish Archaeological Internet Reports 26: 43-52. https://journals.socantscot.org/index.php/sair/ article/view/1028 bicho n. 2004. — The middle Paleolithic occupation of southern Portugal. Settlement Dynamics of the Middle Paleolithic and Middle Stone Age 2: 513-531. binfoRd l. R. 1979. — Organization and formation processes: Looking at curated technologies. Journal of Anthropological Research 35 (3): 255-273. https://www.jstor.org/stable/3629902 boëda e. 1993. — Le débitage discoïde et le débitage Levallois récurrent centripète. Bulletin De La Société Préhistorique Française 90: 392-404. https://doi.org/10.3406/Bspf.1993.9669 b oëda e. 1994. — Le Concept Levallois: variabilité des méthodes. Éditions Du CNRS (coll. Monographies du CRA; 9), Paris, 280 p. boRdes f. 1950. — Principes dune methode d’etude des techniques et de la typologie du Paleolithique ancien et moyen. L’anthropologie 54: 19-34. b oRdes f. 2000. — Typologie du Paléolithique ancien et moyen. CNRS Éditions, Paris, 228 p. bouRguignon l., faivRe J.-p. & tuRq a. 2004. — Ramification des chaînes opératoires : une spécificité du Moustérien ? Paleo 16: 37-48. bRacco J.-p. 1997. — Gestion et exploitation du quartz dans les gisements de l’Arbreda et Reclau Viver (Catalogne, Espagne) : aspects techniques et données sur la transition Paléolithique moyen / Paléolithique supérieur. Préhistoire Anthropologie Méditerranéennes 6: 279-184. bReuil h. & zbyszewski g. 1945. — Contribution à l’étude des industries paléolithiques du Portugal et leurs rapports avec la géologie du Quaternaire. Les principaux gisements des plages quaternaires du littoral d’Estremadura et des terrasses fluviales de la basse vallée du Tage. Comunicações dos Serviços Geológicos de Portugal XXVII: 1-678. bRown k., fa d., finlayson g. & finlayson c. 2011. — Small game and marine resource exploitation by Neanderthals: the evidence from Gibraltar, in bicho n., haws J. & davis l. (eds), Trekking the Shore. Interdisciplinary Contributions to Archaeology. Springer, New York, NY: 247-272. https://doi.org/10.1007/9781-4419-8219-3_10 callahan e. 1987. — An Evaluation of the Lithic Technology in Middle Sweden during the Mesolithic and Neolithic. Vol. 8. Societas archaeologica Upsaliensis, Uppsala, 72 p. caRdoso J. l. 2006. — The Mousterian Complex in Portugal. Zephyrus 59: 21-50. https://revistas.usal.es/uno/index.php/05147336/article/view/5638 caRdoso J. & Raposo l. 1993. — As Indústrias Paleolíticas Da Gruta Da Figueira Brava (Setúbal). 3ª Reunião do Quaternário Ibérico, Coimbra: 451-456. caRdoso J. & Raposo l. 1998. — A gruta da Figueira Brava (Setúbal) no contexto do Paleolítico Médio Final do Sul e Ocidente ibéricos. Encontro sobre Arqueologia da Arrábida, Trabalhos de Arqueologia 14: 7-19. caRdoso J. l., Raposo l. & feRReiRa o. da v. 2002. — A Gruta Nova da Columbeira (Bombarral). Câmara Municipal do Bombarral, Bombarral, 142 p. caRpentieRi m. & aRzaRello m. 2022. — For our world without sound: the opportunistic debitage in the Italian context—a methodological evaluation of the lithic assemblages of pirro nord, Cà Belvedere di Montepoggiolo, Ciota Ciara Cave and Riparo Tagliente. Journal of Paleolithic Archaeology 5: 12. https://doi. org/10.1007/s41982-022-00117-9 583 Figueira Brava Middle Paleolithic quartz assemblage COMPTES RENDUS PALEVOL • 2025 • 24 (28) c uRa s., c uRa p., b eRRuti g. l. f., g Raziano R. & p eça p. 2017. — Estudo tecnológico de três sítios do Paleolítico Médio do centro de Portugal: Ribeira da Ponte da Pedra, Santa Cita e Lagoa do Bando, in aRnaud J. & maRtins a. (eds), Arqueologia em Portugal: 2017 – Estado da questão. Associaçao dos Arqueólogos Portugueses, Lisboa: 319-330. daffaRa s. 2018. — Non-flint raw materials in the European Middle Paleolithic: variability of Levallois and discoid knapping methods and study of the supply areas. Universitat Rovira i Virgili, Tarragona. http://hdl.handle.net/10803/669893 daffaRa s., aRnaud J., beRRuti g., caRacausi s. & aRzaRello m. 2025. — Ugly stones sharpen the wits: the Middle Paleolithic of Piedmont in the context of Western Europe. Archaeological and Anthropological Sciences 17: 172. https://doi.org/10.1007/ s12520-025-02284-6 d elagnes a. & m eignen l. 2006. — Diversity of lithic production systems during the Middle Paleolithic in France, in hoveRs e. & kuhn S. L. (eds), Transitions Before the Transition. Interdisciplinary Contributions To Archaeology. Springer, Boston, MA: 85-107. https://doi.org/10.1007/0-387-24661-4_5 de la peña p. 2015. — A qualitative guide to recognize bipolar knapping for flint and quartz. Lithic Technology 40 (4): 316-331. https://doi.org/10.1080/01977261.2015.1123947 de lombeRa-heRmida a. 2009. — The scar identification of quartz lithic industries. Non-flint raw material use in prehistory. Old prejudices and new direction British. Archaeological Reports: 5-11. de lombeRa-heRmida a. & RodRíguez-Rellán c. 2016. — Quartzes matter. Understanding the technological and behavioural complexity in quartz lithic assemblages. Quaternary International 424 (1): 2-11. https://doi.org/10.1016/j.quaint.2016.11.039 de lombeRa-heRmida a., RodRíguez X.-p., fábRegas R. & moncel m.-h. 2011. — La gestion du quartz au Pléistocène moyen et supérieur. Trois exemples d’Europe Méridionale. L’anthropologie 115 (2): 294-331. https://doi.org/10.1016/j. anthro.2011.02.003 de lombeRa-heRmida a., RodRíguez-álvaRez X. p., peña l., sala-Ramos R., despRiée J., moncel m.-h., gouRcimault g., voinchet p. & falguèRes c. 2016. — The lithic assemblage from Pont-de-Lavaud (Indre, France) and the role of the bipolaron-anvil technique in the Lower and Early Middle Pleistocene technology. Journal of Anthropological Archaeology 41: 159-184. https://doi.org/10.1016/j.jaa.2015.12.002 deschamps m. 2019. — Identification of Quina and Vasconian technocomplexes in Gatzarria Cave (north-western Pyrenees), based on the stratigraphic, taphonomic and technological revision of the Georges Laplace collections. Comptes Rendus Palevol 18 (5): 569-586. https://doi.org/10.1016/J.crpv.2019.04.001 d eschamps m. & z ilhão J. 2018. — Assessing site formation and assemblage integrity through stone tool refitting at Gruta da Oliveira (Almonda karst system, Torres Novas, Portugal): a Middle Paleolithic case study. PLOS One 13 (2): E0192423. https://doi.org/10.1371/journal.pone.0192423 dRiscoll k. 2010. — Understanding Quartz Technology in Early Prehistoric Ireland. PhD dissertation, University College Dublin. dRiscoll k. 2011. — Identifying and classifying vein Quartz artefacts: an experiment conducted at the World Archaeological Congress, 2008. Archaeometry 53 (6): 1280-1296. https://doi. org/10.1111/J.1475-4754.2011.00600.X d Riscoll k. & m enuge J. 2011. — Recognising burnt vein quartz artefacts in archaeological assemblages. Journal of Archaeological Science 38 (9): 2251-2260. https://doi.org/10.1016/j. jas.2011.03.028 eiXea a., villaveRde v. & zilhão J. 2016. — Not only flint: Levallois on quartzite and limestone at Abrigo de la Quebrada (Valencia, Spain): implications for Neandertal behavior. Journal of Anthropological Research 72 (1): 24-57. https://doi. org/10.1086/685265 foRestieR h. 1993. — Le Clactonien: mise en application d’une nouvelle méthode de débitage s’inscrivant dans la variabilité des systèmes de production lithique du paléolithique ancien. Paleo, Revue D’archéologie Préhistorique 5: 53-82. https://doi. org/10.3406/PAL.1993.1104 g eneste J.-m. 1985. — Analyse lithique d’industries moustériennes du Périgord: une approche technologique du comportement des groupes humains au Paléolithique Moyen. PhD disseration, University of Bordeaux I, Bordeaux. geneste J.-m. 1991. — Systèmes de production lithique, gestion des outillages et territoires au Paléolithique moyen où se trouve la complexité ? Techniques &Culture. Revue Semestrielle d’Anthropologie des Techniques 54-55: 419-449. https://doi.org/10.4000/Tc.5013 gonzález-molina i., Jiménez-gaRcia b., maillo-feRnandez J.-m., baquedano e. & dominguez-RodRigo m. 2020. — Distinguishing Discoid and Centripetal Levallois methods through machine learning. PLOS One 15 (12): e0244288. https://doi. org/10.1371/journal.pone.0244288 haws J. a., benedetti m. m., funk c. l., bicho n. f., daniels J. m., hesp p. a., minckley t. a., foRman s. l., JeRaJ m. & gibaJa J. f. 2010. — Coastal Wetlands and the Neanderthal Settlement of Portuguese Estremadura. Geoarchaeology 25 (6): 709-744. https://doi.org/10.1002/gea.20330 haws J. a., funk c. l., benedetti m. m., bicho n. f., daniels J. m., minckley t. a., denniston R. f., JeRaJ m., gibaJa J. f. & hockett b. s. 2011. — Paleolithic landscapes and seascapes of the West Coast of Portugal, in bicho n., haws J. & davis l. (eds), Trekking the Shore. Interdisciplinary Contributions to Archaeology. Springer, New York: 203-246. https://:doi: 10.1007/978-1-4419-8219-3_9 hiscock p. 2015. — Making it small in the Paleolithic: bipolar stone-working, miniature artefacts and models of core recycling. World Archaeology 47: 158-169. https://doi.org/10.1080/0043 8243.2014.991808 inizan m., ReduRon-ballingeR m., Roche h. & tiXieR J. 1999. — Technology and Terminology of Knapped Stone. Cercle de recherches et d’études préhistoriques, Nanterre, 189 p. J aubeRt J. 1997. — L’Utilisation du Quartz au Paléolithique Inférieur et Moyen. Préhistoire Anthropoligie Méditerranéennes 6: 239-259. k nutsson h., k nutsson k., m olin f. & z etteRlund p. 2016. — From flint to quartz: organization of lithic technology in relation to raw material availability during the pioneer process of Scandinavia. Quaternary International 424: 32-57. https://doi. org/10.1016/j.quaint.2015.10.062 knutsson k. & lindgRen c. 1999. — Making sense of quartz. Presentation and results of an experimental analysis applied to quartz from a number of sites in Söodertöom. Interdisciplinary Investigations of Stone Age Sites in Eastern Middle Sweden, The Swedish National Heritage Board, Archaeological excavations Dept, Gävle: 5-36 knutsson k. 2014. — ‘Simple’ need not mean ‘archaic.’ Antiquity 88 (341): 950-953. https://doi.org/10.1017/S0003598x00050894 kobayashi h. 1975. — The Experimental Study of Bipolar Flakes, in swanson E. H. (ed.), Lithic Technology: Making and Using Stone Tools. De Gruyter Mouton, Berlin, New York: 115-128. https://doi.org/10.1515/9783111390376.115 lazuén t., fabRegas R., lombeRa a. & pedRo RodRiguez X. 2011. — La gestión del utillaje de piedra tallada en el Paleolítico Medio de Galicia. El nivel 3 de Cova Eirós (Triacastela, Lugo). Trabajos De Prehistoria 68 (2): 237-258. https://doi.org/10.3989/ tp.2011.11068 leRoi-gouRhan a. 1964. — Le geste et la parole. I. Technique et langage. II. La mémoire et les rythmes. Les Études Philosophiques 20 (3): 360-360. manninen m. a. 2016. — The effect of raw material properties on flake and flake-tool dimensions: a comparison between quartz and chert. Quaternary International 424: 24-31. https://doi. org/10.1016/j.quaint.2015.12.096 584 COMPTES RENDUS PALEVOL • 2025 • 24 (28) Melo M. N. R. et al. maRks a., shokleR J. & zilhão J. 1991. — Raw material usage in the Paleolithic the effects of local availability on selection and economy. University of Kansas Publications in Anthropology 19: 127-140. maRks a., monigal k. & zilhão J. 2001. — The Lithic Assemblages of the Late Mousterian at Gruta da Oliveira, Almonda, Portugal. Les premiers hommes modernes de la Péninsule Ibérique. Actes du Colloque de la Comission VIII de l’UISPP, Vila Nova de Foz Côa, Octobre 1998. Instituto Português de Arqueologia, Lisboa: 145-154. máRquez b., mosqueRa m., péRez-gonzález a., aRsuaga J. l., baquedano e., paneRa J., espinosa J. & gómez J. 2013. — Evidence of a neanderthal-made quartz-based technology at Navalmaíllo rockshelter (Pinilla del Valle, Madrid Region, Spain). Journal of Anthropological Research 69 (3): 373-395. https://hdl. handle.net/20.500.14352/44669 m áRquez b., b aquedano e., p éRez -g onzález a. & a Rsuaga J. l. 2016. — Microwear analysis of Mousterian quartz tools from the Navalmaíllo Rock Shelter (Pinilla del Valle, Madrid, Spain). Quaternary International 424: 84-97. https://doi.org/10.1016/J. quaint.2015.08.052 mathias c., bouRguignon l., bRenet m., gRégoiRe s. & moncel m.-h. 2020. — Between new and inherited technical behaviours: a case study from the Early Middle Paleolithic of Southern France. Archaeological and Anthropological Sciences 12: 146 https://doi.org/10.1007/s12520-020-01114-1 mathias c., beRnaRd-guelle s., bouRguignon l. & viallet c. 2021. — On Knapping Pebbles. Impact of Pebble Morphologies on Debitage Initialisation and Differences with Ramification Processes. 13th International Symposium on Knappable Materials, Tarragona, Spain. https://hal.science/hal-03380640v1 melo m. n. R. 2023. — Estudo tecno-económico da indústria lítica em quartzo do Paleolítico Médio (MIS 5) da Gruta da Figueira Brava (Setúbal, Portugal). M.A. dissertation, Universidade de Lisboa. http://hdl.handle.net/10451/57796 mesfin i., benJamim m.-h., lebataRd a.-e., saos t., pleuRdeau d., matos J. & lotteR m. 2023. — Evidence for Earlier Stone Age ‘coastal use’: the site of Dungo IV, Benguela Province, Angola. PLOS One 18 (2): e0278775. https://doi.org/10.1371/ journal.pone.0278775 mielgo c., huguet R., laplana c., maRtín-peRea d. m., moclán a., máRquez b., aRsuaga J. l., péRez-gonzález a. & baquedano e. 2024. — Intra-site spatial approaches based on taphonomic analyses to characterize assemblage formation at Pleistocene sites: a case study from Buena Pinta Cave (Pinilla del Valle, Madrid, Spain). Archaeological and Anthropological Sciences 16 (5): 1-28. https://doi.org/10.1007/s12520-023-01913-2 m oRa R. 1984. — Estudio tecnológico de los complejos líticos al aire libre de la comarca de la Selva (Avellaners y Diable Coix) y comparación con Arbreda H43 (Serinyà). M.A. thesis, University of Barcelona. m ouRRe v. 1996. — Les industries en quartz au Paléolithique. Ter - minologie, méthodologie et technologie. Paléo, Revue d’Archéologie Préhistorique 8: 205-223. https://doi.org/10.3406/pal.1996.1160 mouRRe v. & thiébaut c. 2008. — L’industrie lithique du Moustérien final de la Grotte du Noisetier (Fréchet-Aure, HautesPyrénées) dans le contexte des Pyrénées centrales françaises. Treballs d’Arqueologia 14: 87-104. mouRRe v., JaRRy m., david c. & lelouvieR l.-a. 2011. — Le débitage sur enclume aux Bosses (Lamagdelaine, Lot, France). Paleo, Revue d’Archéologie Préhistorique, Numéro spécial: 49-62. https://doi.org/10.4000/paleo.1894 muñoz del pozo a., gómez de soleR b., bustos-péRez g., maRía gema c., picin a., blasco R., Rivals f., Rufà a. & Rosell aRdèvol J. 2023. — Analysis and classification of Middle Paleolithic lithic raw materials from Teixoneres Cave: project overview and initial results. Acta Imeko 12 (3): 1-7. https://doi. org/10.21014/actaimeko.v12i3.1494 paRgeteR J. & de la peña p. 2017. — Milky quartz bipolar reduction and lithic miniaturization: experimental results and archaeological implications. Journal of Field Archaeology 42 (6): 551-565. https://doi.org/10.1080/00934690.2017.1391649 p aRgeteR J., k hReisheh n. & s tout d. 2019. — Understanding stone tool-making skill acquisition: experimental methods and evolutionary implications. Journal of human evolution 133: 146166. https://doi.org/10.1016/j.jhevol.2019.05.010 peReiRa t. & benedetti m. m. 2013. — A model for raw material management as a response to local and global environmental constraints. Quaternary International 318: 19-32. https://doi. org/10.1016/j.quaint.2013.04.011 Ramos p., aubRy t., monteiRo-RodRigues s. & sanches m. 2024. — A ocupação do Vale do Côa pelo Homem de Neandertal: o sítio arqueológico da Cardina-Salto do Boi (Vila Nova de Foz Côa, Portugal) como caso de estudo. Estudos Do Quaternário Quaternary Studies 24: 34-42. https://doi.org/10.30893/eq.v0i24.229 Raposo l. & caRdoso J. 2000. — Mousterian Industries of the Gruta da Figueira Brava. Memórias da Academia das Ciências de Lisboa, Classe de Ciências Tomo Xxxviii: 319-337. R odRiguez -a lvaRez X.-p., de l ombeRa -h eRmida a., v alcaRce R. & feRnández l. 2011. — The Upper Pleistocene site of Cova Eirós (Triacastela, Lugo, Spain), in de lombeRa-heRmida a. & fábRegas valcaRce R. (eds), To the West of Spanish Cantabria: The Paleolithic Settlement of Galicia. Archaeopress (BAR International series; 2283), Oxford: 123-133. RodRíguez-Rellán c. 2016. — Variability of the rebound hardness as a proxy for detecting the levels of continuity and isotropy in archaeological quartz. Quaternary International 424: 191-211. https://doi.org/10.1016/j.quaint.2015.12.085 Rolland n. & dibble h. l. 1990. — A new synthesis of Middle Paleolithic variability. American Antiquity 55 (3): 480-499. https://doi.org/10.2307/281279 saville a. & ballin t. b. 2008. — Quartz technology in Scottish prehistory. Scottish Archaeological Internet Reports 26: 1-101. https://doi.org/10.9750/issn.2056-7421.2008.26.1-101 spRy c., kuRpiel R., foley e. & penzo-kaJewski p. 2021. — Revisiting the “Quartz Problem” in lithic studies: a review and new, open-access, experimental dataset. Lithic Technology 47 (2): 171-181. https://doi.org/10.1080/01977261.2021.1981655 tallavaaRa m., manninen m. a., heRtell e. & Rankama t. 2010. — How flakes shatter: a critical evaluation of quartz fracture analysis. Journal of Archaeological Science 37 (10): 2442-2448. https://doi.org/10.1016/j.jas.2010.05.005 teXieR p. 1981. — Désilification (sic) des silex taillés. GéoProdig, portail d’information géographique. http://geoprodig.cnrs.fr/ items/show/159575 thiébaut c., costamagno s., coumont m.-p., mouRRe v., pRovenzano n. & théRy-paRisot i. 2010. — Approche expérimentale des conséquences du piétinement des grands herbivores sur les vestiges archéologiques. Mise en commun des approches en taphonomie = Sharing taphonomic approaches: actes du workshop n° 16 - XVe congrès international de l’UISPP, UISPP, Sep 2006, Lisbonne, Portugal: 109-130. https://shs.hal.science/ halshs-00585961v1 tiXieR J 1978. — Méthode pour l’étude des outillages lithiques: notice sur les travaux scientifiques de J. Tixier. Doctorat D’État, Université de Paris X-Nanterre, 118 p. tiXieR J. & tuRq a. 1999. — Kombewa et alii. Paléo, Revue d’archéologie Préhistorique 11: 135-143. vaqueRo m. 1999. — Intrasite spatial organization of lithic production in the Middle Palaeolithic: the evidence of the Abric Romaní (Capellades, Spain). Antiquity 73 (281): 493-504. https://doi. org/10.1017/S0003598X00065054 vaqueRo m. & Romagnoli f. 2017. — Searching for lazy people: the significance of expedient behavior in the interpretation of paleolithic assemblages. Journal of Archaeological Method and Theory 25: 334-367. https://doi.org/10.1007/s10816-017-9339-x 585 Figueira Brava Middle Paleolithic quartz assemblage COMPTES RENDUS PALEVOL • 2025 • 24 (28) y e Ş ilova g. c., a RRoyo a., v eRgès J. m. & o llé a. 2024. — New approaches to the bipolar flaking technique: qualitative, quantitative, and kinematic perspectives. Journal of Archaeological Method and Theory 31 (3): 1333-1382. https://doi.org/10.1007/s10816-024-09639-8 zilhão J. 2012. — Neandertals from World’s end: Results of recent research, in tuRbón d., fañanás l., Rissech c. & Rosa A. (eds), Biodiversidad humana y Evolución. Universidad de Barcelona, Sociedad Española de Antropología Física, Barcelona: 68-77. zilhão J., angelucci d. e., igReJa m. a., aRnold l. J., badal e., callapez p., caRdoso J. l., d’eRRico f., dauRa J., demuRo m., deschamps m., dupont c., gabRiel s., hoffmann d. l., legoinha p., matias h., monge soaRes a. m., nabais m., p oRtela p., q ueffelec a., R odRigues f. & s outo p. 2020. — Last Interglacial Iberian Neanderetals as fisher-hunter gatherers. Science 367 (6485): eaaz7943. https://doi.org/10.1126/ science.aaz7943 Submitted on 12 May 2025; accepted on 20 September 2025; published on 10 December 2025.