Fallen and Lost into the Abyss? A Mesolithic Human Skull from Sima Hedionda IV (Casares, Málaga, Iberian Peninsula)

Research Article Rafael M. Martinez-Sanchez*, María Dolores Bretones-García, Cristina Valdiosera, Juan Carlos Vera-Rodríguez, Inmaculada López Flores, María D. Simón-Vallejo, Pilar Ruiz Borrega, María J. Martínez Fernández, Jorge L. Romo Villalba, Francisco Bermúdez Jiménez, Rafael Martín de los Santos, Salvador Pardo-Gordó, Miguel Cortés Sánchez Fallen and Lost into the Abyss? A Mesolithic Human Skull from Sima Hedionda IV (Casares, Málaga, Iberian Peninsula) https://doi.org/10.1515/opar-2022-0267 received April 15, 2021; accepted October 19, 2022 Abstract: The presence of scattered prehistoric human bones in caves and sinkholes is common in many regions of Iberia. These are usually interpreted as erratic elements coming from burial contexts, usually collective associations. These burial contexts are very frequent in karst areas of the Iberian Peninsula since the Early Neolithic, mostly in the Late Neolithic, and Copper Age, while findings from earlier chronologies are much more unusual. In this work, we present partial remains of a human skull from the Mesolithic period, recovered from a cave in the Strait of Gibraltar area. Although there is no conclusive evidence pointing to a dismantled burial context, this constitutes an isolated find, where its final location appears to be consistent with gravitational fall followed by water transportation. Keywords: human remains, Mesolithic, Southern Iberia, sinkholes, radiocarbon  * Corresponding author: Rafael M. Martinez-Sanchez, Department of History, University of Cordoba, Filosophy and Letters, Pl. Cardenal Salazar 3, Cordoba, Córdoba 14003, Spain, e-mail: [email protected] María Dolores Bretones-García: Territorial Delegation of Cultural Heritage, C/Martínez-Montañés 8, 23007 Jaén, Spain Cristina Valdiosera: Departamento de Historia, Geografía y Comunicación, Universidad de Burgos, Burgos, Spain Juan Carlos Vera-Rodríguez, María J. Martínez Fernández: Departamento de Historia, Geografía y Antropología, Universidad de Huelva, Huelva, Spain Inmaculada López Flores: Independent Researcher, Sevilla, Spain María D. Simón-Vallejo, Miguel Cortés Sánchez: Departamento de Prehistoria y Arqueología, Universidad de Sevilla, Sevilla, Spain Pilar Ruiz Borrega, Francisco Bermúdez Jiménez: Independent Researcher, Córdoba, Spain Jorge L. Romo Villalba, Rafael Martín de los Santos: Independent Researcher, Málaga, Spain Salvador Pardo-Gordó: Departamento de Geografía e Historia, Universidad de La Laguna, Santa Cruz de Tenerife, Spain ORCID: Rafael M. Martinez-Sanchez 0000-0002-8947-117X; Cristina Valdiosera 0000-0003-4948-2226; Juan Carlos Vera- Rodríguez 0000-0002-5989-2694; María J. Martínez Fernández 0000-0003-3049-0082; Salvador Pardo-Gordó 0000-0002- 1060-1526  Special Issue Published in Cooperation with Meso'2020 –Tenth International Conference on the Mesolithic in Europe, edited by Thomas Perrin, Benjamin Marquebielle, Sylvie Philibert, and Nicolas Valdeyron. Open Archaeology 2022; 8: 892–904 Open Access. © 2022 the author(s), published by De Gruyter. This work is licensed under the Creative Commons Attribution 4.0 International License. 1 Introduction While the Neolithization process in Iberia has been largely explored, little is known about the archaeology of the last hunter-gatherers of the southern region. This has led to an apparent void that only seems to disappear with the irruption of the first Neolithic horizon in this territory. In Andalucía, the oldest Neolithic contexts are found in the coastal and nearby areas, indicated by radiocarbon dates on domestic caprines in Cueva de Nerja (Málaga)(NV2 Silo)and Cueva de la Dehesilla (15c–16b)(Algar, Cádiz). Both radiocarbon dates concur around 5500 cal. BC, marking the timing of the transitional horizon Mesolithic/Neolithic in southern Iberia (Aura Tortosa et al., 2013; García Rivero et al., 2018). After that, during the first centuries of farming strategies in the region, a high representation of archaeological contexts and sites is noted across southern Iberia, with a higher density of data from the last quarter of the sixth-millennium cal. BC (Martín Socas, Camalich Massieu, Caro Herrero, & Rodríguez-Santos, 2018). The apparent scarcity of Epipalaeolithic/Mesolithic contexts is further increased by a gap between the end of the seventh and the first half of the sixth-millennium cal. BC. What caused this phenomenon does not seem to be easy to answer, but it has important consequences in the study and interpretation of the Neolithic process that occurred in the territory. Research on the few available contexts has been undertaken in conjunction with the analysis of the neolithization process in this region (Aura Tortosa et al., 2009). In southern Valencia, the Late Mesolithic is better known in the Jucar and Segura basins, where human occupations have been interpreted in terms of autochthons vs colonists (i.e., Mesolithic vs Neolithic). In this area, the Neolithic occupation of previously empty spaces has been observed, in the case of Falguera and Tossal de la Roca, where a hiatus extends to the Late Mesolithic. Moreover, sites with Late Mesolithic occupation seem to lack Early Neolithic settlements (such as Arenal de la Virgen around the Villena lagoon, Alicante)(Juan Cabanilles & Martí Oliver, 2017). The Andalusian Late Mesolithic is virtually unknown, except for the dates obtained at Retamar (Puerto de Santa María, Cádiz)(Ramos Muñoz & Lazarich González, 2002)and Cañada Honda (Aljaraque, Huelva) (Reis, Campos Jara, & Soares, 2020), in both cases on marine shell and, in Retamar, showing a high standard deviation and a succeeding Early Neolithic Phase. Throughout the region, the evidence of occupation seems to be limited to the coastal strip and no later than the end of the seventh-millennium cal. BC. At Cueva de Nerja (Malaga), Late Mesolithic stratigraphic sequences are found in two of the three excavated areas, the Sala del Vestibulo (NV)and the Sala Torca (NT), with charcoal dates from the mid-seventh millennium and beginning of the sixth-millennium cal. BC (Aura Tortosa et al., 2009, 2013). Similarly, the stratigraphic sequence of Bajondillo rock shelter (Torremolinos, Málaga), provided two dates falling in this chronological range (end of seventh-millennium cal. BC)(Cortés-Sánchez, 2007), associated with scarce lithic and ornaments elements (Martí Oliver, Aura Tortosa, Juan Cabanilles, García Puchol, & Fernández López de Pablo, 2009). Some pre-neolithic series from the north-eastern mountain ranges of Andalucía (Cueva del Nacimiento, Pontones, Jaén)show microblade technology (Asquerino Fernández & López, 1981). For instance, there is a Mesolithic phase identified in Cueva Ambrosio (Almería)(Suárez Márquez, 1981)with no associated 14 C dates, and some microliths (triangles)found in Alcolea (Córdoba) in a redeposit of much later chronology (Martínez Sánchez, 2013). In southern Portugal, besides well-knownshellmiddensitesintheTagus(Muge)and Sado river valleys, occupational contexts are almost exclusively restricted to coastal or near-coastal sites. Oliveirinha, Pedra do Patacho, Fiais, Palheiroes do Alegra or Samouqueira sites on the Alentejo coast, and Castelejo and Rocha das Gaivotas at Cabo de São Vicente are composed of accumulations of biomaterials, mostly malacological elements. However, unlike other Iberian regions, 14 C dates obtained from Portuguese sites do not seem to show gaps in human occupation throughout the Late Mesolithic, although some interruptions havebeenobservedintheseventh-millennium cal. BC (Cabo de San Vicente sites)(Carvalho, 2009; Dean, Valente, & Carvalho, 2012; Valente & Carvalho, 2009). In any case, lithic morphotypes of the Boreal and early Atlantic in southern Iberia remain poorly known, with very few sites having provided abundant material assemblages. There, except on the Alentejo and Algarve coast (Portugal), the Sauveterrian and microblade elements are not clearly distinguished from the geometric component, reflecting the existence of multi-sequential assemblages, A Mesolithic Human Skull from Extreme South Iberia. Sima Hedionda IV 893 sedimentary gaps, and residual collections intermixed with elements of later chronology (Aura Tortosa et al., 2009). A clear example in this regard is that some surface assemblages of inner Andalusia have smallblade industries previously attributed to the Epipalaeolithic, though now identified as Early Neolithic (Asquerino Fernández, 1986, 1987a,b; Gavilán Ceballos, 1987). 2 Sima Hedionda IV Archaeological Context The Sima Hedionda II/IV is a sinkhole in the Sierra de la Utrera karst formation, just over 5 km from the current coastline, in the Municipality of Casares (at the western edge of the Malaga coast)and located in the historic region of the Strait of Gibraltar (Figure 1). Sierra de la Utrera is a small mountain range comprising a calcareous plateau, highly modelled by karstification processes. This is made up of rocks of the Kimmeridgian-Oxfordian lithological top stage (Upper Jurassic, Torcal Formation, up to an altitude of 350 m a.s.l.) and a lower one (base), made up of a base of Middle Jurassic oolite limestones (Endrinal Formation) (Martín-Algarra, 2006). There are several large caves, some of which are archaeological sites, such as Gran Duque Cave, with a rich archaeological record dating from the Early Neolithic (Ferrando de la Lama, 1987). The one known as Sima Hedionda II or Sima Pito was located in the early 1970s, although Figure 1: (a)Iberian Peninsula, indicating southern sites with published Mesolithic human remains. 1: Muge shell middens (Moita do Sebastião, Cabeço da Arruda, Cabeço do Amoreira, Fonte do Padre Pedro, Flor da Beira, and Cova da Onça, Ribatejo); 2: Sado sites (Cabeço das Amoreiras, Arapouco, Cabeço do Pez, Alentejo); 3: Samouqueira (Alentejo); 4: Fiais (Alentejo); 5: Vale Boi (Algarve); 6: Sima Hedionda II/IV (Casares Málaga); 7: Cueva del Esqueleto (Cortes de la Frontera, Málaga); 8:, Cueva del Higuerón-Tesoro (Cala del Moral, Málaga); 9: Cueva de Nerja (Maro, Málaga); 10: Cueva de la Cariguela (Píñar, Granada); 11: Casa Corona (Villena, Alicante); 12: Coves de Santa Maira (Castell de Castells, Alicante); and 13: El Collado (Oliva, Valencia). (b)Aerial view of Utrera Karstic formation, with Sima Hedionda II/IV site (white star)(Casares, Málaga). 894 Rafael M. Martinez-Sanchez et al. it was not until 2005 that it was completely explored. Then, a new system of galleries was found (called Hedionda IV), leading to the discovery of several archaeological remains. Both sinkholes (Sima Hedionda II and IV)are connected with each other forming a large cave system with nearly 100 m of slope and over 2,168 m of passage (Figure 2). The current entrance is located at 91 m above sea level, while its bottom gallery goes below. Between 2014 and 2015, this system was intensively explored through an archaeological survey conducted by one of us (MD Bretones), during which the inner topography was recorded. Inside, and along with different chambers and galleries, several archaeological deposits have been detected, most of which date back to the Early Neolithic (second half of the sixthmillennium cal. BC)(Bretones García et al., in press). Archaeological assemblages are located in different places, often in well-defined horizontal rooms and corridors, though some of them have also been documented at the end of vertical and inaccessible chasms, as a result of intentional displacements or episodic gravitational movements and dragging. In this case, there is evidence for use of the cave prior to the Early Neolithic, making it one of the few sites with Mesolithic human remains from southern Iberia. 3 Sector 7 (SHII/IV/7001–7006)Human Skull Though skeletal material was found in upper areas of the cave, it was in an Early Neolithic context. Of particular interest is a set of human remains placed in a deep area (61 m a.s.l.), within a vertical fissure, 41 m deep, with a narrow base that was impossible to reach and far from the closest evidence for human use. This space was named sector 7, a small gallery currently accessible only from the cave’s main hall, ending in a slope leading to a periodically flooded collector (Figure 3). Figure 2: Plan (left)and elevation (cross-section)(right)of Sima Hedionda IV Cave. The figure shows the finding place (sector 7): a black star in the plan, and a black arrow in elevation (cross-section). Topography, drawings, and reference data by GES-Casares cavers club. A Mesolithic Human Skull from Extreme South Iberia. Sima Hedionda IV 895 We detected five fragments, probably all part of the same skull. The largest bone fragment is a very robust frontal bone that initially seemed to be isolated (7001)(Figure 4). It shows a well-developed glabella and a strong supraorbital notch with numerous nutrient foramina. It also presents very large paranasal Figure 3: (a)Finding of human skull remains in sector 7 (Sima Hedionda II/IV Cave), embedded in a crack. (b)View of the frontal part of the human skull (7001)on-site. (c)View and position of highly cemented skull fragments placed below the frontal bone (once removed): 7003, maxilla fragment; 7004 and 7005, cranial vault fragments; 7006, temporal bone. Figure 4: (a)View in norma frontalis of the frontal bone 7001 and (b)view in norma lateralis. 896 Rafael M. Martinez-Sanchez et al. sinuses. Likewise, it reveals very high porosity around the supraciliary arches and close to the coronal suture (not preserved), and periosteal bulges with irregular and diffuse porosity. Although nasal bones and specific measurement points were lost, interorbital width should be wide, based on the remaining portion (Figures 3–5). The cranial wall is quite thick and the endocranial surface shows a high development of the frontal crest (Figure 5). Other fragments (Figure 3c)were found at the base of the fissure and cemented together by carbonate flowstone. These were two pieces of the cranial vault (7004 and 7005), a portion of the left temporal bone (7006), and a maxilla fragment (7003), retaining part of the right molar row (RM 1 and RM 2 ), showing slight wear in the occlusal plane. The absence of RM 3 could indicate non-eruption or agenesis. All remains are consistent with a single, very robust, adult male individual. Figure 5: CT scan ectocranial views of frontal bone 7001 (by OsiriX DICOM Wiever):(a)norma verticalis,(b)superior part (endocranial),(c)norma frontalis,(d)posterior (endocranial),(e)right norma lateralis,(f)sagittal section, and (g)left norma lateralis. A Mesolithic Human Skull from Extreme South Iberia. Sima Hedionda IV 897 4 Contextual and Taphonomic Insights Fragments 7003–7006 could not be clearly observed, given the intense calcareous formation around them, fixing the bones into the rock wall, so these were not removed. Flowstone masked and cemented the cranial remains, making safe extraction very difficult. Only frontal bone 7001 was found free from the rock and was recovered. The frontal bone surface was observed with a 640 ×480 pixel dinolite digital microscope. No cut marks or other signs of human manipulation were observed. Regarding taphonomic conditions, it shows some ancient (post-mortem)breaks inside and outside, accompanied by fine cracks on the cranial surface that have slightly deformed this bone, mainly on the supraorbital flange. Other taphonomic conditions such as highly cemented surfaces and polished areas are observed. Inside the cranial vault (endocranial), as well as on the external bone surface, we can recognize patchy pigmentations of metallic oxides, most probably manganese. These are accompanied by small cracks and sinuous or star-shaped fractures (Figure 6). Small cracks characterized by curled-up edges, probably were caused by changes in humidity in the karstic environment (Fernández-Jalvo & Andrews, 2016)suggesting swelling and shrinkage of the bone tissue over time. Brightness or polishing located in some areas of the external vault may be caused by slight abrasion with fine sand/clay sediment (Figure 6). This, together with manganese dots on the bone surface (López-González, Grandal-d´Anglade, & Vidal-Romaní, 2006)and the existence of fine sand and its arrangement on the hall floor around the location of the find, suggest episodic flooding and drainage inside the lower part of the cave, involving water transport events. Figure 6: Four different views of the exterior cranial surface in the frontal bone 7001 (640 ×480 pixels Dinolite, 1 mm scale). 1–4: micro-cracks extended over the bone cortex; 1–3: black dots (probably manganese dioxide staining); 1 and 4: slightly bright over the frontal surfaces. 898 Rafael M. Martinez-Sanchez et al. 5 Chronometric Analysis Considering the exceptional nature of this find, we also proceeded to radiocarbon date a sample obtained from RM 2 (dentin), a subsample obtained from the portion selected for archaeogenetic analysis. While initial DNA screening of this individual yielded low quantities of endogenous DNA (<1%)for genomic analysis, it was possible to determine the biological sex as male. Micro-sample dating of collagen confirmed a Mesolithic chronology. The result (Beta-424650)was an AMS date of 7600 ±30 BP (Table 1, Figure 7). Organic content from the tooth sample used for radiocarbon dating and archaeogenetics gave δ 13 C but not δ 15 N results. We then attempted to obtain results from a frontal fragment (3.5 g), using the protocol of Bocherens et al. (1991)and the IACT-CSIC facilities (Stable Isotopes Biogeochemistry Laboratory, Granada, Spain). However, no results were obtained due to the lack of usable collagen; no further sampling was attempted. IRMS δ 13 C values on the tooth (−17.0; Table 1)are similar to results from Mesolithic individuals from Cueva de Nerja (CNA3293.2.1, Sala Torca, level 13, “Pepita”)(Fernández Rodríguez et al., 2020)and burials 3 and 4 at El Collado (Oliva, Valencia). In these cases, while all sites are close to the coast, stable isotope analyses reveal a mostly terrestrial diet (Fernández Rodríguez et al., 2020; Gibaja Bao et al., 2015), like other non-Iberian coastal examples from the Western Mediterranean (Goude et al., 2016). The dating was carried out by Beta in 2015, with ultrafiltration, but % collagen and the C:N value were not reported at that time. We have calibrated the above using two approaches. The first, Model A in Figure 7 and Table 1, follows the IntCal20 (Reimer et al., 2020)atmospheric curve and assumes a fully terrestrial diet. Model B follows a mixed curve and marine reservoir effect. We used a ΔRof 94 ±61, which comes from the information available in the 14CHONO Marine 20 database related to the West Mediterranean. This offset is the same used in other studies concerning the early Holocene of the Iberian Mediterranean coast (Brisset, Bujarchs, Ballesteros Navarro, & Fernández López de Pablo, 2018), including the cases of Mesolithic individuals from Nerja and El Collado (Fernández Rodríguez et al., 2020). The marine diet value of 20% used here is close to those of the Cueva de Nerja individual and El Collado burials 3 and 4. δ 13 C values are −17.35 for Cueva de Nerja and −17.6 for the two El Collado individuals. On the basis of these sites, we are assuming that the Mesolithic population has a marine dietary intake (Salazar García et al., 2014). Table 1 and Figure 7 include all published data for Mesolithic human remains in Andalucía. 6 Discussion Cranial fragments found in Sima Hedionda IV zone 7 (SHII/IV/7001-6), are among the few human remains dated to the Mesolithic in southern Iberia; no further material was found. This is the southernmost site with Mesolithic human remains in Spain. Both location and taphonomy suggest a natural percolation from Table 1: Numeric data of the three 14 C datings on Mesolithic human remains in Andalusia using the two calibration models Site Code BP SD δ 15 N‰ (Air-N2) δ 13 C‰ (V-PDB) BC cal. 68.2% BC cal. 95.4% Model C.Esqueleto Beta-324381(?)7560 40 n.d. n.d. 6452–6401 6475–6266 A 6372–6256 6404–6230 B S.Hedionda 7003-RM 2 Beta-424650 7600 30 n.d. −17 6463–6431 6498–6411 A 6412–6268 6421–6250 B Nerja-Pepita CNA3293.2.1 7907 39 11.55 −17.35 6906–6654 7032–6647 A 6685–6588 6744–6498 B CNA-3293: Nerja “Pepita,”Beta-424650: Sima Hedionda IV (7003-RM 2 ), Beta-324381(?): Esqueleto Cave. Model A: atmospheric curve (Intcal20)(Reimer et al., 2020); Model B: mixed with a 20% of marine curve (Heaton et al., 2020), including aΔRvalue of 94 ±61. Software used, Oxcal4.4.4. A Mesolithic Human Skull from Extreme South Iberia. Sima Hedionda IV 899 above, once skeletonized, including water transport (Belcastro et al., 2021). Location and condition of the cranial finds preclude meaningful comment on their origin, be it burial, funerary assemblage, or chance event such as an accident. No evidence exists that allows comment on cultural elements associated with the find, or their absence. The frontal bone was the only fragment that could be extracted and observed in detail. It is notable for its robusticity, with characteristics defined in similar cases as “palaeomorphic,”showing a great development of the supraorbital region, with a heavy glabella and thick supraorbital margins, and wide supraorbital notches. Analogous skull traits, rare in Neolithic and later Iberian examples, have been observed in other Iberian Mesolithic skulls, such as Cuartamentero Cave (Asturias)(Drak Hernández, 2016; Garralda Benajes, 1982)or the individual 1 (adult male)from the collective burial of Cingle del Mas Nou (Ares del Maestrat, Castelló)(Olaria Puyoles, 2020). Some of these morphologies have been noted as common in other examples from Western Europe (Drak Hernández, 2016),reflecting nutritional and adaptive patterns shared by hunter-gatherers (Katz, Grote, & Weaver, 2017). In Iberia, burial deposits from this period are very rare. Even today, the Muge and Sado sites in Portugal make up the vast majority of the skeletal remains known since the first finds in the 1860s (Peyroteo-Stjertna, 2016). However, in recent times, the discovery of Mesolithic human remains has become increasingly frequent, both as complete burials (La Braña, León)(Vidal Encinas, Fernández Rodríguez, Prada Marcos, & Fuertes Prieto, 2008)or as a result of accidental events (Chan do Lindeiro)(Serrulla Rech & Sanín Matías, 2017). True necropoli or collective burials have also been identified, such as El Collado (Oliva, Valencia) (Gibaja Bao et al., 2015), Casa Corona (Villena, Alicante)(Fernández López de Pablo, Gómez Puche, Ferrer García, & Yll Aguirre, 2011), Cingle del Mas Nou (Ares del Maestrat, Castelló)(Olaria Puyoles, 2020), and Los Canes (Asturias)(Arias Cabal et al., 2009), while other more fragmentary contexts include apparent Figure 7: Multi-plots and sums applying the two calibration models. CNA-3293: Nerja “Pepita,”Beta-424650: Sima Hedionda IV (7003-RM 2 ), Beta-324381(?)(lab code to be confirmed): Esqueleto Cave. Model A: atmospheric curve (Intcal20)(Reimer et al., 2020); Model B: mixed with a 20% of marine curve (Heaton et al., 2020), including a ΔRvalue of 94 ±61. Software used, Oxcal4.4.4. 900 Rafael M. Martinez-Sanchez et al.