Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spain

JASREP-00858; No of Pages 10 Journal of Archaeological Science: Reports xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Journal of Archaeological Science: Reports journal homepage: www.elsevier.com/locate/jasrep

Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spain) Lisa M. Fontes a,b,⁎, Lawrence Guy Straus b,c, Manuel R. González Morales c a b c

EcoPlan Associates, Inc., 701 West Southern Avenue, Suite 203, Mesa, AZ 85210, USA Department of Anthropology, University of New Mexico, MSC01-1040, 1 University of New Mexico, Albuquerque, NM 87131, USA Instituto Internacional de Investigaciones Prehistóricas de Cantabria, Universidad de Cantabria, Avenida de los Castros s/n, 39005 Santander, Spain

a r t i c l e

i n f o

Article history: Received 2 November 2016 Received in revised form 4 March 2017 Accepted 8 March 2017 Available online xxxx Keywords: Upper Paleolithic Cantabrian Lower Magdalenian Spain Lithic technology Mobility Landscape use

a b s t r a c t Paleolithic archaeologists have a longstanding interest in temporal change in prehistoric human behavior, and have often identified changes between archaeological periods based on sampling limitations. This analysis focuses on the Cantabrian Lower Magdalenian period in El Mirón cave, where archaeologists have been able to subdivide the ~ 33 cm thick Level 17 palimpsest into 13 comparable units that provide insight into human behavioral change within an archaeological period. The authors focus on lithic raw material provisioning as a window into changing human-landscape relationships. The Level 17 sublevel lithic assemblages testify that hunter-gatherers decreased their use of presumably local (based on El Mirón project surveys) mudstones and quartzites in favor of regional flint resources. Additionally, portions of exotic flints from southern France and the Ebro Basin increase over the course of the Lower Magdalenian, possibly indicating gradually intensifying social contacts. Microstratigraphic units provide insight into dynamic “Magdalenian” landscapes and how lithic provisioning related to hunter-gatherers' territories, site catchments, regional mobility, and social relationships. © 2017 Elsevier Ltd. All rights reserved.

1. Introduction Paleolithic prehistorians have a longstanding interest in understanding human resilience and adaptation to Late Glacial landscapes and environmental conditions. Much of the Upper Paleolithic archaeological record in western Europe is known from cave and rockshelter settings, where the remnants of human occupations that archaeologists unearth have been influenced by both cultural and natural formation processes (Bailey, 1983, 2007; Straus, 1979). These palimpsest deposits have influenced both archaeological knowledge of long-term human behavioral trends and archaeological understanding of time in Paleolithic archaeology (Bailey, 1983, 2007; Foley, 1981). Palimpsest deposits can be incredibly complex archaeological layers where single occupational surfaces are effectively lost in a buildup of microstratigraphic lenses (e.g., El Mirón), or multi-occupation surfaces sealed by fortuitous formation processes (e.g., La Garma, Pincevent, or Verberie) (Arias et al., 2011; Bailey, 2007; Leroi-Gourhan and Brézillon, 1972; Straus and González Morales, 2012a; Zubrow et al., 2010). As the former type dominates the Upper Paleolithic archaeological record, one of the biggest challenges facing Upper Paleolithic archaeologists is developing ways to dissect and understand palimpsest archaeological deposits so that they ⁎ Corresponding author at: EcoPlan Associates, Inc., 701 West Southern Avenue, Suite 203, Mesa, AZ 85210, USA. E-mail address: [email protected] (L.M. Fontes).

may more precisely investigate human behavioral adaptations to complex and varied Late Glacial landscapes. How archaeologists understand time in prehistory is based on the temporal scale at their disposal and the resolution that timescale has (Bailey, 2007). Upper Paleolithic research concerning human behavioral change generally focuses on relatively coarse stratigraphic units, (i.e., archaeological levels that are temporally assigned to particular cultural technocomplexes based on diagnostic artifact types, usually lithic or osseous industries or portable art items), which are then compared to each other. In some regions, archaeologists have been able to define sub-periods, (e.g., “Lower Magdalenian” and “Upper Magdalenian”), though the number and nature of these subdivisions varies in relation to local cultural histories, which inform regional cultural trajectories within broader archaeological periods (Straus, 2015). Recently, researchers have modeled behavioral change between the major Upper Paleolithic archaeological periods, particularly in relation to climatic and environmental factors (Cascalheira and Bicho, 2013; Schmidt et al., 2012). These studies have advanced archaeological understanding of the roles that climate change may have played in how humans reorganized their adaptive systems, however, these studies provide information about how behavioral change occurred between archaeological periods, rather than within them. Most Upper Paleolithic archaeologists would agree that what they recognize as the difference between “Solutrean” or “Magdalenian” was probably not the result of a sudden human behavioral change, but rather the culmination of

http://dx.doi.org/10.1016/j.jasrep.2017.03.015 2352-409X/© 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Fontes, L.M., et al., Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spai..., Journal of Archaeological Science: Reports (2017), http://dx.doi.org/10.1016/j.jasrep.2017.03.015

2

L.M. Fontes et al. / Journal of Archaeological Science: Reports xxx (2017) xxx–xxx

gradual shifts in human behavioral systems. Focusing on human behavioral change within archaeological periods can help archaeologists pinpoint the nature and timing of Upper Paleolithic cultural transitions. This paper presents a case study from El Mirón cave (Ramales de la Victoria, Cantabria, Spain), which contains an exceptionally well-dated and meticulously excavated Cantabrian Lower Magdalenian (CLM; c. 17,000–14,000 uncal. years BP) palimpsest that provides an ideal context to examine human behavioral change over the course of an archaeological period. This analysis subdivides Level 17 into 13 heuristic units that are then compared to assess how lithic raw material provisioning, and by proxy, landscape use strategies, developed and changed over the course of the Late Last Glacial. 2. The Lower Magdalenian at El Mirón cave The CLM period was first identified at Altamira cave (Cantabria), and later at other sites in modern day Cantabria and eastern Asturias provinces on the northern Atlantic Spanish coast, including El Rascaño, El Juyo, El Castillo, and El Mirón (Barandiarán et al., 1985; Freeman and González Echegaray, 2001; González Echegaray, 1960; González Echegaray and Barandiarán, 1981; Straus and González Morales, 2012b; Utrilla, 1981). This regional Lower Magdalenian archaeological “culture” is identified based on dense palimpsest deposits with unique artifact assemblages that include diagnostic lithic industries rich in bladelets and so-called nucleiform endscrapers (González Echegaray, 1960; see also Domingo et al., 2012 and Straus et al., 2016), antler points (sagaies) with tectiform engravings, and emblematic portable art—striation engraved red deer scapulae with depictions of red deer hinds (González Morales and Straus, 2009). The latter are also associated with similar engravings that were made on the ceilings and walls of several caves in the region, including Altamira and El Castillo (Alcalde del Río, 1906; Breuil and Obermaier, 1935). Archaeologists believe that CLM sites contain remains left by a distinct regional hunter-gatherer band that may have organized their settlements among residential bases and outlying, short-term (?), specialized camps in order to exploit the varied terrestrial and aquatic resources in the environmentally diverse, geographically circumscribed Vasco-Cantabrian region (Fontes et al., 2016; González Morales and Straus, 2009; Straus, 1986, 1992; Straus and González Morales, 2012b). Recent analyses of lithic raw material conveyance among four major CLM palimpsests (Altamira, El Juyo, El Mirón, and El Rascaño) indicate that the groups who occupied these sites shared an economic system, providing further evidence that the hunter-gatherers who occupied CLM sites were indeed a distinct regional group or related local bands (Fontes et al., 2016).

El Mirón cave is located in the Río Asón drainage of eastern Cantabria province, c. 255 m a.s.l. and 150 m above the valley floor at the confluence of the Ríos Calera and Gándara with the Asón, within the western cliff-face of Monte Pando in a foothill range of the Cantabrian Cordillera (Fig. 1; Straus and González Morales, 2012a; Straus et al., 2015a). The cave is surrounded by N1000 m peaks and is only some 20 km inland from the modern coastline, illustrating one of the major geographic features that characterizes the Vasco-Cantabrian region: substantial elevation change within short distances (Straus, 1992). The archaeological site has been the subject of continuous investigation led by L.G. Straus and M.R. González Morales since 1996, and is known for its Lower Magdalenian occupations, which include the human burial discovered in 2010 (Straus et al., 2015a). The cave vestibule is capacious, and Straus and González Morales have conducted excavations in several areas, including the outer vestibule “Cabin”, mid-vestibule trench, rear vestibule “Corral”, and the small area behind an engraved block, where the burial was unearthed (Fig. 2). The CLM deposit in El Mirón cave has been identified in each excavated area in the cave, indicating that these occupations were intensive and spanned the length of the cave vestibule. In the outer vestibule the CLM Level 17, which is the subject of this analysis, is a thick, “chocolate” brown, highly organic layer with no sterile breaks in the palimpsest (Fig. 3; Straus and González Morales, 2012b). The deposit includes: massive quantities of highly fragmented faunal remains, especially red deer and ibex; abundant lithic and osseous (sagaies, needles, awls, etc.) artifacts, including evidence of in situ manufacture; fire-cracked rocks and hearths; charcoal and ochre lenses; perforated ungulate teeth and shells; and extraordinary portable art objects, including an engraved red deer scapulae (González Morales and Straus, 2009; Nakazawa et al., 2009; Straus and González Morales, 2012b; Straus et al., 2015b). These archaeological remains indicate that CLM hunter-gatherers' undertook a diverse array of activities while they occupied El Mirón cave, including: animal butchery, tool manufacture, marrow and grease extraction, cooking, sewing (for making clothes, moccasins, etc.), producing ornaments and art objects, and a ritualized burial (Straus et al., 2015b). This deposit and its evidence for rich and varied residential activities mirrors similar horizons at other CLM hub sites in the region, including Altamira, El Castillo, El Juyo (all in Cantabria), and Santimamiñe (Vizcaya) (Straus et al., 2015b). Level 17 is dated by five radiocarbon assays to: −15,370 ± 80 uncal. BP (GX-32654); −15,450 ± 160 uncal. BP (GX-27115); −15,470 ± 240 uncal. BP (GX-24466);

Fig. 1. The location of El Mirón cave and geographically known flint outcrops within the Vasco-Cantabrian region of north coastal Spain. El Mirón is indicated by a red star. Geographically known lithic source outcrops are indicated by numbered purple squares, as follows: (1) Llaranza, (2) Ojo Guareña, (3) Sonabia, (4) Barrika, (5) Treviño, (6) Urbasa, (7) Bidache, and (8) Chalosse. Locations are approximate. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Please cite this article as: Fontes, L.M., et al., Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spai..., Journal of Archaeological Science: Reports (2017), http://dx.doi.org/10.1016/j.jasrep.2017.03.015

L.M. Fontes et al. / Journal of Archaeological Science: Reports xxx (2017) xxx–xxx

3

Fig. 2. El Mirón cave and excavation plan view. Drafted by E. Torres and modified by R.L. Stauber.

−15,610 ± 90 uncal. BP (OxA-22093); and − 15,700 ± 190 uncal. BP (GX-25853) (see Straus and González Morales, 2003, 2007; Straus et al., 2015b). These dates (chronologically presented above) testify that the Level 17 horizon in El Mirón cave was quickly formed c. 15,500 uncal. years BP following successive, intensive occupations by CLM hunter-gatherers.

3. Methods 3.1. Level 17 sublevels Level 17 ranges from 21 to 38 cm thick (based on changes in depth on the nine measures from centers of squares; Fig. 4), with its thickest portions in its northern section, furthest from the cave wall. The stratigraphy (three profile sections) testifies to a horizontal, flat deposition process in this area of the cave (Fig. 3). The level was excavated using microstratigraphic techniques. First, the area was divided in to meter squares with 50 × 50 cm sub-square units (Fig. 4). Excavators would expose surfaces (see Fig. 5), termed “spits” (arbitrary excavation units), within multiple squares and map large artifacts (bone fragments N3 cm, lithics N1 cm, shell and osseous artifacts, manuports, firecracked rocks, charcoal and ochre samples, among other artifact categories) in situ using an EDM that measured to the nearest millimeter. After artifacts were recorded, excavators would take closing depths for the spit in nine locations per square (Fig. 4). They would then continue excavation, once again opening new spits for several squares simultaneously across the excavated area. Base depths were used as top depths for subsequent spits. These excavation techniques, together with fortuitous formation processes, allow the squares within the Level 17 excavated area to be spatially correlated to each other, forming heuristic sublevel units that can be used to assess hunter-gatherer behavioral change through time within the level. Level 17 sublevels were created by matching the base depth measurements that were recorded during the excavation among the ten squares that comprise the 9.5 square meter area (see Fig. 4). The matching process yielded a total of 13 sublevels. Depths between squares were matched as closely as possible by comparing measures from successive spits to each other. For example, the east coordinate

for square H2 spit 50, 15.490 m, best correlates to square I2 spit 33′s west coordinate, 15.496 m, a difference of six millimeters. This match is better for the H2 spit 50 east coordinate than the west coordinates for I2 spit 32 (15.514 m, a 2.4 cm difference) and I2 spit 34 (15.439 m, a 5.1 cm difference). Thus, both H2 spit 50 and I2 spit 33 correlate to the same sublevel based on absolute depths. This example illustrates the matching process based on a single coordinate; this procedure would be repeated 60 more times to coordinate every depth measure as closely as possible across the 9.5 square meter area (see Fig. 4). This process is not without its limitations. Some excavated spits were thicker or thinner than others based on the density of archaeological finds. For example, if a spit in one square was 2 cm thick, and spits in an adjacent square were 0.5 cm thick, four spits in the latter square would correlate to a single spit in the former square. Correlations were guided by the thickest spit in these cases in order to create sublevels that included all ten of the excavated squares in the El Mirón outer vestibule area. This process also resulted in sublevels of varying thicknesses, but averaging 2.5 cm (Table 1). A second challenge was the absolute thickness of Level 17 itself: the deposit is thicker in its northern and western portions than in its southeastern portion, which is adjacent to a bulge in the cave wall. The sublevels divide the deposit by absolute, level depths rather than by assuming a sloping deposition. Thus, two sublevels at the base of Level 17 bring together only the six northernmost excavation squares, which have the greatest density of material. Finally, it's important to note that the depth measures that were correlated in this process were subject to human error that is inherent in the excavation process. For example, recording of depth measures was subject to the transit operator's and excavators' skill (changing year-to-year in a nearly two decades long excavation process) and the exact positions measured—slightly to the left or right of the previous depth point, on top of a manuport or bone, etc. While these are setbacks, using absolute depth data to create sublevels has provided heuristic units that can be used to explore relative spatiotemporal human behavioral change during the CLM occupations of El Mirón cave. 3.2. Lithic raw materials analysis Lithic raw materials were identified for a total of 33,718 lithic artifacts from Level 17. Two different reference collections were used to

Please cite this article as: Fontes, L.M., et al., Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spai..., Journal of Archaeological Science: Reports (2017), http://dx.doi.org/10.1016/j.jasrep.2017.03.015

4

L.M. Fontes et al. / Journal of Archaeological Science: Reports xxx (2017) xxx–xxx

Fig. 3. Stratigraphic section of the outer vestibule excavation in El Mirón cave, north wall. Drawn by L.G. Straus and redrafted by R.L. Stauber.

classify lithic raw materials from El Mirón. The first reference assemblage was created and used by Straus and colleagues in lithic analyses conducted at the end of excavation seasons each year. Straus created an ad hoc lithic raw material reference collection with materials recovered from the site, which distinguished most of the flints, lutites/ mudstones, quartzites, and other stones commonly found at the site. However, this collection was not exhaustive, leaving some uncommon materials characterized as “unknown flints” or “unknown stones”. These categories were not further classified as individual types, nor

were materials in these categories added to the reference collection as they were recovered. Artifacts from four of the Level 17 sublevels (units 1, 2, 6, and 10) were classified using this system. Fontes developed the second raw material reference collection as part of her dissertation research. During two years of laboratory analyses, Fontes analyzed lithic artifacts from 11 archaeological sites located in the Vasco-Cantabrian region and identified 241 visually distinct lithic raw materials in these collections (see Fontes, 2016). Ad hoc raw material reference collections were made for each archaeological collection

Please cite this article as: Fontes, L.M., et al., Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spai..., Journal of Archaeological Science: Reports (2017), http://dx.doi.org/10.1016/j.jasrep.2017.03.015

L.M. Fontes et al. / Journal of Archaeological Science: Reports xxx (2017) xxx–xxx

Fig. 4. The Level 17 correlations procedure. Excavation squares are indicated by solid grey lines, 50 cm × 50 cm subsquares are delineated by dotted grey lines and denoted “a” through “d” within each square, as labeled in square J2. Squares are labeled following their designations on the site's grid (see Fig. 2). Circles indicate depth points taken in each square during excavation before new spits were declared (NW, N, NE, W, Center, E, SW, S, and SE). Solid black circles indicate depth measures compared in each square/spit to make spatial correlations and create sublevel units. Each of these points was compared to its “pair”, i.e., N depth points in square H3 were compared with S depth points taken in square H4. White-filled circles represent depth points that were not used to make correlations, but which were averaged to determine the average depth change in each sublevel unit. White-filled circles with dotted outlines were not used to make spatial correlations.

analyzed, then later compared to each other and to archaeopetrographic reference collections made by archaeopetrologist A. Tarriño and by archaeologist J. Rissetto. Through this procedure, Fontes nearly doubled the size of the El Mirón raw material reference assemblage, adding every new toolstone she identified to the collection. Many of the materials that she added were samples of geographically known flints (Ojo

5

Guareña; Urbasa; several additional variants of Chalosse, Llaranza, Sonabia, and Treviño); and numerous flints [characterized as “other flints” without geographically known source outcrops] that were not previously recorded. Thus, these kinds of raw materials are likely underreported in Straus' analyses. Fontes used this raw material reference system to analyze nine of the Level 17 sublevels. Vasco-Cantabrian lithology is heterogeneous; flint toolstones can be visually distinguished easily. Thus, Fontes did not perform geochemical analyses. Detailed descriptions of all of the lithic raw materials analyzed here can be found in Fontes (2016). The geographically known outcrops of Bidache, Chalosse, Treviño and Urbasa (190 km, 240 km, 80 km, and 112 linear km from El Mirón, respectively)—all visually distinct flints that are considered “tracers” of prehistoric hunter-gatherer interactions—are discussed in detail in Chalard et al. (2010) and Tarriño et al. (2014). Further information about Cantabrian regional flints Barrika, Llaranza, Ojo Guareña, and Sonabia (40 km, 32 km, 19 km, and 20 linear km from El Mirón, respectively) can be found in Rissetto (2009), Sarabia (1990a, 1990b), Sarabia and Pedro, 1999, Sarabia and Pedro, 2002), and Tarriño et al. (2014). Locations of these outcrops relative to El Mirón cave are shown in Fig. 1. Finally, it is important to note that this analysis is based on researchers' current understanding of flint outcrop locations, however, toolstone outcrops have likely been influenced by geomorphological changes that have occurred since the Last Glacial period. Lower Magdalenian hunter-gatherers may have had access to lithic raw materials that are now submerged due to sea level rise, buried by alluvium or colluvium, or obscured by vegetation. 4. Results 4.1. El Mirón Level 17 The assemblages from Level 17 testify that CLM hunter-gatherers increased the quantity of regional flints—Barrika, Llaranza, and Sonabia—in their lithic toolkits through time (Table 1; Fig. 6). Concurrently, quartzites and mudstones, presumably from local outcrops, decreased in abundance. The quantity of limestones and other flints (18– 22 additional toolstones without geographically known outcrop locations) in CLM assemblages from El Mirón remains consistent throughout the deposit. Extra-regional flints from the Chalosse, Bidache, Treviño, and Urbasa outcrops comprise a consistently small portion of El Mirón Level 17 assemblages (b5%), however, all of these materials increase in abundance in more recent portions of the deposit—especially Urbasa and Chalosse (Fig. 6).

Fig. 5. A sample Level 17 excavation surface recorded during the 2005 field season. Photo credit L.G. Straus.

Please cite this article as: Fontes, L.M., et al., Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spai..., Journal of Archaeological Science: Reports (2017), http://dx.doi.org/10.1016/j.jasrep.2017.03.015

6

L.M. Fontes et al. / Journal of Archaeological Science: Reports xxx (2017) xxx–xxx

Table 1 Raw material weights and assemblage portions, by toolstone type. Samples marked with an “a” were analyzed by Straus and colleagues; Fontes analyzed all others. Weights were measured in grams. Nine center point measurements (see Fig. 4) were averaged for each sublevel to determine the average sublevel thickness. Other flints include all flints Fontes identified in the El Mirón assemblage with unknown geographic locations. Other stones are rare raw materials, including quartzitic sandstone, ochres, amber, and a (possibly) igneous rock that Fontes identified. Unknown flints and stones are two categories Straus and colleagues used in their analyses; the contents of these classifications would likely fall into Fontes' “other stone” and “other flints” categories.

Sublevel thickness (cm)

Barrika Bidache Chalosse Other flints Limestone Llaranza Mudstone Other Stone Ojo Guareña Quartz & calcite Quartzite Sonabia Treviño Urbasa Unknown flint Unknown stone Total

Thickness (cm)

Barrika Bidache Chalosse Other flints Limestone Llaranza Mudstone Other stone Ojo Guareña Quartz & calcite Quartzite Sonabia Treviño Urbasa Unknown flint Unknown stone Total

Sublevel 13

Sublevel 12

Sublevel 11

Sublevel 10a

Sublevel 9

Sublevel 8

4.3

2.9

0.7

1.5

1.3

2.3

Sublevel 7 1.9

wt.

%

wt.

%

wt.

%

wt.

%

wt.

%

wt.

%

wt.

%

882.8 20.9 120.7 306.1 174.7 590.8 1835.6 151 18.7 60.4 1156.6 934 153.9 123.1 0 0 6529.3

13.5 0.3 1.8 4.7 2.7 9.0 28.1 2.3 0.3 0.9 17.7 14.3 2.4 1.9 0.0 0.0

689.8 10.4 85.1 261.9 51.7 444 1293.5 188.2 35.1 33.4 1773.1 489.4 58.7 100.2 0 0 5514.5

12.5 0.2 1.5 4.7 0.9 8.1 23.5 3.4 0.6 0.6 32.2 8.9 1.1 1.8 0.0 0.0

529.6 24.8 61.6 207.8 161.1 411.2 965.2 806.6 18.8 27.7 1150.7 603.1 42.5 59.5 0 0 5070.2

10.4 0.5 1.2 4.1 3.2 8.1 19.0 15.9 0.4 0.5 22.7 11.9 0.8 1.2 0.0 0.0

697.2 15.9 6.7 227.5 288.9 307 239.5 222.5 0 40.1 592.5 254.9 0.1 0 546.8 653.7 4093.3

17.0 0.4 0.2 5.6 7.1 7.5 5.9 5.4 0.0 1.0 14.5 6.2 b0.01 0.0 13.4 16.0

654.9 28.6 94.3 338.5 66 480.3 1923.5 129.7 25 47.9 1082 781.6 78.1 102.1 0 0 5832.5

11.2 0.5 1.6 5.8 1.1 8.2 33.0 2.2 0.4 0.8 18.6 13.4 1.3 1.8 0.0 0.0

1002.7 54.8 88.4 300.9 31.9 648.8 1715 210.3 29.3 164.5 1288.3 729.9 76.5 115 0 0 6456.3

15.5 0.8 1.4 4.7 0.5 10.0 26.6 3.3 0.5 2.5 20.0 11.3 1.2 1.8 0.0 0.0

750 15.3 51.8 252.7 35.8 481.3 399.9 18.4 5.7 38 544.4 555.5 28.8 68.5 0 0 3246.1

23.1 0.5 1.6 7.8 1.1 14.8 12.3 0.6 0.2 1.2 16.8 17.1 0.9 2.1 0.0 0.0

Sublevel 6a

Sublevel 5

Sublevel 4

Sublevel 3

Sublevel 2a

Sublevel 1a

1.4

2.6

4

3.4

3.4

3.3

Level 17 total 33

wt.

%

wt.

%

wt.

%

wt.

%

wt.

%

wt.

%

wt.

%

880.6 63.8 0.5 312.5 87.2 65.9 81.3 9.1 0 4.9 192.5 158.6 2.1 0 599.4 432.2 2890.6

30.5 2.2 0.0 10.8 3.0 2.3 2.8 0.3 0.0 0.2 6.7 5.5 0.1 0.0 20.7 15.0

636.8 49.1 66.9 194.8 66.9 349.6 40.9 14.5 8.5 10.5 184.6 564.3 48.1 85.2 0 0 2320.7

27.4 2.1 2.9 8.4 2.9 15.1 1.8 0.6 0.4 0.5 8.0 24.3 2.1 3.7 0.0 0.0

827.1 54.2 120 244.2 132.9 752.5 330.2 9.1 14.3 79.3 90.6 762.1 53 79.1 0 0 3548.6

23.3 1.5 3.4 6.9 3.7 21.2 9.3 0.3 0.4 2.2 2.6 21.5 1.5 2.2 0.0 0.0

1121.7 36.2 190.5 402.4 0.1 874.9 151.6 2.8 9.2 18.7 106.6 993.4 66.3 174.5 0 0 4148.9

27.0 0.9 4.6 9.7 0.0 21.1 3.7 0.1 0.2 0.5 2.6 23.9 1.6 4.2 0.0 0.0

737.3 9.1 0 693.8 50.3 73.2 64.6 0 0 31.6 37.4 815.6 0 0 1438.5 4.4 3955.8

18.6 0.2 0.0 17.5 1.3 1.9 1.6 0.0 0.0 0.8 0.9 20.6 0.0 0.0 36.4 0.1

1245.6 31.1 0 478.3 41.5 89.4 128.2 0 0 127.9 19.6 209.4 1.5 0 1731.9 46.4 4150.8

30.0 0.7 0.0 11.5 1.0 2.2 3.1 0.0 0.0 3.1 0.5 5.0 0.0 0.0 41.7 1.1

10,656.1 414.2 886.5 4221.4 1189 5568.9 9169 1762.2 164.6 684.9 8218.9 7851.8 609.6 907.2 4316.6 1136.7 57,757.6

18.4 0.7 1.5 7.3 2.1 9.6 15.9 3.1 0.3 1.2 14.2 13.6 1.1 1.6 7.5 2.0

The increase in regional and extra-regional flint quantities and corresponding decrease in local quartzites and mudstones indicates that the CLM hunter-gatherers who occupied El Mirón cave changed their lithic provisioning strategies over time. The groups gradually shifted from a bipartite low-quality (quartzite and mudstone) and high-quality (regional flint) lithic economy to a system focused on procuring high quality flints from regional and exotic outcrops. There are several possible explanations for this economic change: − First, CLM groups may have chosen to optimize, obtaining more flexible flint toolstones because they were maintainable, requiring less time to be spent procuring lithic raw materials. Additionally, flints are more productive toolstones than quartzites and mudstones, especially for bladelet reduction, a major focus of CLM toolkits (see Fontes, 2016). CLM hunter-gatherers may have procured raw materials with greater production potential and use lives to reduce risk in a landscape with patchily distributed resources (Freeman 1973; Straus, 1987); − Second, CLM local territories may have changed due to environmental or social factors. For example, regional environmental changes may have altered comestible resource distributions, causing groups to change their foraging patterns and consequently altering their

procurement strategies (assuming that groups embedded acquiring lithic materials with other mobile activities [Binford, 1980]). It is equally possible that CLM hunter-gatherer populations increased and/or group compositions changed, resulting in different group territories that may have influenced what lithic outcrops El Mirón's occupants may have had access to; − Third, social networks may have developed and played a more prevalent role in CLM hunter-gatherer economies. The increase in extraregional flints is interesting in this regard, particularly because these materials are most abundant in the most recent CLM Level 17 deposits, which are temporally closest to the Middle Magdalenian period—known for its far-ranging social contacts that are evidenced by the distribution of Pyrenean portable art items (Straus, 2013). Raw materials identified in the entire Level 17 sequence indicate a connection (albeit a small one) with France—presumably with groups of people, not just lithic outcrops. What is unique and important about the evidence from the Level 17 sublevels is that the presence of these toolstones gradually increases, perhaps representing contacts that gradually intensified before the Middle Magdalenian artistic “explosion”. If CLM groups were expanding their social networks, they may have received greater access to exotic flints through trade and/or exchange relationships. At the same time,

Please cite this article as: Fontes, L.M., et al., Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spai..., Journal of Archaeological Science: Reports (2017), http://dx.doi.org/10.1016/j.jasrep.2017.03.015

L.M. Fontes et al. / Journal of Archaeological Science: Reports xxx (2017) xxx–xxx

7

Fig. 6. Percents of lithic raw materials in El Mirón Level 17 Sublevels. Raw materials are divided into three groups. Top: Cantabrian regional flints Barrika, Llaranza, and Sonabia. Middle: Extra-regional flints Treviño, Urbasa, Chalosse, and Bidache. Bottom: presumably local raw materials, including mudstone, quartzite, limestone, and other flints (without geographic source information). Only samples analyzed by Fontes are shown here.

there is ample evidence that CLM groups had an important local culture, symbolized in unique portable art items like engraved scapulae (see González Morales and Straus, 2009). Hunter-gatherer groups who inhabited El Mirón cave may have balanced local traditions and “global” interactions within their “Magdalenian” world. None of these explanations is exclusive of the others. CLM groups may have desired the flexibility and maintainability that flint dependence would have offered their economic systems, not simply as a means to optimize their foraging, but to adapt to a changing social and environmental landscape. Flexibility could help groups: optimize their economies; adjust to local territorial disputes or migrating terrestrial resources; and adapt to the rigors of an ever-expanding “Magdalenian” social system (Otte, 2012). 4.2. El Mirón Level 17 in relation to Level 504 Lower Magdalenian deposits were recovered from several locations in El Mirón cave. This section situates the results from Level 17 in context with those from Level 504, a similarly rich deposit from the rear vestibule of the cave that was associated with a Lower Magdalenian

age primary human burial (Fontes et al., 2015; Straus et al., 2015a). Level 504 has been dated by three radiocarbon assays that yielded CLM dates that are penecontemporaneous with those from Level 17: 15,460 ± 40 uncal. BP (MAMS-14585) and 15,740 ± 40 uncal. BP (UG-7217) (Straus et al., 2015a; Straus et al., 2015b). Among geographically known flints, Barrika (41.8%) and Llaranza (26.6%) are the most abundant lithic raw materials in the Level 504 assemblage, followed by Sonabia (3.2%) and trace amounts of Treviño (0.9%) and Urbasa (b 0.01%). Chalosse and Bidache are absent from the collection (Fontes et al., 2016). Quartzites, mudstones, quartzes, calcites, and limestones are present in small portions in the assemblage (Fontes et al., 2015). The Level 504 assemblage has twice the portion of Barrika than the average quantity in the Level 17 sublevels (20%; the maximum portion is 30.5% in Sublevel 6), and less Sonabia toolstone than the Level 17 sample. The portion of Llaranza in Level 504 is similar to that in the more recently formed deposits in Level 17, however, the proportions of extra-regional flints are most similar to samples from the earliest CLM sublevels. These comparisons indicate spatial variation in the CLM deposits in El Mirón cave. It's possible that spatial variances are the result of occupations: with specific functions (i.e., bladelet production); made by different CLM bands; made at different moments within the structure of the overall settlement system, representing different

Please cite this article as: Fontes, L.M., et al., Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spai..., Journal of Archaeological Science: Reports (2017), http://dx.doi.org/10.1016/j.jasrep.2017.03.015

8

L.M. Fontes et al. / Journal of Archaeological Science: Reports xxx (2017) xxx–xxx

moments of a lithic procurement schedule; or, that were spatially delimited into activity areas where some raw materials were elected over others due to their preferred knapping qualities. The lithic raw materials from these two deposits within El Mirón testify that CLM occupations in caves were dynamic settlements similar to those observed at open air sites (see Zubrow et al., 2010). 4.3. Lower vs. Initial Magdalenian lithic provisioning at El Mirón El Mirón also contains an Initial Magdalenian deposit, Levels 119.3– 117 in the “Corral” area of the rear vestibule, that can be compared to that from Level 17. The Initial Magdalenian assemblages are similar in character to those from CLM Levels 504 and 17: all have enormous quantities of knapping debris, including cores, flakes, bladelets, and microdebitage that testify to in situ lithic manufacture in the cave (Fontes et al., 2015; Straus et al., 2014, 2016). The Initial Magdalenian assemblages are bipartite: an “archaic” toolkit with abundant notches, denticulates, and sidescrapers was made on non-flint raw materials—quartzite, limestone, and mudstone; the second toolkit component was bladelet-based, and manufactured from Barrika, Sonabia, and Llaranza, among other flints (Straus et al., 2014). Initial Magdalenian groups undoubtedly separated these lithic toolstone classes based on their performance characteristics (Straus et al., 2014). The earliest deposits in Level 17 testify to the same bipartite raw material split, evidence that CLM hunter-gatherers continued the same lithic provisioning strategies used by the cave's Initial Magdalenian occupants. Based on the Level 17 record, the shift from bipartite to flintbased lithic toolstone provisioning strategies was a gradual one. This suggests that local cultural traditions (i.e., economic habits) influenced how new behavioral systems developed during the Magdalenian period—CLM groups exploited the same lithic outcrops as their predecessors, progressively shifting the quantities they collected. 4.4. Lithic provisioning in El Mirón Level 17 in relation to other CLM palimpsests Fontes et al. (2016) have summarized lithic raw material provisioning based on geographically known flints at three additional CLM sites: Altamira, El Juyo, and El Rascaño, located approximately 60 km, 40 km, and 20 km west of El Mirón, respectively. Among the known flint sources, those that outcrop in the Cantabrian region are most variable in their quantity within CLM lithic assemblages. Barrika toolstone is only 4.5% of the Altamira lithic assemblage, 11.8% at El Juyo and 7.3% at El Rascaño (Fontes et al., 2016). In contrast, it is 18.4% of the entire Level 17 assemblage, ranging from 10.4% to 30.5% in individual sublevels (Table 1). Llaranza flint is similarly variable: 18.9% of the Altamira assemblage, 2.8% in El Juyo, 7.7% at El Rascaño, and 9.6% in the Level 17 collection (range: 1.9% to 21.2% in individual sublevels). Finally, Sonabia toolstone makes up 13.6% of the Level 17 assemblage (with individual sublevels ranging from 5.5% to 24.3%), 9% at Altamira, 37.4% in El Juyo, and 41.3% at El Rascaño (Fontes et al., 2016). These variable toolstone quantities cannot simply be explained by site-outcrop proximity. El Juyo is the site closest to Llaranza in this sample, yet it has the smallest portion of that toolstone among these four sites. El Mirón is closest to the Sonabia and Barrika outcrops (Fontes et al., 2016). While the El Mirón Level 17 assemblage overall has a greater quantity of Barrika than the samples from the other sites, Sonabia is less abundant here, and is instead a significant component of the lithic assemblages from El Juyo and El Rascaño. Fontes et al. (2016) have argued that these proportional differences likely related to how CLM hunter-gatherers used and conveyed lithic raw materials among these sites as part of a regional settlement system. Lithic toolstones from these three outcrops comprise the majority of these four CLM lithic assemblages, and perhaps reflect how local hunter-gatherer groups structured their mobility and procurement schedules in response to environmental and/or social constraints within the Vasco-Cantabrian coastal corridor.

In contrast, extra-regional flints testify to CLM hunter-gatherers' connections to a “Magdalenian” world beyond the Cantabrian Cordillera. Treviño flint occurs in trace amounts in the Altamira (0.6%), El Rascaño (3%), and El Mirón (1.1% overall; range: 0–2.4%) assemblages. However, it is a greater quantity of the El Juyo assemblage (8.6%). Urbasa flint is slightly more common: 3.8% in the Altamira assemblage, 5.3% at El Juyo, 3.4% at El Rascaño, and 1.6% overall in Level 17 (range 0– 4.2%). The Level 17 assemblage also contains more Chalosse and Bidache toolstones than the other sites, where these materials are present in trace amounts or absent (Fontes et al., 2016). As the easternmost of the CLM sites examined here, El Mirón is closer to the extra-regional outcrops by 20–60 km. However, the patterns observed here still cannot be explained by site proximity alone. That extra-regional lithic toolstones regularly occur in CLM assemblages (even in small amounts) indicates that coastal hunter-gatherer groups likely had an expansive social network that yielded lithic raw materials, which were then brought back to coastal residential sites and later circulated through the region in these foragers' mobile toolkits. The consistent, albeit small quantity of these toolstones throughout the Level 17 sublevel sequence testifies that these plausible, expansive social networks would have been relatively stable. Finally, the observed difference between highly variable regional flint quantities and relatively stable amounts of extra-regional flints at CLM sites likely reflects groups' adaptations to local environmental (both social and ecological) circumstances. Since regional flints were more proximal to CLM sites, they could have been procured through normal subsistence activities, which may have seasonally or annually fluctuated, resulting in differing lithic toolstone needs at different sites, depending on their position within the overall settlement system. In contrast, extra-regional flints were unlikely to have been procured as an immediate response to a local environmental constraint: their procurement costs (i.e., distance traveled/time) would likely have been too high. As minor components of CLM toolkits presumably procured through social relationships, these materials may have been obtained regardless of local adaptive conditions, which is why they are present in small, yet stable quantities—circulated and depleted through the course of regular hunter-gatherer movements. 5. The Lower Magdalenian “landscape” Based on lithic raw material conveyance evidence, Fontes et al. (2016) argued that CLM groups likely balanced several different mobility systems as they organized their lithic provisioning behavior. These interlocking scalar economies may have incorporated: site catchment zones; sub-regional territories that brought groups to different ecological zones; long term (seasonal/annual?) ranges where groups gradually shifted within the maximum territorial extent; and inter-territorial networks where groups met to exchange materials, perhaps along with partners and ideas (Fontes et al., 2016). The Level 17 sublevel unit assemblages provide further data about variation in several of these scales. Based on changing proportions of presumably locally outcropping mudstones and quartzites that the hunter-gatherers occupying El Mirón cave used, these groups shifted the sizes of their site catchment zones during the CLM. Such shifts could have been related to seasonal or annual resource variations that directed groups' foraging to different areas within broader ecological zones (e.g., coastal, montane), consequently influencing their local embedded procurement strategies and decreasing the quantity of local stones in their lithic toolkits because they were located outside of the daily foraging radius and costly to procure. Along a similar vein, these shifts could also correspond to increased local resource abundance and decreasing forager mobility—if the site catchment was resource-rich and groups could provision their toolkits through embedded flint procurement before arriving at El Mirón, logistical trips to local outcrops with medium- to low-quality toolstones may not have been necessary. Finally, demographic pressure could have limited groups' access to local toolstone sources (Straus et

Please cite this article as: Fontes, L.M., et al., Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spai..., Journal of Archaeological Science: Reports (2017), http://dx.doi.org/10.1016/j.jasrep.2017.03.015

L.M. Fontes et al. / Journal of Archaeological Science: Reports xxx (2017) xxx–xxx

al., 2000). CLM hunter-gatherers may have changed their lithic economic strategies to adapt to changing environmental and cultural landscapes. The portions of geographically known regional flint sources indicate that CLM groups who occupied El Mirón maintained sub-regional territories that included sites located in coastal and upland areas. These hunter-gatherers likely shifted environmental zones as they adapted to environmental complexity in this geographically circumscribed region. Portable art objects, particularly engraved scapulae with striated depictions of red deer hinds, also support the hypothesis that a CLM “culture” core area existed in the lowlands and highlands in modern Cantabria and eastern Asturias provinces (González Morales and Straus, 2009). Finally, the Level 17 sublevel lithic assemblages provide further evidence that CLM groups participated in inter-territorial networks, evidenced by exotic toolstones from southern France (Chalosse and Bidache) and the upper Ebro Basin of Álava and Navarra (Treviño and Urbasa). Lithic toolstone exchanges were likely only one component of these Magdalenian social networks. Personal ornaments from El Mirón Level 17 also signify long-distance connections: first, a Homalopoma sanguineum shell was recovered from Sublevel 4, linking El Mirón occupants to the Mediterranean, presumably via the Ebro Basin corridor (Straus and González Morales, 2012b); second, a reindeer tooth with lateral incisions recovered from Sublevel 7 (identified by J.M. Geiling), that is similar to ones identified at Roc de Marcamps 2 (Gironde) (Kuntz et al., 2015). An atlatl hook discovered in El Mirón Level 17 (Sublevel 12) provides further evidence of possible social contacts between CLM groups and French Early Magdalenian hunter-gatherers who resided at Roc de Marcamps and Le Placard, sites located approximately 350 km and 450 km NNE of El Mirón, respectively (Cattelain, 2004; González Morales and Straus, 2009). The atlatls from these three sites are nearly identical in form and size, suggesting an exchange of objects and/or ideas between French and Spanish Magdalenian groups occurring approximately 15,000 uncal. BP years ago (Straus and González Morales, 2012b). Lithic raw materials provide further evidence that social linkages were an important component of “Magdalenian” culture—CLM groups likely acquired exotic toolstones, along with ornaments, ideas, mates, etc., through long-distance social networks. The El Mirón Level 17 sublevel lithic assemblages also testify that movement of these materials intensified through time—perhaps reflecting increasingly frequent contacts between hunter-gatherers living in different Magdalenian settlement areas. Portions of exotic toolstones in assemblages increase in the most recent portion of the El Mirón Level 17 sublevel sequence—those deposits temporally closest to the beginning of the Middle Magdalenian. This timing is significant because the Middle Magdalenian is defined in the Vasco-Cantabrian region based on the presence of portable art from the French Pyrenees, including contours découpés (carved hyoid bone cut-outs) and rondelles (perforated and decorated flat bone discs) (Straus and Gonzáles Morales 2012b). In other words, inter-regional contacts are the hallmark of the Middle Magdalenian period, and the El Mirón Level 17 sublevel lithic assemblages indicate that contacts among groups within these networks gradually intensified, providing further evidence of in situ development of “Magdalenian” cultural traditions. 6. Conclusions This analysis demonstrates the importance of understanding palimpsest deposits with as fine-grained units as possible, not only as archaeologists examine previously recovered materials, but as they conduct new excavations and create recording systems for those contexts. The sublevel units used here, while heuristic, provide 13 points of comparison that can be used to examine hunter-gatherer behavioral change. These microstratigraphic units provide a window that archaeologists can use to reconstruct CLM human-landscape adaptations and to

9

explore technological change, toolkit structure and maintenance, and spatial variation within El Mirón cave. The lithic assemblages from El Mirón cave Level 17 testify that the “Magdalenian” world developed and changed through time, even within the archaeological periods that researchers use to define its various stages. Shifts in lithic provisioning have implications for how CLM groups adapted to local environments, reacted to changing social landscapes, and interacted within long-distance networks. Based on the archaeological record from El Mirón Level 17, the “Magdalenian” landscape was a dynamic one where hunter-gatherers balanced social, lithological, geographical, and environmental resources at several complex and overlapping scales.

Acknowledgements Fontes wishes to thank Roberto Ontañón, who helped facilitate transfer of the El Mirón lithic collection from the Museo de Prehistoria y Arqueología de Cantabria to the Instituto Internacional de Investigaciones Prehistóricas de Cantabria at the Universidad De Cantabria for her study. Fontes is grateful to Ana Belen Marín Arroyo for providing laboratory space within the IIIPC where she could conduct her lithic analyses. Fontes' research has been funded by: the National Science Foundation Doctoral Dissertation Improvement Grant #1318485; the American Association of University Women American Dissertation Fellowship; the UNM Latin American and Iberian Institute Ph.D. fellowship; numerous small grants from UNM; and the UNM Fund for Stone Age Research (J. and R. Auel, principal donors). Excavations at El Mirón cave were authorized by the Gobierno de Cantabria and have been funded by the National Science Foundation, National Geographic Society, Fundación M. Botín, Gobierno de Cantabria, Ministerio de Educación y Cultura, L.S.B. Leakey Foundation, Stone Age Research Fund, and the University of New Mexico. We wish to thank those who participated in the El Mirón excavations, whose careful work made this analysis possible. References Alcalde del Río, H., 1906. Las pinturas y grabados de las cavernas prehistóricas de la Provincia de Santander. Blanchard y Arce, Santander. Arias, P., Ontañon, R., Álvarez-Fernánez, E., Cueto, M., Elorza, M., García-Moncó, C., Güth, A., Iriarte, M., Teira, L., Zurro, D., 2011. Magdalenian floors in the lower gallery of La Garma: a preliminary report. In: Gaudzinski-Windheuser, S., Jöris, O., Sensburg, M., Street, M., Turner, E. (Eds.), Site-internal Spatial Organization of Hunter-gatherer Societies: Case Studies From the European Palaeolithic and Mesolithic. Verlag des Römisch-Germanischen Zentralmuseums. Bailey, G.N., 1983. Concepts of time in quaternary prehistory. Annu. Rev. Anthropol. 12, 165–192. Bailey, G.N., 2007. Time perspectives, palimpsests and the archaeology of time. J. Anthropol. Archaeol. 26, 198–223. Barandiarán, I., Freeman, L.G., González Echegaray, J., Klein, R.G. (Eds.), 1985. Excavaciones en la cueva del Juyo. 14. Centro de Investigacion y Museo de Altamira Monografias. Binford, L.R., 1980. Willow smoke and dogs' tails: hunter-gatherer settlement systems and archaeological site formation. Am. Antiq. 45, 4–20. Breuil, H., Obermaier, H., 1935. The Cave of Altamira at Santillana del Mar, Spain. Junta de las Cuevas de Altamira, Hispanic Society of America, and Academic de la Historia Tipográfica de Archivos, Madrid. Cascalheira, J., Bicho, N., 2013. Hunter-gatherer ecodynamics and the impact of the Heinrich event 2 in central and southern Portugal. Quat. Int. 318, 117–127. Cattelain, P., 2004. Un propulseur de la Grotte du Placard (Vilhonneur, Charente, France). 24. Notae Praehistoricae, pp. 61–67. Chalard, P., Ducasse, S., Bon, F., Bruxelles, L., Teyssandier, N., Renard, C., Gardere, P., Guillermin, P., Lacombe, S., Langlais, M., Mensan, R., Normand, C., Simmonet, R., Tarriño, A., 2010. Chalosse-type flint: exploitation and distribution of a lithologic tracer during the Upper Paleolithic, Southern France. In: Brewer-LaPorta, M., Burke, A., Field, D. (Eds.), Ancient Mines and Quarries: A Trans-Atlantic Perspective, pp. 13–22. Domingo, R., Mazo, C., Utrilla, P., 2012. Hunting camps and nucleiform endscrapers in the Cantabrian Lower Magdalenian: a lithic microwear analysis. Quat. Int. 272, 105–110. Foley, R.A., 1981. A model of regional archaeological structure. Proc. Prehist. Soc 47, 1–17. Fontes, Lisa M., 2016. Economies Set in Stone? Magdalenian Lithic Technological Organization and Adaptation in Vasco-Cantabrian Spain. (Doctoral Dissertation). Department of Anthropology, University of New Mexico. Fontes, Lisa M., Straus, Lawrence Guy, González Morales, Manuel R., 2015. Lithic and osseous artifacts from the Lower Magdalenian human burial deposit in El Mirón Cave, Cantabria, Spain. J. Archaeol. Sci. 60, 99–111.

Please cite this article as: Fontes, L.M., et al., Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spai..., Journal of Archaeological Science: Reports (2017), http://dx.doi.org/10.1016/j.jasrep.2017.03.015

10

L.M. Fontes et al. / Journal of Archaeological Science: Reports xxx (2017) xxx–xxx

Fontes, Lisa M., Straus, Lawrence Guy, González Morales, Manuel R., 2016. Lithic raw material conveyance and hunter-gatherer mobility during the Lower Magdalenian period in Cantabria, Spain. Quat. Int. 412A, 66–81. Freeman, L.G., González Echegaray, J., 2001. La Grotte d'Altamira. La Maison des Roches, Paris. González Echegaray, J., 1960. El Magdaleniense III en la Costa Cantábrica. Boletín del Seminario de Arte y Arqueología 26, pp. 1–32. González Echegaray, J., Barandiarán, I., 1981. El Paleolítico Superior de la Cueva del Rascaño. Monografías del Centro de Investigación y Museo de Altamira 3, Santander. González Morales, M.R., Straus, L.G., 2009. Extraordinary early Magdalenian finds from El Mirón Cave, Cantabria. Antiquity 83, 267–281. Kuntz, D., Sécher, A., Costamango, S., Mallye, J., Pétillon, J., Peschaux, C., Pubert, E., Rendu, W., Boudadi-Maligne, M., Laroulandie, V., Barshay-Szmidt, C., Langlais, M., 2015. Le Roc de Marcamps 2 (Prignac-et-Marcamps, Gironde): Nouvelles données sur la subsistence et les traditions tecniques au début du Magdalénien moyen. Bulletin de la Société Préhistorique Française 112-3, pp. 475–516. Leroi-Gourhan, A., Brézillon, M., 1972. Fouilles de Pincevent: Essai d'analyse ethnographique d'un habitat magdalénien (La Section 36). VIIé Supplément Gallia Préhistoire. Editions CNRS, Paris. Nakazawa, Y., Straus, L.G., González Morales, M., Cuenca, D., Caro, J., 2009. On stone-boiling technology in the Upper Paleolithic: behavioral implications from an early Magdalenian hearth in El Mirón, Cantabria, Spain. J. Archaeol. Sci. 36, 684–693. Otte, M., 2012. Appearance, expansion, and dilution of the Magdalenian civilization. Quat. Int. 272-273, 354–361. Rissetto, J.D., 2009. Late Pleistocene Hunter-gatherer Mobility Patterns and Lithic Exploitation in Eastern Cantabria (Spain). (Ph.D. Thesis). University of New Mexico, New Mexico. Sarabia, P.M., 1990a. Approche de l'etude de la distribution stratigraphique du silex de la province de Cantabria (Espagne du Nord). In: Séronie-Vivien, M.R., Lenoir, M. (Eds.), Le silex de sa genese à l'outil, Cahiers du Quaternaire 17, pp. 141–148. Sarabia, P.M., 1990b. L'utilisation du silex dans les industries du Paléolithique de Cantabria (Espagne du Nord). In: Séronie-Vivien, M.R., Lenoir, M. (Eds.), Le silex de sa genese à l'outil, Cahiers du Quaternaire 17, pp. 443–448. Sarabia, Rogina, Pedro, M., 1999. Notas sobre los modelos de aprovisionamiento de materias primas líticas en el Paleolítico Superior de Cueva Morín (Villanueva de Villaescusa, Cantabria). In: Sautuola, V.I. (Ed.), Revista del Instituto de Prehistoria y Arqueología Sautuola. Consejería de Cultura Turismo y Deporte del Gobierno de Cantabria, Santander, pp. 145–154. Sarabia, Rogina, Pedro, M., 2002. La Utilización de Rocas del Grupo Radiolarita en las Industrias Paleolíticas de Cantabria. In: Sautuola, V.I.I.I. (Ed.), Revista del Instituto de Prehistoria y Arqueología Sautuola. Consejería de Cultura Turismo y Deporte del Gobierno de Cantabria, Santander, pp. 11–25. Schmidt, Isabell, Bradtmöller, Marcel, Kehl, Martin, Pastoors, Andreas, Tafelmaier, Yvonne, Weninger, Bernard, Weniger, Gerd-Cristian, 2012. Rapid climate change and

variability of settlement patterns in Iberia during the Late Pleistocene. Quat. Int. 274, 179–204. Straus, L.G., 1979. Caves: a paleoanthropological resource. World Archaeol. 10, 331–339. Straus, L.G., 1986. Late Würm adaptive systems in Cantabrian Spain. J. Anthropol. Archaeol. 5, 330–368. Straus, L.G., 1987. Upper Paleolithic ibex hunting in SW Europe. J. Archaeol. Sci. 14, 163–178. Straus, L.G., 1992. Iberia Before the Iberians. University of New Mexico Press, Albuquerque. Straus, L.G., 2013. After the deep freeze: confronting “Magdalenian” realities in Cantabrian Spain and beyond. J. Archaeol. Method Theory 20, 236–255. Straus, L.G., 2015. Recent developments in the study of the Upper Paleolithic of VascoCantabria. Quat. Int. 364, 255–271. Straus, L.G., González Morales, M.R., 2003. El Mirón Cave and the 14cchronology of Cantabrian Spain. Radiocarbon 45, 41–58. Straus, L.G., González Morales, M.R., 2007. Further radiocarbon dates for the Upper Paleolithic of El Mirón Cave (Ramales de la Victoria, Cantabria, Spain). Radiocarbon 49, 1205–1214. Straus, L.G., González Morales, M.R., 2012a. El Mirón Cave, Cantabrian Spain: The Site and Its Holocene Archaeological Record. University of New Mexico Press, Albuquerque. Straus, L.G., González Morales, M.R., 2012b. The Magdalenian settlement of the Cantabrian region (northern Spain): the view from El Miron Cave. Quat. Int. 272273, 111–124. Straus, L.G., Bicho, N., Winegardner, A.C., 2000. The Upper Paleolithic settlement of Iberia: first-generation maps. Antiquity 74, 553–566. Straus, L.G., González Morales, M.R., Fontes, L.M., 2014. Initial Magdalenian artifact assemblages in El Mirón Cave (Ramales de la Victoria, Cantabria, Spain): a preliminary report. Zephyrus 73, 45–65. Straus, L.G., González Morales, M.R., Carretero, J.M. (Eds.), 2015a. The red lady of El Mirón Cave: Lower Magdalenian human burial in Cantabrian SpainJ. Archaeol. Sci. 60. Straus, Lawrence Guy, González Morales, M.R., Higham, Thomas, Richards, Michael, Talamo, Sahra, 2015b. Radiocarbon dating the Late Upper Paleolithic of Cantabrian Spain: El Mirón Cave date list IV. Radiocarbon 57 (1), 183–188. Straus, Lawrence Guy, Fontes, Lisa M., Domingo, Rafael, González Morales, Manuel R., 2016. Cores, core scrapers, and bladelet production during the Lower Magdalenian occupations of El Mirón Cave, Cantabrian Spain. Lithic Technol. 41 (3), 212–235. Tarriño, A., Elorrieta, I., García-Rojas, M., 2014. Flint as raw material in prehistoric times: Cantabrian Mountain and Western Pyrenees data. Quat. Int. 364, 94–108. Utrilla, P., 1981. El Magdaleniense Inferior y Medio en la Costa Cantábrica. Monografías del Centro de Investigación y Museo de Altamira 4, Santander. Zubrow, E., Audouze, F., Enloe, J. (Eds.), 2010. The Magdalenian Household. State University of New York Press, Albany.

Please cite this article as: Fontes, L.M., et al., Lower Magdalenian lithic raw material provisioning: A diachronic view from El Mirón cave (Ramales de la Victoria, Cantabria, Spai..., Journal of Archaeological Science: Reports (2017), http://dx.doi.org/10.1016/j.jasrep.2017.03.015