the venuses block from arlanpe cave nort

JOSEBA RIOS-GARAIZAR, DIEGO GARATE, RAPHAËLLE BOURRILLON,ASIER GÓMEZ-OLIVENCIA AND THEODOROS KARAMPAGLIDIS

THE VENUSES BLOCK FROM ARLANPE CAVE (NORTHERNIBERIAN PENINSULA): IMPLICATIONS FOR THE ORIGINSAND DISPERSION OF GÖNNERSDORF-LALINDE STYLEDEPICTIONS THROUGHOUT THE EUROPEAN MAGDALENIAN

Summary. In 2011, an engraved limestone block was found in the cave ofArlanpe (Lemoa, northern Iberian Peninsula). One of the figures representedon it was identified as a schematic feminine representation similar to those ofthe Gönnersdorf-Lalinde style. The stratigraphical position of the block is nottotally clear owing to severe disturbance in the Upper Pleistocene depositslocated near the entrance sector of the cave. Nevertheless, the most probablestratigraphical correlation is with Level I, which has been dated to thebeginning of the Middle Magdalenian. This finding extends the distributionrange of this kind of representation to the northern Iberian Peninsula, where,up to now, only two other, less clear, Gönnersdorf-Lalinde style representationshave been found. It also extends its chronological range, pushing it back tothe beginning of the Middle Magdalenian. In this paper, we present thearchaeological context of the engraved block, followed by a detaileddescription of technological and stylistic features. These data will be used todiscuss the implication of this discovery for an understanding of the origins,expansion and diffusion of this kind of feminine representation across Europe.

INTRODUCTION

The representation of feminine figures in both parietal and portable art began during theEarly Upper Palaeolithic (Conard 2009; White et al. 2012), and is one of the most commonthemes in European Upper Palaeolithic artistic expression. These representations are widespreadacross Europe, with a certain degree of variability in means of expression, style, context andsignificance. Such differences are more evident when comparing Initial and Middle UpperPalaeolithic female figurines (Willendorf style) with Late Upper Palaeolithic ones (Gönnersdorf-Lalinde style) (Gaudzinski and Jöris 2015).

The Gönnersdorf-Lalinde style figurines are characterized by several traits, such as therepresentation of profile silhouettes, with exaggerated buttocks, occasionally showing breasts,

OXFORD JOURNAL OF ARCHAEOLOGY 34(4) 321–341 2015© 2015 John Wiley & Sons Ltd. 321

and, when present, very schematic arms and heads. These figurines usually appear formingcompositions of at least two figurines and are present in both parietal and portable art (Bourrillon2009; Bourrillon et al. 2012; Bosinski et al. 2001; Delluc and Delluc 1995). This style offeminine representation is commonly associated with the Late Magdalenian, and is presentbasically only in France and Germany. After a period, during the Middle Magdalenian, when asort of stylistic unity was observed in the Franco-Cantabrian region, the scarcity of this kind ofrepresentation in the Cantabrian region, and by extension in all of the Iberian Peninsula, has beeninterpreted as part of a process of cultural differentiation between the Magdalenian on either sideof the Pyrenees (Fritz et al. 2007).

The particular spatial distribution of Gönnersdorf-Lalinde representations has long beenrelated to the definitive recolonization process that started in France and extended towards thenorthern plains, already free from the influence of ice sheets (Jochim et al. 1999; Terberger andStreet 2002; Demars 2008). The dispersion and persistence of these symbols have been relatedto the need by ‘pioneering’ populations to maintain social networks with the original territories(Gaudzinski and Jöris 2015).

In this paper, new evidence for these types of engravings is presented from Arlanpe Cave(eastern Cantabrian region). A limestone block with one clear feminine engraving and two otherprobable feminine representations, and its association with Middle Magdalenian occupations inthe cave, introduce new elements of discussion about the chronology and geographical origins ofthese types of representations. It also serves as a basis for discussion about the dynamics ofcultural exchanges and contacts between south-west France and the Cantabrian region during theMiddle and Upper Magdalenian. The site’s function, combining the information from lithic andfaunal assemblages with the information derived from the engraved block, will also be assessed.This will be set in relation to the existence of ritualized spaces in the Cantabrian Magdalenian(Arias 2009) in order to give some insights into the function and symbolic meaning of these typesof feminine representations in this particular context.

SITE LOCATION AND DESCRIPTION

The cave site of Arlanpe (UTM coordinates: 30T 519254, 4782262, 204 m.s.l.) islocated on the north-west slopes of the Aramotz massif, 15 km from Bilbao, the capitalcity of the Basque region of Bizkaia, in the north of the Iberian Peninsula (Fig. 1).The landscape around the cave is very mountainous, except on the valley floors of the Arratiaand the Ibaizabal rivers. The natural routes created by these valleys connect the current coastwith the mountain passes that lead to the Alavese Plateau and the Upper Ebro Valley. Thus,Arlanpe is located in the centre of one of the major Palaeolithic communication routes in theeastern Cantabrian region. This area itself acted, with varying intensity throughout thePalaeolithic, as a crossroad between Continental Europe and the Iberian Peninsula(Arrizabalaga 2007).

The cave itself opens in the Albian limestone; it is elongated in plan with three widerspaces (Fig. 2) (maximum dimension 25 x 6 m). The site was discovered in 1963 but systematicexcavation and research only started in 2006. The archaeological work has revealed a complexstratigraphy containing evidence of Late Roman (Gutiérrez Cuenca et al. 2012), Bronze Age,Middle Magdalenian, Late Upper Solutrean (Rios-Garaizar et al. 2013), Gravettian and EarlyMiddle Palaeolithic (Rios-Garaizar et al. 2011) levels.

THE VENUSES BLOCK FROM ARLANPE CAVE

OXFORD JOURNAL OF ARCHAEOLOGY© 2015 John Wiley & Sons Ltd.322

DESCRIPTION OF THE ENGRAVED BLOCK

The block was initially analysed using the classic system of describing engravings (Fritzand Tosello 2007). High quality photos and copies were made using image processing software.Then a high resolution 3D scan was made to obtain a precise digital model of the figure using aNextEngine 3D laser scanner housed at the CENIEH laboratory. The parameters of the scanningwere: 400 points/IN2 per second, error: 0.005′; visual field 3 x 5′. Each block face was scannedat least ten times to avoid gaps. NextEngine © software was used to align the different scanswithout any simplification or interpolation. The resulting mesh was combined with a RGB phototo create a model. This was analysed with 3D Resharper software to obtain transverse andlongitudinal sections. The model was also used to obtain a three-dimensional perspective of theblock for a better understanding of its position and the relationship between the engravings.MeshLab software was used to visualize the model and create ortho-photographs (Fig. 3).

The engraved block is a parallelepipedic fragment of limestone with two wide and longparallel sides, two narrow and long laterals, and two narrow and short extremities. The largersides present quite regular and flat surfaces, as does one of the long laterals. The rest of thelaterals are irregular. On the limestone surface, rudist fossils, parallel cracks and thin calcite

Figure 1Map of location of Arlanpe Cave in the eastern Cantabrian region together with other sites mentioned in the text (J.

Rios-Garaizar).

JOSEBA RIOS-GARAIZAR ET AL.

OXFORD JOURNAL OF ARCHAEOLOGY© 2015 John Wiley & Sons Ltd. 323

Figure 2Plan of Arlanpe Cave, with the position where the engraved block was recovered marked with a star (BURNIA

espeleologi taldea, modified by J. Rios-Garaizar).



crusts can be observed. The maximum dimensions of the block are 78 x 35 x 19 cm, with aweight around 60–70 kg. The engravings are located on one of the wide sides (Side A) and onthe regular long and narrow side (Side B) (Fig. 4).

Side A

Side A has a regular surface about 70 cm long and 38 cm wide. Several parallel cracksrun across the surface. A figure formed by several engraved lines can be observed (A1). Itoccupies a surface of 34.5 x 20 cm. Two of the lines are curved with a sub-parallel disposition,creating a fusiform figure, wider in the middle part, and with the concavity on the right side. Thecurved line on the right is crossed perpendicularly, in the lower part, by a shorter, sinuous, line.Two similar traces, in this case curved, are situated parallel in the upper part, one of themintersecting the main tracing on the left. This combination of engravings forms a singlerepresentation, probably figurative. Beside this composition two more sub-parallel lines (A2), ofsmaller size, have been traced on the upper part of Side A.

Side B

Side B is regular in surface, about 70 cm long and 19 cm wide. Three differentrepresentations can be individualized. The most interesting one is situated on the upper partadjacent to the right side (B1). The figure, 24.5 cm long, is the result of two curved lines forminga fusiform figure (Fig. 5). The line at the back is continuous and forms an ample curve whichrepresents a feminine buttock. The line located at the front intersects the previous one in theupper part and has two straight sections which join in an open angle, more or less in the middle,representing the waist and the contact between the leg and the torso (Fig. 5: A). Right in theintersection between these two lines, another, shorter, one (4.6 cm) was engraved, and below thisline, in parallel, a faint tracing was engraved (Fig. 5: D). These two lines represent the arms. Inthe upper part of the figure, two small traces, partially erased by an ancient fracture, represent aschematic head (Fig. 5: B). The legs are represented by two small traces which emergeminimally on Side A of the block (Fig. 5: C).

Figure 3Different 3D views of the block (scale 30 cm) (T. Karanpaglidis and J. Rios-Garaizar).



Figure 4Sides A and B (photo and copy) (D. Garate).



The second figure (B2) is located beside the first one, along its left side. The tracings areshallow and almost imperceptible. It is composed of several irregular lines, almost straight indelineation. Three of them seem to create a discontinuous line. Another two have a similardirection and form a slight curved line; the shorter one can be traced perpendicularly towards theother figure.

The last figure (B3) was made near the bottom of Side B. It is composed of a 15.5 cmlong, curved continuous line (concavity on the right), opposed to a sub-parallel line formed bytwo tracings. The position of these lines creates a fusiform figure.

Interpretation of the figures

The different figures traced on the limestone block from Arlanpe Cave share somecommon characteristics. They are composed of pairs of sub-parallel curved lines, most of them(except B2) created by deep tracings, and some of them (A1, B1, B2) with transverse traces(simple – B2, or double – A1, B1). Three of them (A1, B1, B3) also create fusiform figures.Among these figures, the most complete and detailed one is B1. In this case, the conventions(fusiform silhouette, V-marked waist joint, reduced breast, pronounced buttocks) fit the graphiccharacteristics of the Schematic Feminine Figures (FFS, Figures Féminines Schématiques;Delluc and Delluc 1995) which are typical in the western European Magdalenian. The same

Figure 5Macrophotographic details of figure B1 (D. Garate).



conventions can be observed at various sites including those of Lalinde (France) andGönnersdorf (Germany) (Bourrillon 2009; Bourrillon et al. 2012; Bosinski et al. 2001; Dellucand Delluc 1995) (Fig. 6). The representation of parallel arms, with no clear perspective, and theschematic head are less common, though present, in these kinds of engravings.

The other figures, especially A1 and B3, can be assimilated within the same ideographicconcept, but they are of a lower definition and have a lesser degree of detail, which hinders anoutright interpretation. However, the association in the same block of a clear FFS and of lessclear fusiform figures, close to the FFS concept, suggests that we can probably interpret theengravings on the Arlanpe block as a composition of anthropomorphic depictions, some of whichare feminine.

ARCHAEOLOGICAL CONTEXT

Stratigraphy of the entrance sector

Arlanpe Cave has been excavated in three different sectors (Fig. 2). The entrance sector,where the block was found, is near the actual entrance of the cave and presents a complexstratigraphical sequence (Fig. 7). It is characterized by a sedimentary hiatus, erosion, animalburrowing and human activity. Seven main episodes can be distinguished:

Figure 6Examples of Schematic Feminine Figures (FFS). a. Les Combarelles (C. Archambeau); b. Lalinde (A. Roussot); c.Magdeleine-la-Plaine (E. Ladier); d. Fronsac (B. and G. Delluc); e. and f. Gönnersdorf (G. Bosinski); g. Andernach

(G. Bosinski); h. Hohlenstein-Bärenhole (G. Bosinski).



1. Cave infilling with yellow-orange sandy endokarstic sediments, totally sterile at the base(Level VIII) but with some evidence of human activity above it (Level VII). The upper partof this level is partially encrusted with carbonate calcium. This sequence shows traces oferosion caused by water.

2. Greenish-brown sediment in which charcoal and land snails are present. It represents the firstintense episode of human occupation of the cave, with many lithic and faunal remains (LevelVI). This level can be attributed to the Early Middle Palaeolithic (EMP).

3. Yellow-orange sandy sediments. Two intact levels have been identified. The lowest one(Level V) is almost sterile archaeologically and in its upper part stalagmite crust fragmentsare very abundant. The upper level (Level IV) has yielded more archaeological material andcan also be attributed to the EMP. Levels IV–V have been used by burrowing animals(probably badgers) to make tunnels, owing to their more compacted nature. These tunnelshave collapsed and were finally filled by an admixture of materials from the upper levels(Level II – Solutrean) together with original material from these levels. The carefulexcavation process has allowed the identification and isolation of these tunnels from the restof the archaeological material found in situ.

4. Dark grey sediments with many limestone slab fragments. This level shows strong evidenceof anthropogenic activities such as fireplaces, consumed faunal remains and lithic artefacts(Rios-Garaizar et al. 2008; 2013). It has been dated to the Late Upper Solutrean (US),c.17,000 BP. The level is partially affected by underlying burrowing, deformation caused bycollapses, and by the excavation of ceremonial pits during Late Roman occupations. Twolevels have been identified in this episode; Level II has been assigned to the intact sediments,while Level III is composed of sediments displaced by burrowing and burrow collapse.

Figure 7Western section of the entrance sector (J. Rios-Garaizar).



5. The sediments situated above Level II are pale brown in colour and contain fewer limestonefragments (Level I). The horizontal extension of this level has been heavily affected by theexcavation of a ceremonial pit (see below) and probably by burrowing. Nevertheless, intactpatches of unmodified sediment were excavated. Archaeological remains are less abundantthan in the previous level. The dating and archaeological composition, which will bediscussed later, point to a Middle Magdalenian (MM) attribution.

6. The exposed Palaeolithic surface was probably used during the Late Bronze Age to deposithuman bodies as part of a funerary ritual. In the central area of the cave, there is evidenceof a small mound that could have a funerary significance. The cave was then used during theLate Roman period for ritual activity. These rituals comprised the excavation of smallceremonial pits (Fosa N and Fosa S), where different kinds of goods were offered (GutiérrezCuenca et al. 2012).

7. The uppermost level consists of mixed sediments belonging to cave entrance dismantlement,burrowing activity and illegal excavations.

Context of the engraved block

The block was first discovered during the 2008 season, in the eastern section of theentrance sector (within subsquares K21.1–2 and K21.3–4). The exposed surface was then almostcompletely covered by mixed sediments corresponding to a filled in ceremonial pit dating backto the Late Roman period (Fig. 8: A). Only the southernmost part of the block was covered byintact sediments corresponding to Level I. At this time the engraved surfaces were not exposed,so the importance of this limestone block was not recognized and it was thus left in place. In the2010 season, the sequence on the eastern profile was cleared and defined. Owing to the proximityof the cave wall it was quite difficult to identify the different levels but we were able to recognizeLevel II sediments (US) and Level I-like sediments above them (Figure 8: B–D). In the photos,the relative position of the block can be easily appreciated, as well as the partial covering ofLevel I sediments (Figure 8: B). Between 2010 and 2011, the eastern section partially collapsedso we decided to open two more subsquares (K21.2 and K21.4), excavating the small volume ofsediment situated between the 2008 excavation and the cave wall. This excavation allowed boththe extraction and the study of the relative position of the block. It was leaning against the cavewall and was partially covered by Level I and mixed sediments. At the moment of extraction themain engraved figure (B1) was noticed and was immediately photographically recorded.

The block was resting, almost vertically, on the long and narrow side withoutengravings. The wide side with decoration (Side A) was leaning against the cave wall, with theintervening deposits consisting of mostly mixed sediments. The decorated long and narrow side(Side B) was positioned on the top, slightly inclined towards the south, with the end with thecomplete FFS (B1) positioned higher than the end with the schematic fusiform depiction (B3),which was partially covered by Level I sediments. The base of the block was embedded in LevelIV, Level II and burrow sediments.

Our interpretation is that the block was found in a modified in situ context. Probably itwas originally in a vertical position, but the excavation of the Roman pit and the deformationcaused by burrow collapses made it lean, also deforming the sediments covering the block. Thisfact casts some doubt on the exact stratigraphical position of the block and the engravings.Nevertheless, given the way it was found, and the stratigraphy of this excavation sector, the most



probable attribution is to Level I. Following this hypothesis, the block was already in place whenMagdalenian people occupied the cave and made engravings on two of its exposed surfaces.

Level I radiocarbon dating

Two herbivore bone samples recovered from Level I were dated by 14C AMS. Theycome from the same subsquare and excavation pit, with a difference in depth of 3 cm, Beta-316472 being above Beta-287336. Calibrated results (Table 1) show no overlap at 2 sigmadeviation, suggesting a long time formation. Nevertheless, both dates can be attributed to thelocal Early Middle Magdalenian. Similar dates have been recently obtained in Santimamiñe(López Quintana and Guenaga Lizaso 2011), Linar (Rasines et al. 2009), Ekain (Altuna et al.2010) and Mirón Caves (Straus et al. 2011).

Archaeological assemblage

The Level I lithic assemblage is composed of 470 pieces. Of these, 246 pieces arefragments smaller than 10 mm, natural fragments and formless fragments. The remaining 224

Figure 8Position of the block in 2010 excavation. A) North section, the black arrow marks the position of the block; notethe Fosa Norte filling cutting the original stratigraphy preserved in the left corner. B) Eastern section, the white

arrow marks the position of a small limestone block which was situated in the upper surface of Level II, just in theinterface between Levels I and II. Note the Fosa Norte filling above the engraved block. C) Top of Level II before

excavation. D) Base of Level II (J. Rios-Garaizar).



comprise blades, flakes, cores and by-products (Table 2). The main raw material used is flint (88per cent) imported from different locations situated between 30 and 50 km away from the site(Flysch, Treviño and Loza varieties). Other local raw materials, such as sandstone or mudstone,were used for specific purposes such as hard hammers, or for obtaining large flakes.

The flint production at the site is characterized by blade technology. Four bladecategories have been identified with regard to blank width. The most abundant are bladelets ofaround 4 mm in width. These products correspond to the negatives of bladelet cores recovered atthe site. The proportion of bladelets to bladelet cores suggests that, although some productionoccurred at the site, the bulk of bladelets was produced elsewhere. This becomes more evidentwhen studying the larger blades, which were usually transformed into formal tools, suggestingthat they were introduced into the site as personal gear.

Formally retouched tools (Table 3) consist of backed bladelets, burins, borers (some ofwhich are microlithic), notches and raclettes. The backed bladelets show simple or doublebacking with steep retouch. Blanks selected to make armatures were mainly straight and narrowbladelets (3–4 mm). The presence of impact fractures on some of these pieces indicates that theywere used in composite hunting weapons. Burins are thick, usually displaying multiple burinfacets and crest-like preparations. Some of them probably acted as bladelet cores. Aside fromretouched tools, there are also some non-retouched macro-tools that were used in knapping andother activities, such as retouchers or ochre grinders.

The faunal assemblage from Level I is dominated by mountain goat (Capra pyrenaica)and chamois (Rupicapra pyrenaica), but other species such as red deer (Cervus elaphus), wolfand bear are also present (Table 4) (Arceredillo-Alonso et al. 2013).

TABLE 1

Radiocarbon dates from Level 1

Level Lab No. Material Uncal(BP)

δ13C (‰) δ15N (‰) CAL BP (95%,IntCal 09)

I Beta-287336 Herbivore bone withanthropic fracture

14,150±60 -20.3 3.4 17,540–16,930

I Beta-316472 Herbivore bone 15,100±60 -19.6 n.a. 18,589–18,025

TABLE 2

Technological classification of Level I

Artefact type Flint Mudstone Sandstone Others Σ

Flake core 2 2Bladelet core 2 2Cortical flakes and blades 14 6 1 21Blade production by-products 14 14Flakes 19 10 1 30Blades and bladelets 131 2 133Burin spalls 10 10Splintered piece fragments 5 1 6Debris 192 13 16 221Natural fragments 1 4 20 25Used non-retouched rocks 5 1 6Total 390 32 9 39 470



TABLE 3

Typological classification of Level 1*

S. Bordes type Sandstone Limonite Mudstone Silex Σ

23. Borer 2 224. ‘Bec’ 1 126. Micro-borer 3 328. Asymmetrical dihedral burin 1 129. Angle dihedral burin 1 130. Angle burin on broken blade 2 234. Burin on straight retouched truncation 1 158. Backed blade (total backing) 1 174. Notch 2 278. Raclette 2 278b. Raclette (atypical) 1 185. Backed bladelet 14 1487. Denticulate backed bladelet 1 189. Notched bladelet 4 490. Dufour bladelet 1 192. Diverse 1 192. Fragment of backed blade 1 192. Retouched tool 1 192. Flake with retouch in ventral surface 1 192. Fragment with abrupt retouch 2 2Macrolithic toolsShaped pebble 1 1Hard hammer 1 1Retoucher 1 1Used slab 4 4Total 6 1 2 41 50

* Following Sonneville-Bordes and Perot 1956.

TABLE 4

Number of remains (NR) and minimum number of individuals (MNI) of the faunal remains found in Level I ofArlanpe Cave

Taxa NR MNI

Artiodactyla indet. 33 –Cervidae indet. 3 –Cervus elaphus 2 2Caprinae indet. 1 –Rupicapra pyrenaica 9 2Capra pyrenaica 5 2Herbivores 53 6Carnivora indet. 7 –Canis lupus 1 1Felidae indet. 1 1Ursus sp. 1 1*Carnivores 10 2+1*TOTAL 63 8+1*

* Individuals based on decidual dentition



Arlanpe Level I can be interpreted as the result of various short occupations that mayhave been infrequent, based on the dates obtained from this level. During these occupations,activities related to weapon replacement were carried out. The strategic position of the site witha good view over the confluence of two rivers (Arratia and Ibaizabal), as well as its location atthe entrance of a calcareous gorge, probably favoured its use for controlling prey movements.Compared to the occupations in Level II (Final Upper Solutrean), where some signs of stabilitywere recognized (Rios-Garaizar et al. 2013), Level I seems basically to fulfil the function of ahunting camp on the basis of the lithic and faunal assemblages.

DISCUSSION

Chrono-cultural attribution

The suggested stratigraphical association between the block and Level I, although notdefinitive, is probably the most parsimonious explanation based on the available evidence. LevelI represents a succession of ephemeral hunting camps, where limited activities were carried out.The dates of this level situate it in an early phase of the Middle Magdalenian. The lithicassemblage recovered in Level I is not very diagnostic, but the presence of abundant backedbladelets, the large proportion of burins compared with endscrapers, and the characteristics ofblade production are quite similar to other regional Middle Magdalenian sites such as Ekain VII(Cazals and Langlais 2005), Santimamiñe levels Csn-Camr (López Quintana and GuenagaLizaso 2011) and Mirón Cave (Straus and González Morales 2012).

The chronology of such kinds of feminine representations in Europe extends fromc.13,680 ± 130 BP (Magdeleine-la-Plaine, France) to 11,700 ± 90 BP (Petersfels, Germany)(Bourrillon et al. 2012; Bosinski et al. 2001; Ladier et al. 2005), that is, between the end of theMiddle Magdalenian and the Late Magdalenian. Unfortunately, most of these representationslack precise dating and their chronology has been estimated by stylistic comparison. Some ofthem, from southern France, can also be attributed to the Middle Magdalenian, as is the case forArlanpe Level I. This is also true of those from Villars (Dordogne), Gourdan and Mas-d’Azil(Pyrenees). The engraved FFS from Gourdan has been attributed to the Middle or UpperMagdalenian by stylistic comparison (Fritz et al. 1993), and the associated archaeologicalcontext has yielded two dates between 14,400 and 13,200 BP (Bertrand et al. 2003; Jaubert1995), that is, within the temporal range of the Middle Magdalenian. At Mas-d’Azil, FFS havebeen recovered from a Middle Magdalenian layer without direct dating, but archaeological layerswith the same cultural attribution excavated in the Salle Piette and in the Galerie des silex yielded14C dates between 13,640 and 13,200 BP (Alteirac and Bahn 1982).

Contextual interpretation

The presence of this kind of representation in a site interpreted as a hunting camp maybe difficult to explain. However, the low intensity of the use of this cave in the Magdalenianleaves plenty of room for multiple uses over a long period of time. The block was probablyoriginally upright, with the images on Side B facing the entrance of the cave, and Side Aorientated in the narrow space between the block and the cave wall. Despite the fact that no otherevidence of symbolic activity has been identified at Arlanpe in the Magdalenian (possibly owingto the limited extent of the excavation), this probably belongs to some sort of ritual structure. The



presence of ritual structures in the Lower and Middle Magdalenian has recently been describedin detail (Arias 2009). It seems that sometimes these occur in the deeper areas of caves wheresymbolic activities were carried out, for example La Garma A, Praile Aitz, Erberua and therecently described Etxeberri (Garate et al. 2012); while others took place closer to occupationareas, for example Erralla, El Juyo and the secondary burial at El Mirón (Straus et al. 2011).Some of these sites include upright stones (Erralla, Juyo, La Garma A) and on many of themthere were anthropomorphous representations (Entrefoces, La Garma A, Juyo). In such levels,pigments and ornaments were also commonly present. In the case of Arlanpe Level I, ochregrinders and ochre fragments have been recovered close to the engraved block, but no ornamentsor portable art are known. What seems different in the case of Arlanpe is the presence of objects,such as burins or backed bladelets, associated with everyday activities. Furthermore, regardingthe faunal assemblage, there is nothing unusual or significant. The animals show a predominanceof mountain species as would be expected from the environment surrounding the cave (cf. Erralla– Altuna and Mariezkurrena 1985).

Regional stylistic parallels

The only two known parallels for Arlanpe in the northern Iberian Peninsula are thefigures from El Linar and Arenaza Caves (San Miguel and Muñoz 2002; Garate 2004). In thesesites, unique schematic feminine representations (FFS) have been recognized. They show afusiform torso, exaggeration of the buttocks, and the waist indicated by a V-like line, with noindication of the extremities or the head. Unlike Arlanpe, both figures are faint engravings oncave walls and are located far away from occupation areas (100 m in Linar, 250 m in Arenaza).None of these figures has been dated directly. Initial work in Linar Cave suggested that thehuman occupations were of Upper Magdalenian age (Moure 1971), but recent excavations inthree different sectors, situated at c.20 m from the FFS engraving (Lasheras et al. 2005), suggestthat the site was occupied during the Middle Magdalenian, with dates between c.15,800 and14,000 BP (Rasines et al. 2009), and is also associated with portable art typical of the MiddleMagdalenian (de las Heras et al. 2008). In the case of Arenaza, the excavations that were carriedout in the 1970s and 1980s revealed a Late Magdalenian–Azilian sequence (Apellaniz 1985), buta recent revision of the lithic industry reveals that level VI has elements that link it to olderphases of the Upper Magdalenian or even the Middle Magdalenian (Rios-Garaizar and SanEmeterio 2012). In the cave of Ardales, located in the south of the Iberian Peninsula, severalpossible FFS have also been described (Ramos et al. 2002; Maura et al. 2009), but so far noconcrete Upper Palaeolithic occupation context has been reported (Cantalejo and Espejo 2013).

Geographical dispersion of FFS among Europe

The distribution of sites with FFS representations extends from the northern IberianPeninsula to the central European plains. Three main concentration areas can be distinguished,one in the Dordogne/Quercy/Pyrenean regions (France), and the other two in western and easternGermany (Fig. 9). As mentioned previously, the older manifestations are found in the southernFrench area, probably beginning during the Middle Magdalenian, while in northern Europe thesefigures are associated with the Late Magdalenian occupations. In this sense, the attribution of theArlanpe block to the Middle Magdalenian reinforces the idea of an origin for these kinds ofimages in south-western Europe and an ultimate expansion towards northern Europe at the end



Figu

re9

Site

sin

Eur

ope

with

FFS.

The

doub

tful

figur

es(3

8,43

,45,

46)

are

ingr

ey.T

heci

rcle

sre

pres

ent

the

mai

nre

gion

alag

greg

atio

ns.(

Bas

emap

:E

urop

ean

Env

iron

men

tAge

ncy,

mod

ified

byR

.Bou

rrill

on.)



of the Magdalenian. This expansion would coincide roughly with the process of colonization ofthe central European plains at the end of the Last Glacial (Otte 1990; Jochim et al. 1999;Terberger and Street 2002; Demars 2006), which probably was made from French–Iberianrefugia. It has also been proposed, from genetic evidence, that Cantabrian populations played amajor role in the colonization of northern Europe (Torroni et al. 2001), but recent studies suggesta more cautious interpretation of this genetic evidence (Garcia et al. 2011).

Thus, the apparent scarcity of FFS in the northern Iberian Peninsula is unexpected. Thisfact has been used to propose the existence of a stylistic frontier during the Upper Magdalenianbetween the Cantabrian region and south-western France (Fritz et al. 2007). The available datasuggest that FFS originated during the Middle Magdalenian, and the discovery of the Arlanpeblock, where three FFS (two of them less perceptible) were depicted probably in the MiddleMagdalenian, reinforces the possibility that the symbol originated in the Franco-Cantabrianregion during this period, and then dispersed from this macro-region towards north-centralEurope during the Late Magdalenian. Many other lines of evidence speak about a certain unityof this macro-region during the Middle Magdalenian, such as the repetition of artisticrepresentations (ie. Contour-découpé, Claviforms, Niaux-style bisons, etc. (Tosello 2003; Garateet al. 2012), of ornament types (Rivero and Álvarez-Fernández 2009), the distribution pattern ofraw materials (Corchón et al. 2009), and the expansion of certain types of bone tools (i.e.fork-based antler points – Pétillon 2007).

CONCLUSIONS

The finding of the Arlanpe block, with engraved figurines, reinforces the presence of FFS in thenorthern Iberian Peninsula, although some doubtful examples were previously recognized, in ElLinar and Arenaza Caves. One of the Arlanpe figures clearly fits the stylistic canon of Lalinde-Gönnersdorf representations, while another two are closely related to this canon and can thus beinterpreted as probable FFS, with a further two being more difficult to interpret. In Arlanpe, thepresence of this engraved block in a site interpreted as a hunting camp is difficult to explain, butis probably linked to the creation of some sort of ritualized space, as has been proposed for otherexamples from the Cantabrian region.

The block at Arlanpe also fills a gap in the distribution of FFS representations insouth-western Europe, which was difficult to explain given the apparent unity, at least insymbolic-artistic manifestations, of the Franco-Cantabrian region during the Middle and UpperMagdalenian. The possible relationship of the Arlanpe block to the Middle Magdalenian levelsof the site reinforces the idea that such images were initially developed in south-western Europeand later diffused to the north, following the colonization routes of the northern plains, alreadyfree of ice. This also opens new questions about the role of populations from the Cantabrianregion in this recolonization process.

Finally, the presence of a clear Lalinde-Gönnersdorf type figure from an Early MiddleMagdalenian context suggests that these kinds of manifestations, typical of the UpperMagdalenian, probably had an older origin than previously thought.

Acknowledgements

We wish to express our gratitude to the anonymous reviewers who helped substantially toimprove the manuscript. The English version was corrected by Charles Bashore. We are grateful to the



Arlanpe excavation and research team for their efforts. Fieldwork was funded by Bizkaiko Foru Aldundia,Harpea Kultur Elkartea and the Barandiarán Foundation, and help provided by Lemoako Udala and theInstituto Internacional de Investigaciones Prehistóricas de Cantabria – IIIPC. The cataloguing of thearchaeological material was funded by Eusko Jaurlaritza. The faunal analysis was funded by grants fromEusko Ikaskuntza. AG-O had a Marie Curie IEF Fellowship when this work was done and is supported bythe Ministerio de Economia y Competitividad (project CGL2012-38434-C03-01) and Gobierno Vasco/Eusko Jaurlaritza (Research Group IT 834-13).

(JR-G, TK) Centro Nacional de Investigación sobre la Evolución Humana (CENIEH)09002 Burgos

SPAINE-mail: [email protected]

(DG, RB) CREAP Cartailhac-TRACES-UMR 5806Université de Toulouse-Le Mirail

31058 Toulouse CedexFrance

(AG-O) Ikerbasque/Dept. Estratigrafía y PaleontologíaFacultad de Ciencia y Tecnología

Euskal Herriko UnibertsitateaUPV-EHU. Apdo. 644

48080 BilbaoSPAIN

REFERENCES

ALTEIRAC, A. and BAHN, P. 1982: Premières datations radiocarbone du Magdalénien moyen de la grotte duMas d’Azil (Ariège). Bulletin de la Société préhistorique ariégeoise 37, 107–10.

ALTUNA, J. and MARIEZKURRENA, K. 1985: Bases de subsistencia de los pobladores de Erralla:macromamíferos. In ALTUNA, J. and MARIEZKURRENA, K. (eds.), Cazadores magdaleniense en lacueva de Erralla (Cestona, País Vasco) (Donostia-San Sebastián, Sociedad de Ciencias Aranzadi –Aranzadi Zientzia Elkartea), 87–117.

ALTUNA, J., MARIEZKURRENA, K., RÍOS, F. and WESBUER, J. 2010: Contorno recortado de ave en el yacimientode Ekain (Deba, País Vasco). In Pleistocene Art of the World, IFRAO Congress, Ariège-Pyénées,France, CD 9.

APELLANIZ, J.M. 1985: Cueva de Arenaza I (San Pedro de Galdames, Vizcaya). Arkeoikuska: Investigaciónarqueológica 1984, 60–2.

ARCEREDILLO-ALONSO, D., GÓMEZ-OLIVENCIA, A. and SAN PEDRO-CALLEJA, Z. 2013: La fauna demacromamíferos de los niveles pleistocenos de la Cueva de Arlanpe. In RIOS-GARAIZAR, J., GARATE, D.

and GÓMEZ-OLIVENCIA, A. (eds.), La cueva de Arlanpe (Lemoa): Ocupaciones humanas desde elPaleolítico Medio Antiguo hasta la Prehistoria Reciente (Bilbao, Diputación Foral de Bizkaia, KobieSerie BAI 3), 123–60.

ARIAS, P. 2009: Rites in the dark? An evaluation of the current evidence for ritual areas at Magdalenian cavesites. World Archaeology 41, 262–94.

ARRIZABALAGA, A. 2007: Frontières naturelles, administratives et épistémologiques. L’unité d’analysedans l’archéologie du Paléolithique (dans le cas basque). In CAZALS, N., GONZÁLEZ URQUIJO, J.E. andTERRADAS, X. (eds.), Frontières naturelles et frontières culturelles dans les Pyrénées préhistoriques.Fronteras naturales y fronteras culturales en los Pirineos prehistóricos (Santander, PUbliCan –Ediciones de la Universidad de Cantabria), 27–37.

BERTRAND, A., DUJARDIN, V. and PINÇON, G. 2003: Les répartitions d’éléments clés de l’industrie en matièredure animale au cours du Magdalénien Moyen en Europe et leur signification. In DESBROSSE, R. and



mailto:[email protected]

THÉVENIN, A. (eds.), Préhistoire de l’Europe. Des origines à l’Age du Bronze. Actes des congrèsnationaux des sociétés historiques et scientifiques, 125ème, Lille, 2000 (Paris, CTHS), 247–69.

BOSINSKI, G., D’ERRICO, F. and SCHILLER, P. 2001: Die Gravierten frauendarstellungen von Gönnersdorf(Stuttgart).

BOURRILLON, R. 2009: Les représentations humaines sexuées dans l’art du Paléolithique supérieureuropéen: diversité, réminiscences et permanences (Mémoire de thèse, Université de Toulouse II leMirail).

BOURRILLON, R., FRITZ, C. and SAUVET, G. 2012: The theme of female representations during the EuropeanUpper Palaeolithic: formal persistence and variations. Bulletin de la Société préhistorique française109, 85–103.

CANTALEJO, P. and ESPEJO, M.M. 2013: Cueva de Ardales (Málaga, España). Patrimonio prehistórico en elsur de la Península Ibérica. In PASTOORS, A. and AUFFERMANN, B. (eds.), Pleistocene Foragers on theIberian Peninsula: Their Culture and Environment. Festschrift in Honour of Gerd-Christian Wenigerfor his Sixtieth Birthday (Mettmann, Neanderthal Museum, Wissenschaftliche Schriften desNeanderthal Museums 7), 243–60.

CAZALS, N. and LANGLAIS, M. 2005: La place d’Ekain (couche VII) au sein du Magdalénien basco-cantabrique: nouvelles contributions sur l’organisation des productions lithiques. Munibe(Antropologia-Arkeologia) 57, 177–91.

CONARD, N. 2009: A female figurine from the basal Aurignacian of Hohle Fels Cave in southwesternGermany. Nature 454, 248–52.

CORCHÓN, M.S., TARRIÑO, A. and MARTÍNEZ, J. 2009: Mobilité, territoires et relations culturelles au début duMagdalénien moyen cantabrique: nouvelles perspectives. In DJINDJIAN, F., KOZLOWSKI, J. and BICHO,N.F. (eds.), Le Concept de Territoires dans Le Paléolithique Supérieur Européen. Actes Du XV Congrèsde l’UISPP (Lisbonne 2006) (Oxford, BAR Int. Ser. 1938), 217–30.

DE LAS HERAS, C., MONTES, R., LASHERAS, J.A., RASINES, P. and FATÁS, P. 2008: Dos rodetes paleolíticosprocedentes de las cuevas del Linar y de Las Aguas, Alfoz de Loredo (Cantabria). Veleia 24–5,161–74.

DELLUC, B. and DELLUC, G. 1995: Les figures féminines schématiques du Périgord. L’Anthropologie 99,236–57.

DEMARS, P.-Y. 2006: L’occupation de l’Europe par les chasseurs du Paléolithique supérieur: une question declimat. M@ppemonde 83.

DEMARS, P.-Y. 2008: Paléogéographie des chasseurs de l’Europe du Paléolithique supérieur: répartition etspécialisation des sites. L’Anthropologie 112, 157–67.

FRITZ, C. and TOSELLO, G. 2007: The hidden meaning of forms: methods of recording Paleolithic parietalart. Journal of Archaeological Method and Theory 14, 48–80.

FRITZ, C., TOSELLO, G. and PINÇON, G. 1993: Les gravures pariétales de la grotte de Gourdan. In DELPORTE,H. and CLOTTES, J. (eds.), Préhistoire, Arts et Sociétés, Actes du 118e Congrès National des SociétésHistoriques et Scientifiques de 1993 à Pau (Paris, CTHS), 381–402.

FRITZ, C., TOSELLO, G. and SAUVET, G. 2007: Groupes ethniques, territoires, échanges; la ‘notion defrontière’ dans l’art magdalénien. In CAZALS, N., GONZÁLEZ URQUIJO, J.E. and TERRADAS, X. (eds.),Frontières naturelles et frontières culturelles dans les Pyrénées préhistoriques. Fronteras naturales yfronteras culturales en los Pirineos prehistóricos (Santander, PUbliCan – Ediciones de la Universidadde Cantabria), 165–81.

GARATE, D. 2004: Nuevas investigaciones sobre el arte paleolítico de la cueva de Arenaza (Galdames,Bizkaia). Munibe (Antropologia-Arkeologia) 56, 3–17.

GARATE, D., BOURRILLON, R. and RIOS-GARAIZAR, J. 2012: La grotte ornée paléolithique d’Etxeberri(Camou-Cihige, Pyrénées-Atlantiques): datation du contexte archéologique de la «salle desPeintures». Bulletin de la Société préhistorique française 109, 637–50.

GARCIA, O., FREGEL, R., LARRUGA, J.M., ALVAREZ, V., YURREBASO, I., CABRERA, V.M. and GONZALEZ, A.M.

2011: Using mitochondrial DNA to test the hypothesis of a European post-glacial humanrecolonization from the Franco-Cantabrian refuge. Heredity 106, 37–45.

GAUDZINSKI, S. and JÖRIS, O. 2015: Contextualising the female image – symbols for common ideasand communal identity in Upper Palaeolithic societies. In WENBAN-SMITH, F., COWARD, F., HOSFIELD,R. and POPE, M. (eds.), Settlement, Society, and Cognition in Human Evolution (Cambridge), 288–314.



GUTIÉRREZ CUENCA, E., HIERRO GÁRATE, J.A., RIOS-GARAIZAR, J., GÁRATE, D., GÓMEZ-OLIVENCIA, A. andARCEREDILLO-ALONSO, D. 2012: El uso de la cueva de Arlanpe (Bizkaia) en época tardorromana.Archivo Español de Arqueología 85, 229–51.

JAUBERT, J. 1995: Datations numériques des gisements des Pyrénées centrales: Ariège, Haute-Garonne(zone pyrénéenne) et Hautes-Pyrénées. Bulletin de la Société Préhistorique de l’Ariège 50, 291–301.

JOCHIM, M., HERHAHN, C. and STARR, H. 1999: The Magdalenian colonization of southern Germany.American Anthropologist 101, 129–42.

LADIER, E., LENOIR, M. and WELTÉ, A.-C. 2005: Relations ou convergences entre Périgord et Quercy: le casdes figures féminines schématiques de type Lalinde-Gönnersdorf dans l’art mobilier. In JAUBERT, J.

and BARBAZA, M. (eds.), Territoires, déplacements, mobilité, échanges durant la Préhistoire,Actes du 126e Congrès National des Sociétés historiques et scientifiques, Toulouse (Paris, CTHS),397–410.

LASHERAS, J.A., MONTES, R., MUÑOZ, E., RASINES, P., DE LAS HERAS, C. and FATÁS, P. 2005: El proyectocientífico Los Tiempos de Altamira: primeros resultados. Munibe (Antropologia-Arkeologia) 57,143–59.

LÓPEZ QUINTANA, J.C. and GUENAGA LIZASO, A. 2011: Revisión estratigráfica del depósito arqueológico dela cueva de Santimamiñe (Kortezubi, Bizkaia): Campañas de 2004 a 2006. Cronoestratigrafía ypaleoambiente. In LÓPEZ QUINTANA, J.C. (ed.), La Cueva de Santimamiñe: Revisión y Actualización(2004–2006) (Bilbao, Diputación Foral de Bizkaia, Kobie Serie BAI 1), 7–70.

LÓPEZ QUINTANA, J.C., GUENAGA LIZASO, A. and SÁENZ DE BURUAGA, A. 2011: Dinámica evolutiva de laindustria lítica tallada en la secuencia estratigráfica de Santimamiñe: campañas de 2004 a 2006. InLÓPEZ QUINTANA, J.C. (ed.), La Cueva de Santimamiñe: Revisión y Actualización (2004–2006) (Bilbao,Diputación Foral de Bizkaia, Kobie Serie BAI 1), 71–110.

MAURA, R., VERA, J.C., CANTALEJO, P., MORENO, A. and ARANDA, C. 2009: La figura humana femenina en elarte parietal paleolítico del sur peninsular: a propósito de las ‘Venus Egabrenses’. Espacio Tiempo yForma Serie I, Nueva época Prehistoria y Arqueología 2, 93–102.

MOURE, J.A. 1971: El magdaleniense superior de la cueva del Linar (La Busta, Santander). Empúries:revista de món clàssic i antiguitat tardana 33–4, 283–4.

OTTE, M. 1990: Relations transculturelles et transrégionales dans l’art mobilier. In CLOTTES, J. (ed.), L’artdes objets au Paléolithique: les voies de la recherche, Colloque Foix-le-Mas-d’Azil, 1987 (Paris,Ministère de la Culture), 185–94.

PÉTILLON, J.-M. 2007: Les pointes à base fourchue de la zone Pyrénéo-Cantabrique: un objet à la charnièreentre Magdalenien Moyen et Magdalenien Supérieur? In CAZALS, N., GONZÁLEZ URQUIJO, J.E. andTERRADAS, X. (eds.), Frontières naturelles et frontières culturelles dans les Pyrénées préhistoriques.Fronteras naturales y fronteras culturales en los Pirineos prehistóricos (Santander, PUbliCan –Ediciones de la Universidad de Cantabria), 245–78.

RAMOS, J., CANTALEJO, P., MAURA, R., ESPEJO, M.M. and MEDIANERO, F.J. 2002: La imagen de la mujer en lasmanifestaciones artísticas de la Cueva de Ardales (Ardales, Málaga). Un enfoque desde la relacióndialéctica producción reproducción social. RAMPAS V, 87–124.

RASINES, P., MONTES, R., LASHERAS, J.A., MUÑOZ, E., DE LAS HERAS, C. and FATÁS, P. 2009: ‘Los tiempos deAltamira’. Un proyecto de investigación de la Cueva de Altamira y su entorno Paleolítico. In LLANOS,A. (ed.), Actas del Congreso ‘Medio siglo de Arqueología en el cantábrico oriental y su entorno’(Vitoria-Gasteiz, Instituto Alavés de Arqueología), 709–27.

RIOS-GARAIZAR, J. and SAN EMETERIO, A. 2012: Avance del estudio de la industria lítica del yacimiento deArenaza I (Galdames, Bizkaia) (Bilbao, Repport Arkeologi Museoa).

RIOS-GARAIZAR, J., IRIARTE-AVILÉS, E., GARATE, D., GOMEZ-OLIVENCIA, A. and SAN PEDRO CALLEJA, Z. 2008:Nuevos datos sobre la transición entre el Solutrense Superior y el Magdaleniense inferior en la Regióncantábrica: la cueva de Arlanpe (Lemoa, Bizkaia). Sautuola 14, 95–104.

RIOS-GARAIZAR, J., GARATE, D., GÓMEZ-OLIVENCIA, A., IRIARTE-AVILÉS, E., ARANBURU-ARTANO, A.,ARCEREDILLO-ALONSO, D., GARCÍA, A., IRIARTE-CHIAPUSSO, M.J., MORENO, J., MURELAGA, X., ORTÍZ,J.E., TORRES, T., SAN PEDRO-CALLEJA, Z. and ZAPATA-PEÑA, L. 2011: The Lower to Middle Palaeolithictransition in northern Iberia: new data from Arlanpe Cave. Antiquity 85, 329.

RIOS-GARAIZAR, J., GARATE, D., GÓMEZ-OLIVENCIA, A., ARCEREDILLO-ALONSO, D., IRIARTE-AVILÉS, E.,GARCÍA-MORENO, A. and SAN PEDRO-CALLEJA, Z. 2013: El final del Solutrense en el oriente cantábricoa través de las ocupaciones de la cueva de Arlanpe (Lemoa, Bizkaia). Zephyrus 72, 15–38.



RIVERO, O. and ÁLVAREZ-FERNÁNDEZ, E. 2009: Evidencias de contactos intergrupales en Europa: elementosde adorno y arte mobiliar en el Magdaleniense Medio. In FRITZ, C., BOURRILLON, R., PETROGNANI, S.,GARATE, D. and SAUVET, G. (eds.), L’Art des Sociétés Préhistoriques. Rencontres Internationales.Doctorants et Post-doctorants. 1er édition, Avril 2008, Toulouse (Varilhes, Préhistoire, Art et Sociétés64), 149–56.

SAN MIGUEL, C. and MUÑOZ, E. 2002: El Linar. In ASOCIACIÓN CÁNTABRA PARA LA DEFENSA DEL PATRIMONIO

SUBTERRÁNEO (ed.), Las cuevas con arte paleolítico en Cantabria (Santander), 83–8.SONNEVILLE-BORDES, D. and PEROT, J. 1956: Lexique typologique du Paléolithique Supérieur. Bulletin de

la Société préhistorique française 53, 547–59.STRAUS, L.G. and GONZÁLEZ MORALES, M.R. 2012: The Magdalenian settlement of the Cantabrian region

(northern Spain): the view from El Mirón Cave. Quaternary International 272–3, 111–24.STRAUS, L.G., GONZÁLEZ MORALES, M.R. and CARRETERO, J.M. 2011: Lower Magdalenian secondary human

burial in El Mirón Cave, Cantabria, Spain. Antiquity 85, 1151–64.TERBERGER, T. and STREET, M. 2002: Hiatus or continuity? New results for the question of pleniglacial

settlement in Central Europe. Antiquity 76, 691–8.TORRONI, A., BANDELT, H.-J., MACAULAY, V., RICHARDS, M., CRUCIANI, F., RENGO, C., MARTINEZ-CABRERA, V.,

VILLEMS, R., KIVISILD, T., METSPALU, E., PARIK, J., TOLK, H.-V., TAMBETS, K., FORSTER, P., KARGER, B.,FRANCALACCI, P., RUDAN, P., JANICIJEVIC, B., RICKARDS, O., SAVONTAUS, M.-L., HUOPONEN, K.,LAITINEN, V., KOIVUMÄKI, S., SYKES, B., HICKEY, E., NOVELLETTO, A., MORAL, P., SELLITTO, D., COPPA,A., AL-ZAHERI, N., SANTACHIARA-BENERECETTI, A.S., SEMINO, O. and SCOZZARI, R. 2001: A signal, fromhuman mtDNA, of postglacial recolonization in Europe. American Journal of Human Genetics 69,844–52.

TOSELLO, G. 2003: Pierres gravées du Périgord Magdalénien (Paris, XXXVI supplément à GalliaPréhistoire).

WHITE, R., MENSAN, R., BOURRILLON, R., CRETIN, C., HIGHAM, T.F.G., CLARK, A.E., SISK, M.L., TARTAR, E.,GARDÈRE, P., GOLDBERG, P., PELEGRIN, J., VALLADAS, H., TISNÉRAT-LABORDE, N., DE SANOIT, J.,CHAMBELLAN, D. and CHIOTTI, L. 2012: Context and dating of Aurignacian vulvar representations fromAbri Castanet, France. Proceedings of the National Academy of Sciences 109, 8450–5.



the venuses block from arlanpe cave nort

Documents