The Gravettian of El Castillo revisited (Cantabria, Spain)

Quaternary International 359-360 (2015) 462e478

Contents lists available at ScienceDirect

Quaternary International journal homepage: www.elsevier.com/locate/quaint

The Gravettian of El Castillo revisited (Cantabria, Spain)
s a, Jose
-Manuel Maíllo-Ferna
ndez b, *, Pedro Castan ~ os c, Federico Bernaldo de Quiro a Ana Neira

n, Campus de Vegazana s/n, 24071 Leo
n, Spain Area de Prehistoria, Universidad de Leo
n a Distancia (UNED), Paseo Senda del Rey, 7, 28040 Madrid, Spain Dpto. Prehistoria y Arqueología, Universidad Nacional de Educacio c Sociedad de Ciencias Aranzadi, Geo-Q, Leioa, Bizkaia, Spain a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 2 September 2014

The long sequence of El Castillo cave contains Units 12 and 14, attributed to the Gravettian. This paper presents a revision of the lithic industry and the fauna recovered during H. Obermaier's 1910e1914 excavations of the site, as well as a number of new datings that enable us to chronologically place the above Gravettian occupations as one of the oldest in Europe. Unit 14 is dated between 34 and 33 ka cal BP and Unit 12 between 30 and 28 ka cal BP. The cave's oldest Gravettian level, Unit 14, presents techno-typological features typical of the first phases of the Gravettian in the Cantabro-Pyrenean region, such as Noailles burins, although it also shows some common elements with the Evolved Aurignacian. The youngest Gravettian unit, Unit 12, is characterised by laminar production from bipolar prismatic cores and a greater, albeit still discreet, presence of dorsal pieces. In terms of the fauna, Unit 14 is represented by red deer, chamois and horse, whereas Unit 12 is represented by red deer and horse, a hint as to what would later become the characteristic composition of Late Upper Palaeolithic faunal assemblages. The lithic and chronological characteristics of the Gravettian at El Castillo and the Cantabro-Pyrenean region lead us to believe in a mosaic formation of this techno-complex. © 2014 Elsevier Ltd and INQUA. All rights reserved.

Keywords: Gravettian El Castillo Fauna Noailles Lithic technology

1. Introduction Despite not being as ‘high-profile’ as other Palaeolithic periods, as is the case of the Middle-to-Upper Palaeolithic Transition and the replacement of Neanderthals by modern humans, the Gravettian has been the object of a relatively discreet, yet intense debate. Over the last century its own existence was debated in the so-called ‘Aurignacian battle’ (Breuil, 1912). Later, the discussion centred around whether or not it was coetaneous with the Aurignacian (Peyrony, 1933, 1936, 1946), its identification as an independent techno-complex (Garrod, 1938), and, even today, the debate continues to be centred around its regional identification and origin(s) ~ a, 2012). The Gravettian, whose geographical expansion ranges (Pen from Europe's Atlantic coasts to Siberia, presents cultural homo~ a, 2012): Homo geneity based on three of its characteristics (Pen sapiens as the author of these industries, as evidenced by the numerous burials attributable to this techno-complex in the

* Corresponding author.
s), jlmaillo@geo. E-mail addresses: [email protected] (F. Bernaldo de Quiro ~ os), ana. uned.es (J.-M. Maíllo-Fern
andez), [email protected] (P. Castan [email protected] (A. Neira). http://dx.doi.org/10.1016/j.quaint.2014.07.060 1040-6182/© 2014 Elsevier Ltd and INQUA. All rights reserved.

central-eastern part of the continent; the generalisation of ‘artistic’ manifestations, of which the so-called ‘Venus’ are worth noting; and the characterisation of the lithic industry based on abrupt retouch and dorsal pieces as its typical elements. The fact that the Gravettian is the first techno-complex clearly made by H. sapiens in Europe allows us to study the cultural processes of replacement, acculturation, and/or cultural replacement without a discussion on anthropological issues distorting such a process, as is the case when looking at earlier periods. The transition between the Aurignacian and the Gravettian or the Gravettian and Solutrean have not, however, generated the epistemological debate triggered by, for example, the Middle-to-Upper Palaeolithic or MesolithiceNeolithic Transitions. In this way, part of the current debate on the Gravettian centres, on the one hand, around the identification of common traits, and, on the other, on the marked regionalisation of its lithic industries. Despite giving the impression of being a monolithic and homogeneous techno-complex, the Gravettian has been, from a historiographic point of view, the techno-complex with the most subdivisions in the Palaeolithic of Western Europe. The common traits conceiving the Gravettian as a Pan-European techno-complex were proposed by Otte (1985), considering the Gravettian as a


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

homogeneous process emerging from a Central European migration, sensu Garrod. This hypothesis, which suggests a Central European origin to the Gravettian, and links European and Siberian industries, has recently been argued by other authors (Kozlowski, 2005; Simonet, 2009; Moreau, 2010) although some researchers consider these common elements to be very general, but none~ a, 2011a; pp. 20). On theless valid as part of the ‘bigger picture’ (Pen the other hand, some authors suggest, although still shyly, a mosaic conception of the Gravettian, where, from some common traits, and following a detailed study of the technology and typology, regional trends can be observed beyond the Gravettian facies (Klaric, 2006; ~ a, 2011a). Pen The cave of El Castillo represents one of the key sequences to understand the Palaeolithic of the Cantabrian region. This is especially true in the case for the Gravettian, a techno-complex better known in the eastern part of the region (Basque Country), which ~ a, appears to develop differently to that of the western part (Pen 2011a), and which shows with a clear polymorphism, as attested by a number of recently published sequences such as the Cuco ~ oz, 2013), or with few diagnostic rockshelter (Rasines and Mun
n (Heras et al., 2013, Gonza
lezfeatures such as Altamira o Miro Morales and Straus, 2013). In this study, we present the chronological framework of the two Gravettian units from the cave of El Castillo -12 and 14-; the preliminary study of the fauna from the collection excavated by Obermaier; and a revision of the lithic
gico Nacional (MAN) in collection housed in the Museo Arqueolo Madrid. 2. El Castillo cave El Castillo cave is located in Cantabria, in northern Spain, and was discovered by H. Alcalde del Río in November 1903 (Fig. 1). The site was excavated by the Institut de Pal
eontologie Humaine (I.P.H.) after an agreement was reached between Prince Albert I of Monaco and Alcalde del Río. The I.P.H.’s excavation was initially directed by H. Breuil, H. Obermaier and J. Bouyssonie between 1910 and 1914 although it was mostly undertaken by H. Obermaier and P. Wernert.

463

Visits to the site by famous researchers, such as P. Teilhard de Chardin, M. Burkkit, E. Hern
andez-Pacheco or the Count of Vega de Sella, contributed towards increasing the site's popularity (Cabrera
s, 1979, 1984). When the excavations were completed, the Valde archaeological material was deposited at the I.P.H. in Paris until its return to Spain in the 1970s, where it was later studied (Cabrera
s, 1979, 1984). Small portions of this collection, however, Valde were divided mostly between the Museo de Arqueología de Cantabria
gico Nacional (MAN) and a small part (MUPAC), the Museo Arqueolo went to Paris' I.P.H. However, since Obermaier's excavations, samples have been sent to numerous European and American museums. El Castillo's sequence is composed of stratigraphic units showing human occupation interspersed with sterile units. The stratigraphic sequence is approximately 21 m deep and, based on the study of the collections of the aforementioned excavations
s, 1984; Cabrera-Valde
s et al., 2001; Bernaldo de (Cabrera-Valde
s et Maíllo-Ferna
ndez, 2009; Bernaldo de Quiro
s et al., Quiro
ndez et al., 2011), it is composed of the 2010; Maíllo-Ferna following archaeological units (Fig. 1): Late Acheulean (Units 26, 25, 24); Mousterian (22 and 21); Transitional Aurignacian (Unit 18); Archaic Aurignacian (Unit 16); Gravettian (14 and 12); Solutrean (10); Lower Madgalenian (Unit 8); Upper Magdalenian (Units 7 and 6); Azilian (Unit 4); Bronze Age (Unit 2); and Middle Ages (Unit 1) (Fig. 2). This sequence was used by Breuil as one of the arguments to corroborate his hypothesis regarding the seriation of the Euro
lez Echegaray, 2013; pp. 26). pean Upper Palaeolithic (Gonza The first great revision of the site was undertaken by Cabera
s (1979, 1984), who updated the stratigraphy and carried out Valde the first modern study of the collections. From the wide series of units discovered, in the present study we only focus on those described as Gravettian, Units 12 and 14 (Fig. 2), according to the
s, and which correspond to revised stratigraphy by V. Cabrera-Valde those labelled by Obermaier as Aurignacian a and b (Units 12 and 14 respectively). The reason why these Gravettian levels were referred to Alpha and Beta Aurignacian is linked to the division of the Upper Palaeolithic devised by H. Breuil: Aurignacian, Solutrean and


n; 8. La Vin ~ a; 9. Fig. 1. Map showing the location of main Gravettian sites cited in the text. 1. El Castillo; 2. Cueva Morín; 3. Pendo; 4. La Garma; 5. El Cuco; 6. Altamira; 7. El Miro ~ a; 12. Santimamin ~ e, Antolin ~ ako Koba; 13. Bolinkoba; 14. Amalda, Irikaitz; 15. Lezetxiki; 16. Aitzbitarte III; 17. Ametzagina; 18. Alkerdi. Cueto de la Mina; 10. La Riera; 11. Sopen

464


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

Fig. 2. Stratigraphy of El Castillo cave.

Magdalenian. The Aurignacian was made up of the Old Aurignacian (Chatelperronian), Middle Aurignacian (Proper Aurignacian), and Recent Aurignacian, which would correspond to the current Gravettian (Breuil, 1911). Unit 12, 0.5 m in thickness, is silty and has medium-sized blocks towards its base. Unit 14 also has a silty composition and contains big blocks, the majority of which are flat. Its thickness is in excess of 1 m in many parts of its sequence. These two Gravettian units are separated by Unit 13, which is made up of silt and large blocks, is between 0.5 and 0.8 m in thickness and is archaeologically sterile
s, 1984; pp. 396). (Cabrera-Valde 3. Radiometric dates In 1913, during H. Obermaier's excavations at El Castillo, Nels C. Nelson was sent to Spain by the American Museum of Natural History (AMNH) of New York to collect a number of material samples to be used in a diorama of El Castillo being put together at the US museum. The samples, which were taken with H. Obermaier,

consisted mainly of various boxes containing material recovered from each of the archaeological units identified during the excavation then taking place, as well as rock, vegetation, etc. samples that could be of use when putting together the aforementioned diorama (White, 2006; Tejero et al., 2010). It is not necessary to highlight the importance of these samples given they represent the only material reference from these earliest excavations, classified according to the Units assigned by Obermaier himself. They are, therefore, of uttermost importance in the present revision. In March 2011, we took, with the AMNH's permission, a number of samples from these boxes. In order to be completely sure of the reliability of the Units' assignations, we took two bones from each of the Gravettian units. The samples were sent to Beta Analytic Inc. to be dated. Together with these samples, there is another one that had been sent earlier: we obtained this sample whilst cleaning a profile at the cave, which we considered to be Unit 12, during development works at El Castillo in 2002. The data obtained allow us to note that, firstly, there is a clear internal coherence amongst the samples originating from the same


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

units, and this confirms our identification of Unit 12 made during the 2012 profile cleaning (Table 1, Fig. 3). Secondly, these data enable us to place both Units within the early stages of the Cantabro-Pyrenean Gravettian. The dates cover a large span of the development of the Gravettian industries in the Cantabrian region, at least in its better ~ a, 2009). The results known phases (Rasilla and Straus, 2004; Pen obtained from Unit 14 are clearly situated towards the upper part of the sequence, and we think it relevant to confirm the presence of Noailles burins, as is also the case at other regional sequences. Table 1 Radiometric dating from Units 12 and 14 at El Castillo calibrated using CalPal2007 (Weninger et al., 2008). Unit Sample

Method Material B.P.

12 12 12 14 14

AMS AMS AMS AMS AMS

Beta Beta Beta Beta Beta

242617 298431 298430 298433 298432

Bone Bone Bone Bone Bone

Cal B.P.

24.070 25.520 25.920 29.600 29.740

± ± ± ± ±

150 140 140 180 190

28,882 30,424 30,919 33,938 34,058

Source ± ± ± ± ±

391 348 345 282 244

Profile AMNH AMNH AMNH AMNH

4. Faunal remains The Gravettian fauna is represented by 1102 specimens distributed between Units 14 and 12. Obermaier's references to the fauna centre around the two most abundant species (horse and red deer), together with the great bovids and some carnivores (bear, hyena, wolf, and fox). A reindeer antler fragment from Unit 14 and the presence of chamois in Unit 12 were also noted. The results from the faunal revision, however, provide some new data, especially with regard to the ungulates in Unit 14. The revision confirms the predominance of horse and red deer, but in no way does it corroborate the ‘very abundant’ description made of the great bovid remains, as written in the notebooks from the final excavation season (Table 2). It is most surprising that they considered this taxon abundant when it has such small frequencies. Table 2 Number of identified specimens (NISP) for each of the ungulate species and their corresponding relative frequencies. Unit 14

Equus caballus Bovini Capra pyrenaica Rupicapra rupicapra Megaceros g. Cervus elaphus Capreolus capreolus Sthphanorhinus hemitoechus Probosc. ind. Ursus s. Canis l. Vulpes v. Crocuta c. Panthera leo Panthera pardus Lynx sp. Felis silvestris TOTAL

Unit 12

NISP

%NISP

NISP

%NISP

79 26 26 126 2 161 18 10 1 13 10 15 3 1 6

17.8 5.8 5.8 28.3

319 2

54.7 0.3

45

7.7

36.2 4.0 2.0

213 3 1

36.5 0.5 0.2

497

2 7 2 3 1 6 1 605

The differences in chamois frequencies are even more striking, as this species is simply noted as being ‘present’ by Obermaier, although it is represented by an important number of elements

465

both cranial (17 horn cores and 16 mandible fragments) as well as from the limbs (12 humeri; 14 tibiae; 14 metatarsals, amongst others). This omission is not a minor matter given that the relative proportion of chamois remains (25.4%) is the highest for the whole of the occupation of El Castillo. This relative proportion is only surpassed by that of red deer, and is very much above that of horse, which, according to the excavators, was very abundant. Although present in much lesser frequencies, the presence of Megaloceros and a proboscidean, most likely a mammoth, is worth noting. In Unit 12, horse and red deer together represent 87.9% of the total number of recovered specimens. This concentration/specialisation based around two species amongst the ungulate fauna can be considered a ‘preview’ of what would later emerge as the typical subsistence model of the Late Upper Palaeolithic. 5. The lithic industry The lithic collection from El Castillo has undergone a number of alterations since its excavation, which has led to the loss of some of its parts. The first, common to all sites excavated at the start of the twentieth century, is the excavation and recovery of the material itself. Although Obermaier and his team were very thorough when collecting material and with the stratigraphic control of the sites they excavated (as was the case at El Castillo and also at Hornos de ~ a), they nevertheless could not escape the theoretical conla Pen ditionings of the prehistoric discipline at the time. Therefore, part of the material was not collected because it went by unnoticed or because, according to the norms in place then, certain types of pieces did not exist in some techno-complexes, such as bladelets in the Aurignacian sensu Breuil. Its second alteration is as a result of the dispersal of El Castillo's material through time. As was common amongst different researchers and institutions then, the shipment of small collections to research centres and museums around the world led to the breaking up of part of the collections recovered during the 1910e1914 excavations. The final destinations of some of these lots are unknown to us and prevent us from fully understanding the true/full techno-typology of some of El Castillo's Units. This is the case of Units 12 and 14, which concern us here. This problem, however, has been partially solved thanks to H. Breuil's typed up lists and Boussonie's drawings, which depict many of the ‘missing’ pieces, a large number of which are of uttermost importance when interpreting these Units. The present study has re-examined a collection from El Castillo
gico Nacional (MAN). We have also deposited at the Museo Arqueolo
s (1984) data on the retouched pieces relied on Cabrera-Valde stored at the at the Museo de Arqueología de Cantabria (MUPAC), as well as on J. Boussonie's drawings of the 1910e1914 excavated material, which are deposited at the MAN. 5.1. Unit14 (Beta Aurignacian) It is a relatively rich unit, with 113 tools classifiable according to Sonneville-Bordes/Perrot's typology (Table 3). However, we only have 53 knapping remains, cores and five unknapped lithic pieces (anvils, retouchers). This could indicate the biased character of this Unit, although Obermaier noted the scarcity of archaeological
s, 1984; pp. 71). The original material within it (Cabrera-Valde counts from the excavation note 600 pieces, including d
ebitage,
s, 1984; pp. 258). In the anvils, hammers, etc. (Cabrera-Valde collection studied by us, the predominant raw material amongst the retouched pieces is flint (56%), followed by limestone (24%), quartzite (15%), and sandstone (3%) (Table 4). Amongst the unretouched pieces, the most represented raw materials are limestone (53%), flint (32.6%), sandstone, quartzite, and schist (4% each)


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

466

(Table 5). The knapped blanks do not generally show a cortical part or they possess it in less than a third of the dorsal surface (69.3% amongst the unretouched material and 79% amongst the retouched (Tables 6a and 6b).

Table 3 Typological list from El Castillo cave's Unit 14. Unit 14 MAN 1. Simple endscraper 2. Atypical endscraper 3. Double endscraper 5. Endscraper on retouched flake 11. Carinated endscraper 12. Atypical carinated endscraper 13. Thick nose endscraper 14. Flat nose endscraper. 15. Core endscraper 23. Borer 24. Atypical borer 27. Straight dihedral burin 28. D
ejet
e dihedral burin 30. Burin on angle.
burin 32. Busque 34. Burin on straight truncation 35. Burin on oblique truncation 41. Multiple burin 42. Noailles burin 43. Core-like burin 44. Flat burin 48. Gravette point 58. Complete backed piece 60. Straight truncated piece 61. Oblique truncated blade 65. Piece retouched in one edge. 66. Piece retouched in two edge. 67. Aurignacian blade. 74. Notch. 75. Denticulate. 76. Splintered piece. 77. Sidescraper. 93. Diverse Total

MUPAC

2 3 1 2 7 3 2 1

1 1

1 1

5 2 1 3 2 1 3 1

1 1 1

1 1 1

1 1

1 3 6 4 4 3 1 6 11 8 6 3 96

1 3 1 2

17

Total

%

2 4 1 3 7 3 2 1 1 6 2 2 4 3 1 3 1 1 1 2 1 1 3 6 5 7 3 2 8 11 8 6 3 113

1.8 3.5 0.8 2.6 6.2 2.6 1.8 0.8 0.8 5.3 1.8 1.8 3.5 2.6 0.8 2.6 0.8 0.8 0.8 1.8 0.8 0.8 2.6 5.3 4.4 6.2 2.6 1.8 7.1 9.7 7.1 5.5 2.6 100

Table 4 Inventory of the retouched material from Unit 14. Sandstone Blade Bladelet Indet. Flake Flake-blade Core Neo-crest Tectonic fragment Total

Quartzite

Flint

Limestone

Total

1

17 3 1 28 2

2

20 3 2 57 2 7 1 4 96

10

1 19

4

3

1 1 4 56

15

24

Table 5 Inventory of the unretouched material from Unit 14. Sandstone Quartzite Iron oxide Schist Flint Limestone Total Cobble 6 Chunck Blade Bladelet Flake Core Slab Semi-crest Semi-tablette Total 6

1

2

1 1

2

1

2

1 2 2 5 3 1 1 1 16

3 1 12 10

26

9 4 2 3 18 14 1 1 1 53

Table 6a Cortical distribution, Unit 14, unretouched blanks. >66%

100% Cobble Chunck Blade Bladelet Flake Core Slab Semi crest Semi tablette Total

<33%

33e66%

0%

9 1

3 4

3 2 3 12 6 1

3 2

2

1 16

1

2

1 28

6

Total 5 4 2 3 18 14 1 1 1 53

Table 6b Cortical distribution, Unit 14, retouched blanks. Retocado Blade Badelet Indet Flake Flake-blade Core Neo-crest Tectonic frag. Total

100%

1

>66%

2

33e66%

4

<33%

0%

III

Total

1

3

16 3

1 5

1 10

20 3 2 57 2 7 1 4 96

4 3 8

2

4

7

18

35 2 3 1 1 61

The retouched pieces assemblage is characterised by the greater presence of endscrapers as opposed to burins (IG ¼ 21.2, IB ¼ 16.8). The Aurignacian Group is more abundant (AG ¼ 15) than the Perigordian Group (PG ¼ 9.7). Amongst the endscrapers, the carinated, thick, and nosed stand out (12.3%), followed by simple endscrapers, on flake and atypical (Fig. 4: 6e8; Fig. 5: 12). Amongst the burins, those from the dihedral family are more numerous (IBd ¼ 7.9) than those produced on truncation (IBtr ¼ 3.5). Although less numerous, burins are more typologically varied: straight dihedrals, deviated on angle, fracture, with straight truncation, oblique and a number of core-like burins
and Noailles burins are worth noting (Fig. 5: (Fig. 5: 2, 5). A busque 1) although in Obermaier's field notes he mentions the presence of
s, 1984; pp. 260). 15 Noailles (Cabrera-Valde
s, 1984; pp. 247) two Gravette In earlier works (Cabrera-Valde points are noted. We have only identified one of them and two piece fragments with total backed, but because these are fractured this prevents us from being more specific (Fig. 4: 4e5). Abrupt retouch is better represented amongst the truncation group (9.7%). Substrate pieces (sidescrapers, notches and denticulates) comprise 22.1% of the total retouched pieces (Fig. 4: 9e10, 12e14), which represents a significant presence, as does the 7.1% of splintered pieces (Fig. 4: 1e3). Pieces with retouch on one or two edges represent 10.6% of the assemblage, and from which we should highlight two Aurignacian blades (Fig. 5: 7, 10, 13). From a technological point of view, we can only touch on some aspects of the chaînes op
eratoires employed given the small size of the MAN's sample. The disparity between the retouched and unretouched material of the collection under study (96 vs 49) is very significant. Roughly speaking, we can infer laminar and flake kinds of production. We believe laminar production, including blades and bladelets, was important despite the small number of pieces represented (22 blades and six bladelets), which represent respectively 15.3% and 4.15% of the collection (Fig. 5: 1e11). In terms of the collection's cores, there is great variety in their shape, but we believe they were looking for blanks with similar characteristics, blanks with little curvature, no torsion, regular,


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

Fig. 3. Calibration curve of El Castillo's datings using CalPal-HULU 2007 (Weninger et al., 2008).


e pieces; 4e5; Gravette points, 6e8, endscrapers; 9e10, denticulates; 11, borer; 12e14, notches (drawings J. Boussonie). Fig. 4. Lithics from Unit 14: 1e3, ecaille

467


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

468

narrow and obtained, in the majority, from unipolar cores (Fig. 5). Considering the small size of the sample, tentatively speaking, we believe that the production between blades and bladelets could be responding to a continuum starting from the prismatic cores, whereas a specific chaîne op
eratoire of bladelets existed from carinated and burin-type cores. In this unit we have inferred four groups of chaînes op
eratoires: prismatic with unipolar or bipolar exploitation, others of a carinated endscraper type and carinated burin type, chaînes op
eratoires specific to flakes and, tentatively speaking, a chaîne op
eratoire of bipolar flaking. 5.1.1. Unipolar/bipolar prismatic chaînes op
eratoires Production is carried out in blanks of a cubic morphology from fractured pebbles. The mise en forme consists of blade crest in the confluence between the surface of exploitation and one of the two flanks and, from there, it opens towards the opposite side (Fig. 5: 9). Such surfaces of exploitation are rectangular with the core flanks at right angles to them. The striking platforms are prepared
ratoires. The and resharpened using core tables on both chaînes ope
bitage is of unipolar direction, cores are more majority of the de numerous (n ¼ 5) than bipolar prismatics (n ¼ 2) (Fig. 6:4; Fig. 7:4; Table 7). The analysis of the blade and bladelet scars also supports this claim given 17 out of the 22 blades and five out of six bladelets show it, whereas bipolar scars are very scarce (Fig. 5: 5e11). We believe there is a continuum between the production of blades and bladelets.

Table 7 Cores from Unit 14. Core type Burin core Carinated endscraper core (width table) Carinated endscraper core (triangular table) Carinated endscraper core (lateral notch) Bipolar prismatic core Unipolar prismatic core Pyramidal Centripetal Polyhedrical Unipolar core (chopper type) Total

3 3 2 2 2 5 2 4 2 2 27

5.1.2. Endscraper/carinated burin-type chaînes op
eratoires The bladelets obtained from these two chaînes op
eratoires are of two kinds: rectilinear blanks and distal curvature on the cores with a wide d
ebitage surface (some of great size), whereas the carinated with triangular surface of exploitation or lateral notch, like the burin-type cores, could have bladelets of smaller size and torsion based on the cores' scars (Fig. 5: 3; Fig. 7: 1e3). The mise en forme of these methods is simple: in the carinated core the surface of exploitation begins to be knapped without prior preparation whereas the carinated with notches take advantage of the intersection between the notch and the surface of exploitation. 5.1.3. Flakes chaînes op
eratoires The majority of flint flakes do not present specific characteristics of particular chaînes op
eratoires, except in the case of a few pieces, which can be linked to the production of core crests or core tablets from laminar cores (Fig. 5: 14e15). The limestone flakes, however, can mostly exclusively be linked to unipolar/ pyramidal chaînes op
eratoires (n ¼ 3) and discoidal/centripetal (n ¼ 4) or polyhedric cores (Fig. 6: 1e3, 5). The limestone pieces

are larger than those produced on other raw materials as a result, without a doubt, of the greater size of the nodules in the surroundings. 5.1.4. Bipolar flaking (on anvil) The sample is too small to confidently suggest bipolar flaking. Even though some of the pieces classified as splintered pieces could be flakes or cores from this kind of bipolar flaking, based on ~ a, 2011b), at premicroscopic criteria defined experimentally (Pen sent we can only speculate of its possible existence (Fig. 4: 1e3). Amongst the unknapped lithic material there are two slate retouchers and three sandstone anvils showing obvious use wear. The bone industry is very poor and we only have two oval-sectioned sagaie of which only the mesial part is preserved. 5.2. Unit 12 e Alpha Aurignacian The youngest Gravettian is Unit 12, termed Alpha Aurignacian (Obermaier 1916). It has a fairly abundant lithic collection with 135 retouched tools classifiable according to Sonneville-Bordes/Perrot's typology, and only 47 knapping remains, amongst cores, flakes, blades, bladelets, and burin spalls (Tables 8 and 9). This collection is
s, 1984; pp. similar to that recovered by Obermaier (Cabrera-Valde 273). This clearly indicates the selection of pieces during excavation. There is a greater variety of raw materials in this Unit, although flint is the most used raw material, both amongst the unretouched and retouched material. Flint is followed, by a considerable margin, by the rest of the raw materials, such as quartzite, quartz (alpha and hyaline), and limestone. Blanks are in their majority non-cortical or present less than a third of cortex on their dorsal side: 80.7% amongst the unretouched material and 85.2% amongst the retouched, especially the blades and bladelets (Tables 10a and 10b).

Table 8 Inventory of retouched material from Unit 12. Quartzite Cobble Core rejuvenation flake Blade Indet Flake Flake-blade Core Neo-crest Semi-crest Total

Flint 1 11 1 11 1 1 1 1 28

1 1

2

Limestone

Total

1

3 1 12 1 13 1 1 1 1 32

1

2

Table 9 Inventory of unretouched material from Unit 12. Quartzite Core crest Cobble Chunck Burin spall Blade Bladelet Flake Core Semi crest Rejuvenation core flake Total

Quartz

Flint

Limestone

3 1

1 3 1

2

4

2 1 5 6 6 8 1 2 34

1

4 2

7

Total 3 1 3 1 5 7 13 11 1 2 47


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro Table 10a Cortex distribution, Unit 12, retouched blanks. 100% Cobble Core rejuvenation flake Blade Indet. Flake Flake-blade Core Neo-crest Semi-crest Total

Table 11 (continued ) MAN <33%

33e66%

0%

1

1

1 1 1 2

1

11 9 1

1 1 2

1

1 22

7

Total 3 1 12 1 13 1 1 1 1 32

Table 10b Cortex distribution, Unit 12, unretouched blanks. 100% Core crest Cobble Chunck Burin spall Blade Bladelet Flake Core Semi crest Rejuvenation core flake Total

>66%

33e66% 1 1

1 1 1

1

2 1

3

3

3

<33%

1

1 8 3 1 14

0%

Total

3

3 1 3 1 5 7 13 11 1 2 47

1 1 5 4 2 6 1 1 24

Amongst the retouched assemblage, the endscraper index is slightly lower than that of burins (IG ¼ 11.8, IB ¼ 13.3). Amongst the endscrapers (Fig. 8: 1e4), the most common are the atypical (3%) and the carinated (5.2%). Amongst the burins, the dihedral are more numerous than those produced on truncation (Table 11; Fig. 8: 5; Fig. 9: 2e4, 8). Obermaier notes a burin that could be Noailles were it not for its large size. Table 11 Typological list from El Castillo cave's Unit 12. MAN 1. Simple endscraper 2. Atypical endscraper 5. Endscraper on retouched flake 6. Endscraper on Augnacian blade. 8. Endscraper on flake 9. Circular endscraper 10. Nail endscraper 11. Carinated endscraper 12. Atypical carinated endscraper 15. Core endscraper 16. Rabot 17. Endscraper-burin 19. Burin-truncated blade 20. Borer-truncated blade 23. Borer 24. Atypical borer 26. Microborer 28. D
ejet
e dihedral burin 29. Dihedral burin on angle 31. Multiple dihedral burin 34. Burin on straight truncation 35. Burin on oblique truncation 40. Multiple burin on truncation 41. Multiple burin 44. Flat burin 58. Complete backed blade 59. Partial backed blade 60. Straight Truncated piece 61. Oblique truncated blade 65. Piece retouched in one edge.

1

1

1 1 3 1 1 3 2 1 2

469

MUPAC 4 1 1 1 1 1 1 4 1 2 1 1 2 2 2 1 1 4 1 1 2 1 1

6 5

Total 4 1 1 1 1 1 2 4 1 2 1 1 2 3 2 1 1 4 2 2 3 2 2 2 3 2 1 6 7

% 3 0.7 0.7 0.7 0.7 0.7 1.5 3 0.7 1.5 0.7 0.7 1.5 2.2 1.5 0.7 0.7 3 1.5 1.5 2.2 1.5 1.5 1.5 2.2 1.5 0.7 4.4 5.2

66. Piece retouched in two edge. 73. Pic 74. Notch 75. Denticulate 76. Splintered piece. 77. Sidescraper 84. Truncated bladelet 93. Diverse Total

2 5 2 5 1 32

MUPAC

Total

%

4 1 13 13 9 12 1 2 103

4 1 15 18 11 17 1 3 135

3 0.7 11.1 13.3 8.1 12.6 0.7 2.2 99.6

Pieces with abrupt retouch, characteristic of this technocomplex, are present in very small numbers: five backed pieces (3.7%) (Fig. 9: 5e7, 11; Fig. 10: 11e12). On the other hand, truncated pieces represent 5.1% of the assemblage (a total of seven pieces), the majority of which are oblique truncations (Fig. 10: 13). It is in this Unit that Obermaier noted the appearance of a Font-Robert point, a
s, 1984; pp. 278). piece now lost (Cabrera-Valde Substrate pieces are, once again, the most represented typological group (37%), with denticulates, sidescrapers and notches present in similar numbers. This fact, which should be considered with caution given the nature of the assemblage, represents an important datum when assessing the formation of the Gravettian and its relationship to earlier techno-complexes where substrate pieces continue to be abundant, thus debunking the argument that there are no flake chaînes op
eratoires during the Upper Palaeolithic
ndez, 2012). We should highlight, given the phylo(Maíllo-Ferna cultural affiliation they imply, the role of the splintered pieces sensu lato, with an 8.1% representation. Obermaier notes in the original list the presence of a number of Gravette points which
s (1984, pp. 273) associates with backed blades. Cabrera-Valde There was also a gibbeuse bladelet which has not been found in any of the preserved collections (Fig. 9: 6). There is little in the way of a microlaminar industry: in this Unit only a truncated bladelet from the MUPAC collection was noted. From a technological point of view we can only note some features of the chaînes op
eratories used given the sample's small size. This Unit's collection, stored at the MAN, has a larger number of unretouched than retouched pieces (47 vs 32). However, by adding the retouched material from MUPAC, the number of these pieces increases to a total of 135. Generally speaking, we can infer laminar and flake kinds of production (Table 12). Table 12 Cores from Unit 12. Cores Unit 12

N

Indeterminate Bipolar prismatic Carinated endscraper core Bipolar core (sur enclume) Unipolar (flakes) Total

1 7 1 1 3 13

5.2.1. Bipolar/unipolar prismatic chaînes op
eratoires Looking at the blanks from the collection under study here, together with Boussonie's original drawings, we can infer that production aimed for rectilinear, elongated blanks, with little or no torsion. The blanks used as cores are flint cubic fragments and, in some cases, on flake. The mise en forme is carried out from blade crests and semi crests, which, given their morphology, were likely made at the confluence between the surface of exploitation and one of the core's flanks (Fig. 9: 1e2; Fig 10: 1, 10, 16). These are, in every case, at right angles to the surface of exploitation. The surfaces of exploitation are rectangular and narrow. These are resharpened by

470


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

Fig. 5. Lithics from Unit 14: 1, 2, 4e5, burin spall, 3, burins; 6, 8, 11, blades; 7, 10, 13, Crested blade; 9, Pieces retouched on one edge; 12, double endscraper; 14e15, core preparation
ndez). flakes; 14, borer; 15, sidescraper (drawn by J-M. Maíllo-Ferna

means of partial crests and sometimes using resharpening core flakes. Striking platforms are prepared and resharpened by means of core tablets. Amongst the bipolar prismatic cores we believe there is a continuum between the production of blades and bladelets as attested by the various large-sized blade crests depicted in Boussonie's drawings (Fig. 9: 1,2). There is also a specific bipolar prismatic production for bladelets as attested by a number of cores on flake (Fig. 11: 2). Bipolar prismatic exploitation presents a large number of d
ebitage methods, which always make use of the rectangular and narrow surface of exploitation and, taking advantage of the core's cubic morphology, we find cores with various alternating surfaces of
bitage (two, three and even four surfaces of exploitation in their de exploitation on a single core), on occasion alternating bipolarity with unipolarity (Fig. 11). No prismatic cores of unipolar exploitation have been documented, however, the majority of blades and bladelets have unipolar scars (18/24), whereas pieces with bipolar or distal scars are

very scarce (3/24), but very common amongst the neo-crests and semi-crests (Fig. 10: 10). The absence of unipolar prismatic cores, together with the
bitage, can be abundance of blanks, which suggests this kind of de explained by means of two hypotheses: a) There were unipolar cores present that were not documented during the excavation; b) the blade cores were knapped in a bipolar manner and, when they reached the bladelet format, they were reconverted into bipolar
bitage cores. This latter hypothesis is likely the most plausible de given that a number of the few laminar blades present are longer than 6 cm and show bipolar scars, although given the small size of the collection this can only be contemplated as a hypothesis (Fig. 10: 14e15). 5.2.2. Carinated endscraper-type chaînes op
eratoires Only one core and one bladelet were obtained using this chaîne op
eratoire so we can only acknowledge the presence of this type of


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

471

5.2.3. Burin-type chaînes op
eratoires None of the few burins found in MAN's collection can be considered cores. However, besides a burin spall, there are three bladelets in the collection (out of a total of six for the whole collection) and one bladelet crest, which result from this kind of chaîne op
eratoire. Basing ourselves on Boussonie's original drawings, where he depicted a number of burins that may have been used as cores, the exploitation is undertaken using narrow surfaces of exploitation, which could indicate a certain degree of torsion of the obtained bladelets. Some of the bladelet pieces have a crest, although it is most common to take advantage of the edge between the ventral and
bitage. The surfaces of exploitation are dorsal sides to begin the de rectangular or slightly triangular and narrow (Fig. 9: 8).

Fig. 6. Lithics from Unit 14: 1e3, endscraper on flakes; 4, bipolar prismatic core; 5, Polyhedrical core (drawn by J. Boussonie).


bitage at El Castillo tentatively. Looking at Boussonie's drawings, de where a few other examples are represented (Fig. 8: 6), this kind of chaîne op
eratoire presents wide surfaces of exploitation with bladelet scars with distal curvature, rectilinear and lacking torsion.

5.2.4. Flake chaînes op
eratoires These represent the most common type of d
ebitage in the collection under study. As was the case in the previous Unit 14, the majority of flint flakes can be linked to the preparation of laminar cores. Chaînes op
eratoires of a unipolar kind (chopper kind) and centripetal on limestone and quartzite, however, have been documented. The mise en forme is simple: the d
ebitage begins directly on one of the cortical surfaces, maintaining the same role until the core is abandoned. In centripetal schemes, the surface of exploitation alternates between the d
ebitage surfaces. The flakes obtained tend to be more or less short with unipolar scars (Fig. 8: 7; Fig. 9: 8e9). One of the unipolar cores is a recycled anvil. It was not possible to link the quartz and rock crystal d
ebitage to any chaîne op
eratoire in particular. 5.2.5. Bipolar flaking (on anvil) There are only two pieces in the MAN collection that could be included as part of this kind of knapping strategy (there are nine


ndez). Fig. 7. Cores from Unit 14: 1e2, carinated endscraper core; 3, carinated burin core; 4, Bipolar prismatic core (bladelets) (drawn by J-M. Maíllo-Ferna

472


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

Obermaier's notes, a large-sized Noailles, now lost), a small number of dorsal pieces and a possible Font-Robert (also lost). Laminar production was carried out using bipolar prismatic cores and, to a lesser extent, unipolar prismatic cores in a possible continuum between blades and bladelets. The specific production of bladelets is also bipolar primastic and from a burin type and carinated endscraper. There is a mostly exclusive exploitation of flakes for secondary raw materials (limestone, sandstone, quartz, etc.) from cores of a unipolar or centripetal kind. Both Units present two common characteristics: firstly, the presence of splintered pieces and possible bipolar flaking, representing 7.1% of Unit 14 and 8.1% of Unit 12; secondly, the importance of substrate pieces (sidescrapers, denticulates and notches) amongst the retouched material, representing 22.1% of Unit 14 and 37% of Unit 12. Additionally, in both Units the type of mise en forme
bitage are very similar in terms of the laminar production, and de especially the prismatic kind. 6. El Castillo's Gravettian within its Cantabrian context

Fig. 8. Lithics from Unit 12: 1e4, endscraper; 5, burin core; 6, carinated endscraper core, centripetal core (drawn by J. Boussonie).

more pieces in the MUPAC collection). However, we cannot be sure that these kinds of pieces were cores or pieces employed in wedge activities given the similarity of the marks to their surfaces and the ~ a, 2011b: 95). Amongst the unknapped small size of the sample (Pen lithic industry are a hammer on limestone and two anvils on quartzite, one of them reconverted into a core. According to Obermaier's descriptions, the bone industry was composed of a thick point from a curved awl, another elongated awl with an engraving, and a number of mesial and basal awl fragments. In terms of mobile art, this is the Unit where the pebble with a feline figure engraving was found. 5.3. General features of the Gravettian at El Castillo The Gravettian of El Castillo is represented by two Units. Unit 14, the oldest, is characterised by larger numbers of endscrapers than burins, Noailles burins (one in the collection under study here, 15 noted in the original listing), and a few Gravette points and truncations. Technologically speaking, there is a production of blades and bladelets in continuum from prismatic unipolar cores and very few prismatic bipolar cores. There are specific chaînes op
eratoires for bladelets on carinated endscraper, carinated burin, and bipolar prismatic cores. On the other hand, the assemblage possesses other characteristics that link it to the Evolved Aurignacian, such as the
bitage of bladelets from carinated cores, some large in size, and de
-type burins. the appearance of a few Aurignacian blades or busque Flakes on secondary raw materials are knapped using discoid/ centripetal and unipolar/pyramidal chaînes op
eratoires. Unit 12 is characterised by a more-or-less equal number of endscrapers and burins (possibly including, according to

Until a few years ago, the weight of French Palaeolithic archaeology historiography was strongly felt in scholars' understanding of the Iberian Peninsula's Palaeolithic sequence, especially that of the Cantabrian region. Despite this, the introduction of the Perigordian model to the Upper Perigordian/Gravettian was not free of problems, both methodological and techno-typological. In this way, one of us pointed towards a clear relationship between the Cantabrian Gravettian with the Perigord. However, it presented certain anomalies, such as its strong techno-typological links with
s, 1982; pp. 227), the small the Aurignacian (Bernaldo de Quiro number of the dorsal pieces, and, lastly, its lack of typological links ~ a, with the French model (McCollough, 1971; Arrizabalaga and Pen 2013). These anomalies were also noted in the Mediterranean Gravettian, where self-identity processes are based on the appearance of backed pieces (Gravettes and microgravettes), the abundance of endscrapers, and little diachronic variability, further deepening the idea of this techno-complex’s regionalisation
n, 2013). (Villaverde et al., 2010; Villaverde and Roma The number of studies on the Gravettian along the Cantabrian coast has increased over the last decade, partly as a result of the appearance of new sequences and the revision of old collections using modern criteria. Therefore, this new corpus of Gravettian levels and data, although still small, allows us to begin shedding more light on the Gravettian's chronological and technotypological framework thanks to the works at Aitzbitarte III's exterior sequence, Bolinkoba (Iriarte-Chiapusso and Arrizabalaga, ~ ako Koba (Aguirre, 2013), Ametzagaina (Calvo et al., 2013), Antolin ~ oz, 2013), La Vin ~ a (Martínez and 2013), El Cuco (Rasines and Mun
n (Gonza
lez-Morales and Rasilla, 2013; Wood et al., 2014), El Miro Straus, 2013), Altamira (Heras et al., 2013), Alkerdi (Cava et al., 2009), Cueto de la Mina (Rasilla and Santamaría, 2006) and ~ a (Pinto-Llona et al., 2012) amongst others as well as sumSopen ~ a, 2011a). mary works on the region (Arrizabalaga, 1995; Pen The first thing we need to bear in mind is the Cantabrian Gravettian's strong links with that of the Western Pyrenees. Sites such as Gargas or Isturizt present strong links in their chronologies and techno-typologies with Cantabrian sequences (Foucher, 2013; Normand et al., 2013). The Cantabrian-Pyrenean Gravettian chronology can be placed at 34/33 ka cal BP (29/27 ka BP) based on the dates from Aitzbitarte ~ ako Koba levels III's exterior sector, Zatoya level IIbam, Antolin ne, Tuto de Camalhot, Tarte
and Lab þ Lmbk, Alkerdi 2, Gargas, Enle ~ a, 2013; Foucher, 2013). Some El Castillo 14 (Arrizabalaga and Pen authors have noted a chronological dichotomy in the Gravettian of the Cantabrian region, older in the western than central area,


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

473

Fig. 9. Lithics from Unit 12: 1, Core crest; 2, burin on core crest; 3e4, burins; 5e7, 10e11, backed pieces; 8, burin core; 9, flake (drawn by J. Boussonie).

perhaps as a result of level preservation differences between the ~ a, 2011a). Now, with El Castillo's new dates, both areas regions (Pen are now on the same level chronologically speaking. Therefore, in the Cantabrian region we find a very old Gravettian similar to the dates obtained from the Central European ‘nuclear Gravettian zone’ dated to 30/28 ka BP from the sites of Willendorf II level 5 (Austria), Molodova V levels 10e9 (Ukraine) around 30.5 ka BP or Geissenklosterle (Germany), whose first Gravettian occupations have been dated around 29.2 Ka BP (Haesaerts et al., 2007). This CantabroPyrenean Gravettian presents a series of distinctive features which, as we have already mentioned, unify this techno-complex in the region and distances it from the classical Perigord classification (Foucher et al., 2008). In its first stages it is characterised by the ~ a, 2011a), appearance of Noailles burins (Arrizabalaga, 1995; Pen which persist in smaller numbers during the rest of the Gravettian and can even be found in the Solutrean, such as at Cueva Morín
lez-Echegaray and Freeman, 1971), whereas these, in the (Gonza classic sequence, would pertain to the middle part of the Gravettian (Djindjiand et al., 1999). In addition, it also presents variable numbers of backed pieces, although generally not so numerous as in other regions. The presence of Font-Robert points is non-existent in the Pyrenees and incidental in the Cantabrian region such as at

Irikaitz (Arrizabalaga and Iriarte, 2011) whereas the rest of the points found in the region either pertain to old excavations such as those at Castillo or Cueva Morín (which are untraceable) or are ~ a (Martínez and catalogued as shouldered pieces, such as at La Vin Rasilla 2013). The Gravettian is present in the region until 23/ 24 ka cal BP (21 ka BP) based on typological observations. In this context the lithic industry at El Castillo perfectly encompasses the framework outlined above. Unit 14, the oldest, presents typological characteristics which place it in an early stage of the regional Gravettian, such as its Noailles burins (up to 15 if we go by H. Obermaier's notes), although far from the percentages ~ ako Koba or Isturitz (Aguirre, found at other sites such as Antolin 2013; Normand et al. 2013), Gravette points and truncated pieces. It, however, also presents characteristics that bring it closer to the
burins, the Evolved Aurignacian, such as the appearance of busque
bitage of bladelets from carinated endscraper cores unipolar de type, the appearance of Aurignacian blades. The carinated endscrapers represent a characteristic typical, up until now, of the western Cantabrian region such as El Castillo or Cueva Morín (Gonz
alez-Echegaray and Freeman, 1971) and unknown in such numbers in the Basque Country or Pyrenees. Unit 12, more recent, presents an ample use of bipolar prismatic d
ebitage, especially for


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

474

the last five years the first comparative analyses for this cultural
horizon have been undertaken both for marine (Alvarez, 2007) and
~ os and Alvarez, terrestrial fauna (Castan 2013). Additionally, there is an increase in the amount of information available from new sites, some recently published, as is the case of Aitzbitarte III (Altuna and Mariezkurrena, 2011). There are others, such as level VII of Cueto de ~ ako Koba which la Mina and the Gravettian levels from Antolin remain unpublished. Our present comparative analysis is based on the relative frequencies (%NISP) of ungulates present within the different samples from the Cantabrian coast. Species showing frequencies lower than 5% have been omitted. This is the case of wild boar, reindeer, Megaloceros, rhinoceros, and mammoth. The data used originate from 16 levels from 12 sites inconsistently distributed amongst the different territories of the region from Asturias to the Basque Country (Table 13).

bladelets, backed pieces (including a gibbeuse bladelet) and pieces now lost such as a possible atypical Noailles (due to its large size) and a Font-Robert point. With these techno-typological data and backed up by the radiocarbon dates obtained from the Unit (34/33 ka cal BP) we can state that Unit 14 corresponds to the first phases of the Gravettian, with reminiscences of the Evolved Aurignancian. Unit 12, on the other hand, with the characteristics presented above, could belong to a middle stage in the Cantabrian Gravettian; not much more can be said about it given this techno-complex’s polymorphism in the region. We cannot be more specific regarding this Unit as the internal sequence of the Cantabro-Pyrenean Gravettian is not yet well known beyond these two generic phases (one with a greater abundance of Noailles and the other with more dorsal pieces, although the Noailles do not disappear completely) nor is their chronology nor finer technological characterisation. In this manner

Table 13 Distribution of the number of identified specimens (NISP) and percentage of ungulates in the different levels and sites of the Cantabrian Region. Equus

Morin IV Morin Vb Pendo V-Va Santim VIII Bolinkoba VI Amalda V Amalda VI C. Mina G,H, VIII Riera I La Garma A Lezetxiki II Castillo 12 Castillo 14 Antolinako K. Aitz III G.antiguo Aitz III G. reciente

Bovini

Capra

Rupicapra

Cervus

Capreolus

Total

NISP

%

NISP

%

NISP

%

NISP

%

NISP

%

NISP

%

39 42 33 56 32 8 101 123 116 51 3 319 79 2 2 5

9.9 6.4 14.9 8.7 4.2 0.6 2.9 28.6 32 22.9 1.5 54 9 18.1 0.5 0.3 0.8

19 46

4.8 7

13 16

3.3 2.4

1 4

0.3 0.6

17 37 9 99 8 62 18 9 2 26 19 340 221

2.6 4.8 0.7 2.8 1.7 17.1 8.1 4.4 0.3 6 5.1 54.7 36.5

56 634 172 236 56 18 5 16

8.7 82.6 13.2 6.8 13 5 2.2 7.8

30 32 1022 2769

4.7 4.2 78.3 79.2

12.5 2.3 0.5 1.4 0.4

26 109

6 29.5

1

0.2

1.1 1.3 64.6 7.7 28.9 15.9 22.5 42.6

69.1 81.1 84.6 73.8 3.4 7.2 7.8 56.3 44.4 35.4 19.9 36.7 36.9 47 20.4 19.3

49 15 1 9 3

4 3 133 45 126 59 140 258

271 531 187 474 26 94 274 242 161 79 41 213 161 174 127 117

17 1 2 67 4 2 18 7 13 3

0.5 0.2 0.6 30 1.9 0.3 4.1 1.9 2.1 0.5

we must note the polymorphic character of the Gravettian in the central and western portions of the Cantabrian region with Units lacking diagnostic pieces of this techno-complex such as the Cuco rockshelter, where the most characteristic pieces are those of substrate and continuous retouch, but with a Gravettian ‘chronol~ oz, 2013); or the typologically ogy’ (28 ka cal BP) (Rasines and Mun
n 120 or Altamira 8 Gravettian, but scarce assemblages of El Miro
lez-Morales and Straus, 2013; Heras et al., 2013). On the (Gonza other hand, in a greater framework of analysis, the data from El Castillo, together with other sites such as Aitzbitarte III (RíosGaraizar et al., 2013), provide data in favour of a mosaic formation of the Gravettian, at least in the Cantabro-Pyrenean zone, as ~ a, 2011a, Ríoshas already been suggested by other authors (Pen Garaizar et al., 2013). Until recently no revisions of faunal assemblages had been published on the subject matter of Gravettian subsistence strategies in the Cantabrian region. In the last few decades of the twentieth century there were on-off references to this subject matter within studies of specific settlements. This is the case of II at Lezetxiki (Altuna, 1972), IV and Vb at Morín (Altuna, 1971, 1973), V-Va at El ~ os, 1982), Pendo (Fuentes, 1980), G and H at Cueto de la Mina (Castan ~ os, 1983), VIII at Santimamin ~ e (Castan ~ os, VI at Bolinkoba (Castan 1984), I at La Riera (Altuna, 1986), V and VI at Amalda (Altuna, 1990). This topic has also been covered, in a broader way, in PhDs
s, 1982), the on the Early Upper Palaeolithic (Bernaldo de Quiro Middle-to-Upper Palaeolithic Transition (Soto-Barreiro, 2003) and ~ a, 2011a). In in some lithic revision works on the Gravettian itself (Pen

392 655 221 642 764 1305 3496 430 363 223 206 581 436 370 622 605

There is a lack of uniformity in the distribution of data throughout the Cantabrian region with poorer samples derived from Asturias and Bizkaia. The assemblages originating from the eastern end of the region (Guipuzkoa) are more abundant than those found in the rest of the territories under consideration here. The differences from a spatial point of view are not exclusively quantitative in character. Differences in the relative frequencies of some species can also be observed. The most outstanding variations are observed between the horse and the chamois. In the case of the horse, the relative proportions in the number of identified specimens show a clear decrease from the western to the eastern edges of the region (Fig. 12). This fall in numbers can be considered quite gradual, although the most abrupt fall takes place between Cantabria and Bizkaia. The opposite trend is true for chamois, which has a mostly residual presence in Asturias, but then goes on to become the most predominant species in the two levels at Amalda. In this case, the increase in chamois numbers towards eastern Cantabria is not gradual. On the one hand, there is an important leap between Asturias and Cantabria-Bizkaia. The most abrupt increase, however, takes place between Cantabria-Bizkaia and Gipuzkoa. Undoubtedly the size of Amalda's assemblage plays an important role in this matter. It is worth noting that the predominance of chamois in Wurm III assemblages in Gipuzkoa was already observed over two decades ago (Altuna 1990). However, although the highest numbers for chamois are found in the most eastern part of the Cantabrian territory, there are isolated data from other areas that point in the same direction. From this point of


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

475

Fig. 10. Lithics from Unit 12: 1, Semi crest; 2, bladelet; 3e4, 6e7, bladelets from burin core; 5, bladelet from carinated endscraper core; 8, retouched blade; 9, burin; 10, crest blade; 11e12, backed pieces; 13, truncation; 14e15, 17, blades; 16, crest blade (drawn by J-M. Maíllo-Fern
andez).

view, it is interesting to note that, at a site as stratigraphically strong and fauna-rich as El Castillo, the highest relative frequencies of chamois correspond to Aurignacian and Gravettian occupations. The most important aspects to note here, however, are the notable differences in hunting patterns of Palaeolithic groups. One of the most cited interpretative hunting strategy models for the Palaeolithic in the Cantabrian region contemplates the evolution from a rather unselective kind of hunting during the Mousterian to a specialised capture of a single species, which reaches its maximum expression in the Magdalenian. During the Early Upper Palaeolithic, the transition between these two subsistence models would have taken place. As a result, the fauna from the Gravettian, together with that from the Aurignacian levels, is of great interest to us here when considering this hypothesis. The presence of specialised hunting strategies would be suggested by more than 70% of the total number of identified specimens pertaining to a single

ungulate species. And non-specialised hunting would require the presence (15% or more) of at least three different species. This criterion was suggested more than two decades ago, and its application is of relevance and utility to us here (Altuna, 1990). With the data currently available to us we can note that in the faunal assemblages of the 16 levels under consideration here there are seven in which the predominance of a single species would suggest specialised hunting during these periods. In four of them ~ e) the predominant spe(two at Morín, El Pendo and Santimamin cies is red deer (Table 13). At Bolinkoba, given its rocky setting, the predominant species is goat, and at two levels in Amalda, chamois. In the remaining levels there are four in which there are three species whose relative frequencies exceed 15% of the total assem~ ako Koba and two at Aitzbitarte III). These blage (Castillo 14, Antolin would represent a diversified kind of hunting strategy. The three best represented species are never the same three except at the two

476


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

In summary, the data indicate that during the Gravettian levels with three well-represented species slightly exceed those with a single species. In this way, in terms of the species selected, a wide range of hunting strategies can be inferred during the Gravettian, as opposed to the very specialised tactics observed during the Solutrean, and especially during the Magdalenian, whose cultural horizons are more regularly dominated by a single species. Therefore, based on these data, a transitional model based on these two kinds of hunting strategies can be suggested here as it explains, reasonably well, most of the faunal data available to us at present. However, a modern revision of the Mousterian faunas has indicated that the proportion of levels with a singles species is in equilibrium or even exceeds those with a wider faunal spectrum. In the future this will force us, when more data are available, to analyse in greater detail the possible influence of other factors, such as the setting and functionality of each site, in the interpretation of the hunting strategies of Palaeolithic groups. 7. Conclusions The revision of El Castillo's Gravettian, despite the problems relating to the time when it was initially excavated and the dispersal of its lithic collections amongst several institutions, allows us to put forward a number of significant conclusions.

Fig. 11. Bipolar prismatic cores from Unit 12 (drawn by J-M. Maíllo-Fern
andez).

levels at Aitzbitarte III. In the case of the remaining four (Cueto de la Mina, La Riera, La Garma and Unit 12 at El Castillo), only two species represent more than 15% of the assemblage and, in three of them, these are horse and red deer.

- The lower part of the Gravettian sequence (Unit 14), with a date of 34/33 cal ka BP, and the presence of Noailles burins, links this Unit to the first instances of the Gravettian in the CantabroPyrenean region with a chronology similar to that of the beginning of this techno-complex in the eastern area of the Cantabrian region. - Unit 14 has some characteristics reminiscent of the Cantabrian Evolved Aurignacian, such as the production of bladelets from
burin or carinated cores, some typical Aurignacian tools (busque Aurignancian blades), as well as a number of splintered pieces/ bipolar flaking. Therefore, we can view this Unit as being close to the transition between both techno-complexes. - Unit 12 presents different techno-typological characteristics to those from Unit 14, such as the absence of Noailles and the greater presence, although still not widespread, of abrupt retouch. - The fauna from Unit 14 is dominated by the remains of red deer and chamois, and by red deer and horse in Unit 12, the two pairs representative of faunal assemblages found later in the Final Upper Palaeolithic. The revision of the fauna has enabled us to note that Obermaier's original observations on the fauna do not correspond with the fauna observed from Unit 14. - The Gravettian lithic industry from El Castillo can be placed within the Gravettian framework of the Cantabro-Pyrenean region and divided into two phases: a first with Noailles burins and a second with a greater presence of abrupt retouch. This scheme moves away from the Perigordian model where Noailles are typical of the middle part of the Gravettian. Additionally, given the chronological evidence, the Cantabro-Pyrenean Gravettian would be found around 34/33 ka cal BP, during the same chronological timeframe as the Central European Gravettian. This evidence allows us to evaluate, with some argumentative weight, the hypothesis of a multiple origin in the formation of the Gravettian. Acknowledgements

Fig. 12. Relative frequencies of Equus and Rupicapra remains in different territories of the Cantabria region.

This work was carried out with funding from project HAR2012
n General de Proyectos de Investigacio
n, 35214 of the Subdireccio
n General de Investigacio
n Científica y Te
cnica of the Direccio


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro

Ministry of Economy and Competitiveness of Spain. We are grateful to Dr Ian Tattersal, Conservator at the Anthropology Department of the American Natural History Museum, for providing us with all manner of facilities when we carried out our sampling. We would
gico Nacional also like to thank the staff at the Museo Arqueolo (MAN), especially Dr Carmen Cacho and Juan Antonio Martos, who also provided all manner of facilities so we could carry out our ~ a provided help with the analysis of research. Dr Paloma de la Pen the datings and provided insightful comments on the text. We are
s for translating the article from its grateful to Dr Pía Spry-Marque original version in Spanish.

References ~ ako koba: aproximacio
n Aguirre, M., 2013. Ocupaciones gravetienses de Antolin preliminar a su estratigrafía, cronología e industrias. In: Heras, C., Lasheras, J.A., Arrizabalaga, A., Rasilla, M. (Eds.), Pensando el Gravetiense: nuevos datos para
n canta
brica en su contexto peninsular y pirenaico, Museo y centro de la regio
n de Altamira, Monografías, vol. 23, pp. 229e241. Santander. Investigacio
rico de Morín (Santander). Altuna, J., 1971. Los Mamíferos del yacimiento prehisto
lez-Echegaray, J., Freeman, L.G. (Eds.), Cueva Morín. Excavaciones In: Gonza
ricas de Santander, vol. 6, 1966e68, Publicaciones del Patronato Cuevas prehisto pp. 67e398. Santander.
ricos de Guipúzcoa. Altuna, J., 1972. Fauna de mamíferos de los yacimientos prehisto
logo de los mamíferos cuaternarios del cant
Con cata abrico y Pirineos occidentales. Munibe 24, 1e464.
lezAltuna, J., 1973. Fauna de Mamíferos de la Cueva Morín (Santander). In: Gonza Echegaray, J., Freeman, L.G. (Eds.), Cueva Morín. Excavaciones 1969, Pub
ricas de Santander, vol. 10, licaciones del Patronato Cuevas prehisto pp. 279e290. Santander. Altuna, J., 1986. The mammalian faunas from the prehistoric site of La Riera. In: Straus, L.G., Clark, G. (Eds.), La Riera Cave. Stone Age Hunter Gatherer Adaptations in Northern Spain, Anthropological Research Paper 36. Arizona State University, Tempe, pp. 237e274, 421e479; 481e497.
n precedente de Macromamíferos durante el Altuna, J., 1990. Caza y alimentacio
n, A., Mariezkurrena, K. (Eds.), Cueva Paleolítico de Amalda. In: Altuna, J., Baldeo de Amalda (Zestoa, País Vasco). Ocupaciones Paleolíticas y Postpaleolíticas, Sociedad de Estudios Vascos Serie B 4, pp. 149e192. San Sebastian. Altuna, J., Mariezkurrena, K., 2011. Estudio de los macromamíferos del yacimiento
n de la entrada). In: Altuna, J., Mariezkurrena, K., de Aitzbitarte III (Excavacio Ríos, J. (Eds.), Ocupaciones humanas en Aitzbitarte III (País Vasco) 33.600e18.400 BP. Servicio Central de Publicaciones del Gobierno Vasco, Vitoria-Gasteiz, pp. 395e478.

n de los moluscos marinos en la cornisa cant
Alvarez, E., 2007. La explotacio abrica durante el Gravetiense: Primeros datos de los niveles E y G de La Garma ~ o, Cantabria). Zephyrus 60, 43e58. (Omon Arrizabalaga, A., 1995. La industria lítica del Paleolítico superior inicial en el oriente cant
abrico (PhD thesis). Universidad del País Vasco, Spain. Arrizabalaga, A., Iriarte, Mª J., 2011. Lower and Upper Palaeolithic settlements in Irikaitz (Zestoa, Basque Country, Spain). Deconstruction of a Pleistocene archaeological site in the eastern Cantabrian Range. Cuaternario Y Geomorfología 25, 105e120. ~ a, P. de la, 2013. El registro de la industria lítica como base para Arrizabalaga, A., Pen
n del Gravetiense canta
brico. In: Heras, C., Lasheras, J.A., una organizacio Arrizabalaga, A., Rasilla, M. (Eds.), Pensando el Gravetiense: nuevos datos para
n canta
brica en su contexto peninsular y pirenaico, Museo y centro de la regio
n de Altamira, Monografías, vol. 23, pp. 367e388. Santander. Investigacio
s, F., 1982. Los inicios del Paleolítico Superior Cant
Bernaldo de Quiro abrico. Min
n general de Bellas Artes y Archivos, Madrid. isterio de Cultura. Direccio
s, F., Maíllo-Ferna
ndez, J.M., 2009. Middle to Upper Palaeolithic at Bernaldo de Quiro Cantabrian Spain. In: Camps, M., Chauhan, P.R. (Eds.), A Sourcebook of Palaeolithic Transitions: Methods, Theories and Interpretations. Springer, pp. 341e359. EEUU.
s, F., Maíllo-Ferna
ndez, J.M., Neira, A., 2010. La cueva de El CasBernaldo de Quiro tillo: perspectivas desde el siglo XXI. In: Monografíes del SERP 8. SERP, Barcelona, pp. 292e310.
olithique Supe
rieur et leur signification. Breuil, 1912. Les subdivisions du Pale
s International d’Archeologie pre
historiques. Compte Rendu de la XIV Congre session, Ginebra, pp. 165e238.
s, V., 1979. El yacimiento de la Cueva de “El Castillo”. (Puente Viesgo, Cabrera-Valde Santander) (PhD. thesis). Universidad Complutense de Madrid, Spain.
s, V., 1984. El yacimiento de la Cueva de “El Castillo”. (Puente Viesgo, Cabrera-Valde Santander). C.S.I.C, Madrid.
s, V., Maíllo-Ferna
ndez, J.M., Lloret, M., Bernaldo de Quiro
s, F., 2001. Cabrera-Valde
olithique supe
rieur dans la grotte du Castillo (CanLa transition vers le Pale tabrie, Espagne): la couche 18. L'Anthropologie 105, 505e532. Calvo, A., Tapia, J., Arrizabalaga, A., Iriarte-Chiapusso, M.J., 2013. El yacimiento de Ametzagaina (Donostia, País Vasco). Un campamento gravetiense al aire libre
brico. In: Heras, C., Lasheras, J.A., Arrizabalaga, A., Rasilla, M. (Eds.), en el Canta
n cant
Pensando el Gravetiense: nuevos datos para la regio abrica en su contexto

477


n de Altamira, Monopeninsular y pirenaico, Museo y centro de Investigacio grafías, vol. 23, pp. 243e254. Santander. ~ os, P., 1982. Estudio de los Macromamíferos del yacimiento prehisto
rico de Castan Cueto de la Mina (Asturias). Boletín del Instituto de Estudios Asturianos 105e106, 43e63. ~ os, P., 1983. Estudio de los Macromamíferos del yacimiento prehisto
rico de Castan Bolinkoba. Kobie 13, 261e298. ~ os, P., 1984. Estudio de los Macromamíferos de la cueva de Santimamin ~e Castan (Vizcaya). Kobie (Serie Paleoantropología Y Ciencias Naturales) 14, 235e318.
~ os, P., Alvarez, Castan E., 2013. Nuevas aportaciones a las bases de subsistencia de
brico. In: Heras, C., Lasheras, J.A., origen animal durante el Gravetiense canta Arrizabalaga, A., Rasilla, M. (Eds.), Pensando el Gravetiense: nuevos datos para
n cant
la regio abrica en su contexto peninsular y pirenaico, Museo y centro de
n de Altamira, Monografías, vol. 23, pp. 331e347. Santander. Investigacio Cava, A., Elorrieta, I., Barandiar
an, I., 2009. El Gravetiense de la Cueva de Alkerdi (Urdax, Navarra): an
alisis y contexto de su industria lítica. Munibe 60, 51e80.
olithique supe
rieur en Europe. Djindjiand, F., Kozlowski, J.K., Otte, M., 1999. Le pale Armand Collins, Paris. se chrono-culturelle sur le Gravettien des Pyre
ne
es: conFoucher, P., 2013. Synthe
flexions sur la stabilite
re
gionale des traditions techniques. In: stat et re Heras, C., Lasheras, J.A., Arrizabalaga, A., Rasilla, M. (Eds.), Pensando el Grave
n canta
brica en su contexto peninsular y tiense: nuevos datos para la regio
n de Altamira, Monografías, vol. 23, pirenaico, Museo y centro de Investigacio pp. 151e168. Santander. Foucher, P., San Juan-Foucher, C., Sacchi, D., Arrizabalaga, A., 2008. Le Gravettien des
ne
es. Pale
o 20, 331e356. Pyre Fuentes, C., 1980. Estudio de la fauna de el Pendo. Madrid. In: Gonz
alez-Echegaray, J. (Ed.), El yacimiento de la cueva de El Pendo (Excavaciones 1953-1957), Bibliohteca Praehistorica Hispanica, vol. 17, pp. 215e237. Garrod, D., 1938. The Upper Palaeolithic in the light of recent discovery. Proceedings of the Prehistoric Society 4, 1e26.
lez-Echegaray, J., Freeman, L.G., 1971. Cueva Morín. Excavaciones 1966e68. Gonza
ricas de Santander, vol. 6. In: Publicaciones del Patronato Cuevas prehisto Santander.
lez Echegaray, J., 2013. El Gravetiense cant
Gonza abrico. In: Heras, C., Lasheras, J.A., Arrizabalaga, A., Rasilla, M. (Eds.), Pensando el Gravetiense: nuevos datos para
n cant
la regio abrica en su contexto peninsular y pirenaico, Museo y centro de
n de Altamira, Monografías, vol. 23, pp. 25e26. Santander. Investigacio
lez-Morales, M., Straus, L.G., 2013. La ocupacio
n Gravetiense de la cueva de El Gonza
n (Ranales de la Victoria, Cantabria) y el contexto del arte paleolítico Miro
n. In: Heras, C., Lasheras, J.A., Arrizabalaga, A., temprano de la cuenca del Aso
n canRasilla, M. (Eds.), Pensando el Gravetiense: nuevos datos para la regio
brica en su contexto peninsular y pirenaico, Museo y centro de Investigacio
n ta de Altamira, Monografías, vol. 23, pp. 305e316. Santander. Haesaerts, P., Borziac, I., Chirica, V., Damblom, F., Koulakovska, L., 2007. Cadre
o 19, stratigraphiqie et chronologique du Gravettien en Europe Centrale. Pale 31e52. Heras, C., Montes, R., Lasheras, J.A., 2013. Altamira: nivel gravetiense y cronología de su arte rupestre. In: Heras, C., Lasheras, J.A., Arrizabalaga, A., Rasilla, M. (Eds.),
n canta
brica en su contexto Pensando el Gravetiense: nuevos datos para la regio
n de Altamira, Monopeninsular y pirenaico, Museo y centro de Investigacio grafías, vol. 23, pp. 501e516. Santander. Iriarte-Chiapusso, M.J., Arrizabalaga, A., 2013. El Gravetiense de Bolinkoba (Bizkaia) a la luz de las excavaciones antiguas y recientes. In: Heras, C., Lasheras, J.A., Arrizabalaga, A., Rasilla, M. (Eds.), Pensando el Gravetiense: nuevos datos para
n cant
la regio abrica en su contexto peninsular y pirenaico, Museo y centro de
n de Altamira, Monografías, vol. 23, pp. 217e227. Santander. Investigacio Klaric, L., 2006. Regional groups in the European Middle Gravettian: a reconsideration of the Rayssian technology. Antiquity 81 (311), 176e199.
olithique supe
rieur et Me
solithique entour le MediKozlowski, J.K., 2005. Pale
rrane
en. L'anthropologie 109, 520e540. te Maíllo-Fern
andez, J.-M., 2012. Missing lithics: the role of flakes in the Early Upper Palaeolithic of the Cantabrian Region (Spain). In: Pastoors, A., Peresani, M. (Eds.), Flakes Not Blades: the Role of Flake Production at the Onset of the Upper Palaeolithic in Europe. Wissenschaftliche Schriftendes Neanderthal Museums, Mettmman, pp. 69e84.
s, F., S
Maíllo-Fern
andez, J.M., Garralda, M.D., Bernaldo de Quiro anchezFern
andez, G., 2011. The Middle-Upper Palaolithic transition in the Cantabrian Region: a mosaic model. In: Derevianko, A.P., Shunkov, M.V. (Eds.), Characteristic Features of the Middle to Upper Paleolithic Transition in Eurasia. Asian Palaeolithic Association, Novosibirsk, pp. 135e149.
n y novedades. In: Martínez, L., Rasilla, M., 2013. El Gravetiense en Asturias: revisio Heras, C., Lasheras, J.A., Arrizabalaga, A., Rasilla, M. (Eds.), Pensando el Grave
n canta
brica en su contexto peninsular y tiense: nuevos datos para la regio
n de Altamira, Monografías, vol. 23, pirenaico, Museo y centro de Investigacio pp. 291e303. Santander. McCollough, M.R.C., 1971. Perigordian facies in the Upper Paleolithic of Cantabria (PhD thesis). University of Pennsylvania, USA. € sterle. The Swabian Gravettian in its European context. Moreau, L., 2010. Geißenko Quart€ ar 57, 79e93. re, J., Simonet, A., 2013. El Gravetiense de la cueva Normand, Ch, Goutas, N., Lacarrie de Isturitz: nuevas investigaciones, nuevos datos. In: Heras, C., Lasheras, J.A., Arrizabalaga, A., Rasilla, M. (Eds.), Pensando el Gravetiense: nuevos datos para
n cant
la regio abrica en su contexto peninsular y pirenaico, Museo y centro de
n de Altamira, Monografías, vol. 23, pp. 171e193. Santander. Investigacio

478


s et al. / Quaternary International 359-360 (2015) 462e478 F. Bernaldo de Quiro


sil. Comisio
n de Investigaciones Prehisto
ricas y Obermaier, H., 1916. El Hombre Fo
gicas, Madrid. Paleontolo Otte, M., 1985. Le Gravettien en Europe. L'Anthropologie 89, 479e503. ~ a, P. de la, 2009. Revisio
n crítica de los conjuntos líticos gravetienses y su Pen
gico en la Península Ibe
rica. Complutum 20 (1), 29e53. contexto arqueolo ~ a, P. de la, 2011a. Sobre la unidad tecnolo
gica del Gravetiense en la Península Pen
rica: implicaciones para el Paleolítico superior inicial (PhD thesis). UniIbe versidad Complutense de Madrid, Spain. ~ a, P. de la, 2011b. Sobre la identificacio
n macrosco
pica de las piezas astilladas: Pen propuesta experimental. Trabajos de Prehistoria 68, 79e98. ~ a, P. de la, 2012. A propo
sito del Gravetiense… El paso de cultura a tecnocomPen plejo: un caso ejemplar de pervivencia particularista. Complutum 23 (1), 41e62.
re. Bulletin Peyrony, D., 1933. Les industries aurignaciennes dans le bassin de le Veze
te
Pre
historique Française, 543e559. de la Socie
rigordien et l’Aurignacien. Bulletin de la Socie
te
Peyrony, D., 1936. Le Pe
historique Française 33, 616e619. Pre Peyrony, D., 1946. Une mise au point au subjet de l’Aurignacien et du Perigordien.
te
Pre
historique Française, 232e237. Bulletin de la Socie Pinto-Llona, A.C., Clark, G., Karkanas, P., Blackwell, B., Skinner, A.R., Andrews, P., ~ a RockReed, K., Miller, A., Macías-Rosado, R., Vakiparta, J., 2012. The Sopen shelter, a new site in Asturias (Spain) bearing evidence on the Middle and Early Upper Palaeolithic in Northern Iberia. Munibe 63, 45e79.
n cant
Rasilla, M., Straus, L.G., 2004. El poblamiento en la regio abrica en torno al último
ximo glacial: Gravetiense y Solutrense. Kobie (Serie de Anejos) 8, 209e242. ma Rasilla, M., Santamaría, D., 2006. El Gravetiense del Abrigo de Cueto de la Mina
ndez, J.M., Baquedano, E. (Eds.), (Posada de Llanes, Asturias). In: Maíllo-Ferna
nea en homenaje a Victoria Cabrera, Zona Arqueolo
gica, 7, vol. I, Miscela pp. 422e439. Alcal
a de Henares. ~ oz, E., 2013. Los niveles gravetienses del abrigo de El Cuco (Castro Rasines, P., Mun Urdiales, Cantabria). In: Heras, C., Lasheras, J.A., Arrizabalaga, A., Rasilla, M.
n canta
brica en su (Eds.), Pensando el Gravetiense: nuevos datos para la regio
n de Altamira, contexto peninsular y pirenaico, Museo y centro de Investigacio Monografías, vol. 23, pp. 255e277. Santander. ~ a-Alonso, P. de la, Maíllo-Ferna
ndez, J.M., 2013. El final del Ríos-Garaizar, J., Pen ~ aciense y el comienzo del Gravetiense en la regio
n cant

n Aurin abrica: una visio


gica. In: Heras, C., Lasheras, J.A., Arrizabalaga, A., Rasilla, M. (Eds.), tecno-tipolo
n canta
brica en su contexto Pensando el Gravetiense: nuevos datos para la regio
n de Altamira, Monopeninsular y pirenaico, Museo y centro de Investigacio grafías, vol. 23, pp. 389e402. Santander.
ne
es. Des armes aux socie
te
s (PhD Simonet, A., 2009. Les gravettiens des Pyre
de Toulouse-Le Mirail, France. thesis). Universite
trica, ecología y clima del Paleolítico Soto-Barreiro, M.J., 2003. Cronología radiome
n de Altamira, monoCant
abrico. In: Museo Nacional y Centro de Investigacio grafía, vol. 19. Madrid. Tejero, J.M., Avezuela, B., White, R., Ranlett, S., Quam, R., Tattersall, I., Bernaldo de
s, F., 2010. Un pedazo de la Prehistoria ca
ntabra en Nueva York. Las Quiro Colecciones de la Cueva de El Castillo (Puente Viesgo, Cantabria) en el American Museum of Natural History (Nueva York, EEUU). Munibe 61, 5e16. Villaverde, V., Rom
an, D., Martínez, R., Badal, E., Bergad a, M.M., Guillem, P.M.,
rez, M., Tormo, C., 2010. El Paleolítico superior en e País Valenciano. NovePe dades y perspectivas. In: Mangado, X. (Ed.), El Paleolítico Superior Peninsular Novedades del siglo XXI, Monografies SERP 8, pp. 85e113.
rica: Villaverde, V., Rom
an, D., 2013. El Gravetiense de la vertiente mediterr
anea ibe
n y perspectivas. In: Heras, C., Lasheras, J.A., Arrizabalaga, A., estado de la cuestio
n canRasilla, M. (Eds.), Pensando el Gravetiense: nuevos datos para la regio
brica en su contexto peninsular y pirenaico, Museo y centro de Investigacio
n ta de Altamira, Monografías, vol. 23, pp. 39e59. Santander. White, R., 2006. From Puente Viesgo to Central Park West: Hugo Obermaier, Nels Nelson and the American Museum of Natural History's collections from Can
nea en tabrian Spain. In: Maíllo-Fern
andez, J.M., Baquedano, E. (Eds.), Miscela
gica 7, vol. I. Museo Arqueolo
gico homenaje a Victoria Cabrera, Zona Arqueolo
de Henares, pp. 58e77. Regional, Alcala €ris, O., Danzeglocke, U., 2008. CalPal-2007. Cologne Radiocarbon Weninger, B., Jo Calibration & Palaeoclimate Research Package. http://www.calpal.de/ (accessed 11.10.10.). Wood, R.E., Arrizabalaga, A., Camps, M., Fallon, S., Iriarte-Chiapusso, M.-J., Jones, J., Maroto, J., de la Rasilla, M., Santamaría, D., Soler, J., Soler, N., Villaluenga, A., Higham, T.F.G., 2014. The chronology of the earliest Upper Palaeolithic in ~ a. Journal northern Iberia: new insights from L’Arbreda, Labeko Koba and La Vin of Human Evolution. http://dx.doi.org/10.1016/j.jhevol.2013.12.017.