lable at ScienceDirect

Journal of Archaeological Science 55 (2015) 209e218

Contents lists avai

Journal of Archaeological Science

journal homepage: http : / /www.elsevier .com/locate/ jas

Diet and lifestyle in Bronze Age Northwest Spain: the collective burialof Cova do Santo

Olalla L�opez-Costas a, b, *, Gundula Müldner c, Antonio Martínez Cortizas d

a Group Earth System Sciences, Department of Geography. Fac. Xeografía e Historia, Universidade de Santiago de Compostela, Santiago de Compostela,15782, Spainb Laboratory of Anthropology, Department of Legal Medicine, Toxicology and Physical Anthropology, Faculty of Medicine, University of Granada, Granada,18012, Spainc Department of Archaeology, University of Reading, Whiteknights, PO Box 227, Reading, RG6 6AB, UKd Departamento de Edafoloxía e Química Agrícola, Fac. Bioloxía, Universidade de Santiago de Compostela, Campus Vida, 15782, Santiago de Compostela,Spain

a r t i c l e i n f o

Article history:Received 4 June 2014Received in revised form11 January 2015Accepted 17 January 2015Available online 28 January 2015

Keywords:Bronze AgeIberiaStable isotopesPrehistoric dietCave burialMilletPaleopathology

* Corresponding author. Present address: Departmeof Reading, Whiteknights, PO Box 227, Reading, RG6 6

E-mail addresses: olalla.lopez@usc.es, olallalc@gm

http://dx.doi.org/10.1016/j.jas.2015.01.0090305-4403/© 2015 Elsevier Ltd. All rights reserved.

a b s t r a c t

A multidisciplinary investigation of the collective burial of Cova do Santo is presented as a novelapproach to understand daily life during the Bronze Age in Northwest Iberia. The research is focused onthree main aspects: i) taphonomy and patterns of disposal, ii) paleopathology and -demography as in-dicators of health status and lifestyle, and iii) stable isotope analysis to reconstruct paleodiet and toinvestigate the timing of the introduction of millet to the Iberian Peninsula. Osteological analyses wereperformed on 64 bones (61 human and 3 animal); additionally, bone collagen was extracted from 15samples (13 human and 2 animal) and analyzed for its carbon and nitrogen stable isotopes composition.The radiocarbon age of the human remains is consistent with the Middle Bronze Age (c. 1890 to 1600 calBC). The recovered remains belonged to a minimum number of 14 individuals with an estimated age atdeath of forty years or younger. This relatively young age is in contrast to a high prevalence of degen-erative joint disease in the group. The isotopic results suggest a very homogeneous diet, which wasalmost exclusively based on C3 plants and terrestrial animal products. Overall, the data suggest that thestudied population belonged to a period prior to the introduction of spring or summer-grown crops suchas millets. The collective burial from the cave of Cova de Santo, Galicia, currently represents the largestassemblage of prehistoric human remains from Northwest Spain and the relatively good preservation ofthe bones offers a unique opportunity to investigate daily life in Northern Iberia during the Bronze Age.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

The gradual introduction of new crops during the Bronze Agehad a profound impact on the way of life of prehistoric societies inWestern Europe, specifically in bringing about an increase in sed-entism and territoriality (Bettencourt, 1999; Tereso et al., 2013). Theso-called spring or summer-grown crops, such as millets, withshort lifecycles and crucial nutritional value (James et al., 2011;Rachie, 1975), represented a clear advantage in regions whereharsh winters frequently ruined crops with longer growing

nt of Archaeology, UniversityAB, UK. Tel.: þ34 881813315.ail.com (O. L�opez-Costas).

seasons. Spring/summer cultivation may also have enabled to freeup large areas of arable land for winter grazing, bestowing furtherbenefits (Sherratt, 1980). Nevertheless, and despite the importanceof these new crops, there are still few data on when and how theywere introduced to prehistoric Europe and how they impacted onpeople's everyday life (Lightfoot et al., 2013; Tafuri et al., 2009;Weber and Fuller, 2007). Understanding these issues is of partic-ular importance for those areas situated far from the centers ofplant domestication, such as the Iberian Peninsula, where agricul-tural innovations may have taken longer to arrive.

Based on paleobotanic evidences, millets were introduced toIberia in the Middle Bronze Age and increased in importancethrough the Late Bronze Age until they became staple crops in theIron Age (Bux�o and Piqu�e, 2008: 161). This process may have beenmost pronounced in NW Iberia where millet had a continuous

O. L�opez-Costas et al. / Journal of Archaeological Science 55 (2015) 209e218210

presence in the human diet until the arrival of maize in the 16th

century AD (Carballo Arceo, 2006; Vazquez Varela, 1994). There areonly few carpological remains of millets from this area dated to theMiddle and Late Bronze Age, but the number of finds, Panicummiliaceum L. in particular, increases from the Iron Age. Neverthe-less, it is acknowledged that their distribution in paleobotanic as-semblages is likely affected by differential and often poorpreservation (Fig.1 n�5e7) (Bettencourt, 2001; Tereso, 2012; Teresoet al., 2013). As a consequence, there is an ongoing debate to whatextent millets were already established as a regular part of the dietin the Bronze Age (see for example Teira Bri�on, 2010; Tereso, 2012).To address this question, several paleobotanical investigations havebeen conducted (e.g. Bettencourt, 1999, Bettencourt, 2001; TeiraBri�on, 2010; Tereso, 2012; Tereso et al., 2013); however, evidencefrom the study of human remains and specifically isotope analysishas not yet been brought into the debate.

The most direct way to understand the social and biologicalcharacteristics of a population, including its diet and health status,is to analyze the human remains and the burial contexts in whichthey were deposited. Indeed, skeletal remains frequently yield in-formation which is rarely accessible through other types of evi-dence (Larsen, 2002). For more than a century, paleodemographyand paleopathology have provided crucial data to address healthand well-being of past human populations. Recent bio-archaeological work has also included stable isotope analysis ofhuman remains as a means of reconstructing diet at population andindividual level (e.g. Fuller et al., 2012; Tafuri et al., 2009). Theanalysis of carbon and nitrogen isotopes in bone collagen allowsidentifying the presence of C4 plants in the food web (O'Leary,1988; Vogel and van der Merwe, 1977). As millets were the onlyC4 plants available for human consumption in prehistoric inlandSpain, isotope data are particularly suited to determine theircontribution to the diet. Regular consumption of millet will in-crease the d13C ratio of the consumer's bone collagen, while thed15N ratio should remain unaffected, allowing to distinguish C4plants from the consumption of aquatic resources which aregenerally also enriched in 15N (Hedges and Reynard, 2007;Schoeninger and DeNiro, 1984). In diets with sufficient proteinintake, bone collagen stable isotope ratios, which reflect a long-term average of diet, are biased towards the isotopic compositionof dietary protein. As millet and other cereals have low to mediumprotein content (around 10%), a significant intake of these crops isrequired to detect a significant change in collagen isotope ratios

Fig. 1. A map of the geographical areas discussed in the text, including the location of Covet al., 1996), Valdavara I and II (3) (Vaquero Rodríguez et al., 2009) and Cova do Savalleiro (4)found are also numbered: from the Middle Bronze Age the site of Sola (5) (Bettencourt, 1999)(7) (Tereso et al., 2013). Iberian Chalcolithic and Bronze Age sites fromwhere stable isotope2010; McClure et al., 2011); La Garma (10) (Arias Cabal, 2005); Motilla de Azuer (12); (N�aje

(Hedges, 2003). Rather than being a weakness of the method, thismakes stable isotope analysis particularly suitable for determiningwhen millet became a dietary staple (Lightfoot et al., 2013; Tafuriet al., 2009).

This paper presents the results of osteological and isotopic an-alyses of human remains from the collective burial of Cova do Santo(SE Galicia) (Fig. 1, n�1), as an alternative approach to understanddiet and lifestyle during the Bronze Age in Northwest Iberia. Wefocused our research on three main aspects: i) taphonomy andpatterns of disposal, ii) paleopathology and demographic profile asindicators of health status and life conditions, and iii) stable isotopeanalysis to reconstruct paleodiet and the possible introduction ofmillet in the Bronze Age. Our results are contextualized with dataavailable for Chalcolithic and Bronze Age burials of other areas ofIberia.

2. Material and methods

2.1. Collective burials in North-western Iberia

There is a large number of Bronze Age megalithic burial moundsin Galicia (Bettencourt, 2008), but these have rendered no humanand only few animal bones to date. Similarly, the Bronze Age cistsand the numerous stone built chests (see review in Bettencourt,2008), only very rarely contain human remains, a fact usuallyexplained by the dominance of acidic soils in the region (L�opez-Costas, 2012). Since any finds of human remains from prehistoricGalicia are therefore highly exceptional, bioarchaeological in-vestigations are not usually possible and most of what is assumedabout the demography and lifestyle of Bronze Age populations hasbeen inferred by analogy with later periods.

The Cova do Santo cave is located at 510m a.s.l. in one of the fewkarstic regions of NW Spain, in the western slopes of the river Silvalley (Fig. 1, n�1, Fig. 2). Nowadays, the area is sparsely populated;however, pass control of the Sil river valley was of considerablestrategic importance in the past, since it formed the mainconnection between the Castilian plateau of central Spain and theAtlantic coast (via the rivers Mi~no and Sil) (Ruiz-G�alvez Priego,1998). From the number of human bones recovered here, Cova deSanto represents the largest prehistoric assemblage of NW Spain.Other caves with prehistoric material have been discovered in thearea, namely the sites of Valdavara (Lugo province), Cova doSavalleiro (Lugo) and Pala da Vella (Ourense province) (Fig. 1). The

a do Santo (1), and other Galician burial caves: Pala da Vella (2) (Fern�andez Rodríguez(Fern�andez Rodríguez and Ramil Rego, 1995). The places where millet seeds have been, and from the Late Bronze Age the sites of Santinha (6) (Bettencourt, 2001) and As Laiasdata are available: Bolores (8) (Lillios et al., 2010); Cova Pastora (9) (García Puchol et al.,ra Colino et al., 2010); and Castillejo de Bonete (12); (Salazar-García et al., 2013).

Fig. 3. Ground plan of the cave modified fromM�endez Fern�andez and Martínez L�opez�sreport (2009). Triangles identify the external entrance and accesses between cham-bers. Chamber C was inaccessible and its shape has been estimated.

O. L�opez-Costas et al. / Journal of Archaeological Science 55 (2015) 209e218 211

Bronze Age strata of the two caves of Valdavara produced 64 boneswhich belonged to a minimum number of five individuals (VaqueroRodríguez et al., 2009). Two fragments of right femora, which havenot been dated, were found in Cova do Savalleiro (Fern�andezRodríguez and Ramil Rego, 1995). Finally, a range of archaeolog-ical materials of probably of Neolithic age, including pottery, hu-man and animal bones, were discovered in Pala da Vella, located afew kilometers Northeast of Cova do Santo (Fig. 1), although thesehave not yet been studied in detail (Fern�andez Rodríguez and VillarQuinteiro, 2003; Fern�andez Rodríguez et al., 1996; P�erez Ortiz andFern�andez Rodríguez, 2005).

Inhumation in caves was a common practice in Northern Iberiafrom the Neolithic to the EarlyMedieval period, especially along theBasque Country and Cantabria where more than 350 caves withhuman remains have been recorded to date (Arias Cabal andArmendariz Guti�errez, 1998; Armendariz Guti�errez, 1990;Onta~n�on Peredo and Armendariz Guti�errez, 2005). The numberdecreases significantly to the west, with few known caves in Galiciaand Asturias. A unique phenomenon on the Atlantic coast ofNorthern Spain is the deposition of the dead inside mines, some-times with mining tools (see Arias Cabal and Armendariz Guti�errez,1998). In Portugal, there are also examples of shelters and cavesused for burial, although less than in Northern Spain (Bettencourt,2011). Even though the sites span a large geographical area, and fewcomplete osteological studies and radiocarbon dates are available.The need for further bioarchaeological studies of prehistoric sitesand burial caves in NW Iberia has already been emphasized in thearchaeological literature (Vaquero-Rodríguez et al., 2011; VaqueroRodríguez et al., 2009).

2.2. Cova do Santo and its bone assemblage

The Cova do Santo cave, whose original entrance collapsed, cantoday only be accessed through a narrow vertical shaft (M�endezFern�andez and Martínez L�opez, 2009). The interior is divided intothree chambers (Fig. 3). The largest of these, “chamber A”, opens tothe outside and its floor is partially covered by large stones fallenfrom the ceiling (Figs. 3 and 4B). A large number of animal andhuman remains are scattered on surfaces, many of them in crevices.Various pieces of Bronze Age pottery were reportedly found inchamber A by a group of potholers and some residents of Pardoll�an(Fig. 2) (M�endez Fern�andez, 2006; M�endez Fern�andez andMartínez L�opez, 2009; Su�arez Otero, 2014). The smaller chambers

Fig. 2. Aerial view of the site with the location of the river that borders it and the nearest vParcelas Agrícolas, Consellería do Medio Rural e do Mar. Xunta de Galicia).

B and C adjoin chamber A to the southwest and are both situated ata lower level. In chamber B, human, other mammals and bird re-mains as well as snail-shells were again found scattered on theground, while other bones had been placed in a large rock crevice atthe narrow western end (Figs. 3 and 4A,C). Chamber C, whose levelis the lowest of the three (Fig. 3), is not currently accessible forsafety reasons; nevertheless, several crania could be observed on itsfloor through the narrow passage leading to it from chamber A.

Since its inclusion in the Galician autonomic catalog of heritagein 2004 (M�endez Fern�andez, 2006), few surveys of the cave'sinterior have been conducted. The instability of the ceiling hasprevented any systematic archaeological investigations from takingplace; consequently archaeological work in the form of surveys waslimited to a few hours and no excavations were possible. The mostextensive survey was conducted in the summer of 2008, duringwhich a number of bioarchaeological finds were collected. Thecollection of the bones was influenced by the cave instability andthe short periods of granted access, so only those ones placed onthe surface were picked up. Every find, including bone, was pho-tographed in situ and their position and the osteological andpaleopathological main features were carefully recorded (seeTable 1). The recovered bones (61 human and 3 animal) werecarefully cleaned before full taphonomic, paleopathological andosteological analyses were performed. Of the human bones, twocranial fragments and a femurwere sent for radiocarbon dating andwere therefore not available for other analyses. Three identifiableanimal bones were also recovered, two from ovicaprids and one

illage, Pardoll�an (above), obtained from SIXPAC (Sistema de Informaci�on Xeogr�afica de

Fig. 4. Photographs of the inside of the cave. (A) Floor of chamber B with a mixture of human and bird remains and snail-shells; (B) View towards the northern end of chamber A. Inthe foreground a number of large rocks from the partially collapsed ceiling are visible with human remains both above and below them; (C) The rock crevice at the western end ofchamber B which contained numerous animal and human bones, most of them covered by a thick calcareous patina. The cranium of a subadult individual which was subsequentlysent for radiocarbon dating can be seen at the back.

Table 1Inventory of the number of human bones (left and right) recovered from Cova doSanto in the 2008 survey. The wide majority of the bones were complete. More thanone fragment belonged to the same bone were count as one. Crania withoutmandible were count as one. Non-listed bones were absent.

Element Chamber A Chamber B Total

Crania 1 5 6Mandible 3 3Vertebra 5 4 9Rib 5 1 6Scapula 1 1 2Clavicle 1 1Os coxae 4 1 5Humerus 1 1Radius 3 3Ulna 1 1 2Femur 4 6 10Tibia 4 2 6Fibula 5 1 6Tarsal 1 1Total 38 23 61

O. L�opez-Costas et al. / Journal of Archaeological Science 55 (2015) 209e218212

from pig. One of the ovicaprid bones was radiocarbon dated. Apreliminary report of the demographic profile and other osteo-logical features was previously published (Lopez-Costas, 2008).

Three human and one animal bone were radiocarbon-dated byaccelerator mass spectrometry at Beta Analytic Inc and OxfordRadiocarbon Accelerator Unit. The human bones were selectedfrom reliable archaeological contexts (two semi-buried and theother from a crevice); the animal bone was a surface find. The AMS14C dates obtained on human samples provided ages in the rangefrom 1890 to 1600 cal BC (2s), which place the deposition of thehuman remains into the Bronze Age, specifically the local MiddleBronze Age (Table 2). In contrast, the animal bone provided an ageof ~3000 cal BC, corresponding to the Late Neolithic. The fact thatthree randomly selected human samples were dated within thesame, relatively tight time-frame suggests that the Middle BronzeAge may have been a period of high activity at Cova do Santo,although it is difficult to know if the cave was also used for burialsbefore or afterward. The earlier date obtained for the ovicapridcould mean that the animal was not related to the human assem-blage (many animals die in caves due to falls or disorientation) orthat it was part of the grave goods of some earlier burial. Thechronology of the human remains suggests depositions belonged toa same burial context which took place during Middle Bronze Age.This supports the interpretation of the Cova do Santo site as acollective burial.

Because the human remains were found disarticulated, eachbone was analyzed individually. Preservation was evaluatedfollowing the abrasion/erosion scale devised byMcKinley (2004). Inaddition, all taphonomic alterations were classified according toBotella et al. (1999). The minimum number of individuals (MNI)was estimated using the recommendations by Roberts (2009; page120). Markers for sex and age estimation and pathological lesionswere recorded using the international standards most commonlyused in Spain (for a summary see M�arquez-Grant et al., 2011).Specifically, sex was estimated using established markers on theinnominate and cranial bones, and metric analyses of long bonescompared to Iberian reference samples (Aleman et al., 1997;

Trancho et al., 1997). International methods for innominate bone(auricular surface and pubic symphysis morphology), 4th rib, dentalwear and cranial suture closurewere used to estimate age in adults,and dental development and epiphyseal fusion in subadults (seee.g. Buikstra and Ubelaker, 1994). In addition, subadult age in thepreserved postcranial remains (ilium, femur and tibia) was alsoassessed with the Iberian standards for growth and maturity(Lopez-Costas et al., 2012; Rissech andMalgosa, 2005; Rissech et al.,2008). Pathological features were evaluated macroscopically,recording the presence, appearance and distribution of abnormalbone formation and destruction in each element.

2.3. Stable isotope analysis

Stable isotope analysis of bone collagen was performed on 15samples, representing 14 different individuals (12 humans and 2

Table 2AMS radiocarbon dates from Cova do Santo. Conventional radiocarbon dates were calibrated using the Oxcal v4.0 (C. Bronk Ramsey) and the ‘INTCAL13’ dataset (RadiocarbonVol 55e4) (y, years old; Indet.¼ indeterminate). Datesmarked (*) were obtained fromM�endez Fern�andez andMartínez L�opez unpublished report (2009) and samples were notavailable for other analyses. OxA-25075 sample belongs to cranium 112.

Code d13C 14C Age 2s cal BC Probability Species/bone Age/Sex

*Beta e 243023 �19.0‰ 3400 ± 40 1781e1611 89.2% Human/cranium (Fig. 5) 0e6 y/Indet.*Beta e 243024 �19.7‰ 3440 ± 40 1881e1658 95.4% Human/femur >20 y/Indet.*Beta e 243025 �21.7‰ 4400 ± 40 3115e2908 86.7% Ovicaprid/tibia Adult/Indet.OxA-25075 �19.7‰ 3486 ± 31 1892e1739 92.1% Human/cranium (sample 112) 20e30 y/female

O. L�opez-Costas et al. / Journal of Archaeological Science 55 (2015) 209e218 213

animals). The selection of bones was based on MNI. In order toavoid resampling the same skeleton, a carefully selection of humanremains for isotopic analysis was done. Human adult samples weretaken from 4 left and 1 right femora, 4 coxal bones, with differentanatomical conformation and no articular-fit at level of the ace-tabulum and one cranium. The other samples belong to subadultswith very different stage of maturity. One human rib was alsoadded as it was analyzed in a preliminary step to assess collagenpreservation.

Collagen extraction followed the method described by Longin(1971) with modifications recommended by Collins and Galley(1998), according to the protocol described in Britton et al.(2008). Carbon and nitrogen stable isotope ratios were measuredin duplicate on a Europa 20e20 isotope ratio mass spectrometercoupled to a Sercon elemental analyzer. Analytical error was ±0.2‰or less for both elements (1 s.d.), and calculated from repeatedanalysis of internal standards calibrated to international referencematerials.

2.4. Statistical analysis

Descriptive statistics (means and standard deviations) werecomputed for each group of samples. The comparisons among sex(male/female) and age groups (subadult/young/middle adults)were made by a non-parametric ManneWhitney U (U-test) andKruskaleWallis (KeW) tests, to account for small group sizes. Non-parametric tests were selected since a probability distributioncannot be assumed with the available sample sizes. SPSS.16 wasused to perform the statistical analysis.

3. Results and discussion

3.1. Patterns of disposal and taphonomy

The disposal of the skeletal remains in the cave was analyzedlooking for patterns that would help understanding the mortuarypractices at Cova do Santo. The human bone assemblage is sum-marized in Table 1. All remains were found disarticulated andscattered across the different chambers of the cave (Fig. 4A, B),suggesting that they are in secondary context. Systematic excava-tions, once they become possible, may still yield articulated burialsin the lower, undisturbed layers and it is therefore not possible atthis time to say with certainty whether the present disarticulatedstate of the assemblagewas the result of the disturbance of primaryburials or of deliberate deposition during the mortuary use of thecave. Nevertheless, the thick calcareous patina that covered thebones found in rock crevices indicates that these, at least, had beenplaced there a considerable time ago (Fig. 4C). This and the fact thatsmaller bones were mostly missing from the assemblage thereforesuggest that the cave may have been used for secondary burial. Inthe absence of systematic excavation it is, once again, difficult toestablish whether the remains were moved to the cave after thebodies had been skeletonized elsewhere or whether these had beenprimary burials inside the cave, which were later re-arranged. In

any case, the burial seems collective in naturewithout evident signsof differentiation between skeletal elements or individuals. Theimpression that no individual was given special treatment is alsosupported by the location of the pottery, which was not placedwithspecific skeletons but clustered on the same rock slab in Chamber A(M�endez Fern�andez and Martínez L�opez, 2009).

The patters of disposal at Cova do Santo showcertain similaritieswith nearby Galician sites with human remains. At Valdavara IeIIand Pala da Vella layers of scattered human remains were found(Fern�andez Rodríguez et al., 1996; Vaquero Rodríguez et al., 2009).At least five individuals were present at Valdavara, but their skel-etons were incomplete and the assemblage was interpreted asbelonging to a secondary multiple burial, where largest bones weremoved and deposited there, while the smaller ones were left in anunknown primary location (P�erez Ortiz and Fern�andez Rodríguez,2005; Vaquero Rodríguez et al., 2009). Nevertheless, in a recentcritical review of the burial evidence from Northern Iberian caves,Onta~n�on and Armendariz (2005) suggest that most of the findsrepresent primary burials that were later displaced and movedaround the chambers. They concede, however, that this interpre-tation must be tentative because of the incomplete nature of theosteological data. Evidence for this scenario is much better at sitesin Northern Portugal, wheremost skeletons were found articulated,except for some specific secondary manipulations. In Gruta Grande,for example, one of the skulls was found separated from the rest ofthe skeleton (Delgado, 1887 cited by Bettencourt, 2011). In sum-mary, while it is likely that the human remains from Galician cavesrepresent the remains of primary burials, more research and spe-cifically modern excavations of primary contexts are necessary toestablish this for certain.

3.2. Preservation of skeletal remains

The studied remains were generally well preserved, even themore fragile ones such as vertebrae. Most of them were completeand only a small number of specimens presented minor post-mortem breaks. These latter ones may have been crushed in situby previous visitors to the cave -they were all situated along thepaths in the cave. No differences in taphonomic damagewere founddepending on sex or age at death and no signs of animal activity orerosion were observed.

The majority of bones (53/58 or 91%) showed little surfaceerosion (grade 0 on McKinley's, 2004, scale), which is remarkableconsidering that 50% was the maximum value observed in otherGalician skeletal assemblages from historic periods (L�opez-Costas,2012), and demonstrates the favorable preservation conditionsinside the cave. Apart from a few root impressions, which werepresent exclusively on bones recovered from near the caveentrance, the most relevant kind of surface alteration was thepresence of calcareous patinas on 21 out of 58 (36%) of the speci-mens. Overall, the state of preservation and the patinas suggest thatmost of the bones had been lying in the places they were recoveredfor a long period of time.

O. L�opez-Costas et al. / Journal of Archaeological Science 55 (2015) 209e218214

3.3. Minimum number of individuals (MNI) and demographicprofile

Although the number of individuals that had once been buriedin chambers A and B was likely higher, the analyzed elementsbelonged to a minimum number of 14 individuals: six women, fourmen, three subadults and one individual of indeterminate sex andage (see Table 3). The nature of the assemblage did not enable us toassign bones to the same individual with any certainty.

The remains of at least two children (4e6 and 6e8 years) wereidentified based on two mandibles and a right ilium. Other ele-ments such as a fragmented neurocranium, a tibia and a femurwere estimated to have a similar age at death but it was notpossible to assign them to either of the children with any certainty,due to the overlap in their age ranges. Several postcranial bones(right ulna and clavicle, a left femur, a vertebra and both tibiae)showed equivalent stages of epiphyseal fusion. According to theinternational standards and Iberian data for femur development(Rissech et al., 2008; Scheuer and Black, 2000) they may havebelonged to at least one adolescent (17e20 years) individual,probably a female (Aleman et al., 1997; Trancho et al., 1997). A semi-visible fusion line in a male femur was estimated at 18e25 yearsold, following the Iberian standards of cessation of femur growth(Rissech et al., 2008).

The MNI of adult skeletons was calculated based on the numberof femora and innominate bones. The adult skeletons were evenmore difficult to individualize because of the imprecise sex andwide age at death estimation ranges obtained from most of theskeletal elements. In order to avoid the possible bias in young adultage estimation, the adult demographic profile includes only ele-ments used in the MNI, which were aged on the basis of epiphysealfusion, auricular surface and pubic symphysis degeneration. Bothsexes and almost every age group were represented in Cova doSanto, as they were in most of the caves from the Basque Countryand Cantabria (Onta~n�on Peredo and Armendariz Guti�errez, 2005).Themain characteristic of the demographic profile of Cova do Santois the relatively young age, since none of the examined elementsshowed signs of an advanced age at death (i.e. older thanz40 yearsold). The best represented age group was that of young adults(20e34 years old), with at least four individuals, followed by sub-adults (Table 3). However, the small MNI and disarticulated natureof the assemblage do not allow drawing conclusions on general lifeexpectancy.

3.4. Paleopathology

A notable feature is the relatively high incidence of degenerativejoint diseases (DJD), spondylosis and osteoarthritis (OA): presenceof porosity and osteophytes (22/26 of observed joints), given theyoung age of the sample. Most of them were cases of mild (13) or

Table 3Age and sex distribution based on the obtained MNI (y, years old).*The age ranges ofthese individuals are intersections between the subadult and young adult sets. Theyhave been included in the most probable group according to their bone features.Changes on this classification may affect the total count per age group.

Age Number Description

Subadult (<19 y) 3* 4e6 y, 6e8 y and a female 17e20* yYoung adult (20e34 y) 4* 2 Males: 18e25* y and 30e34 y

2 Females: 18e25* y and 20e30 yMiddle adult (35e49 y) 2 2 Males: 35e40 y and 35e40 yOld adult (>50 y) 0Indeterminate 5 4 Females

1 Indeterminate sex

medium (8) OA, without signs of eburnation; and six of the sevenpreserved vertebral bodies showed marginal osteophytes. There isan agreement in the osteological literature that DJD and specificallyOA are multietiological factors, amongst which aging is one of themain influences in the onset and severity of the lesions (Weiss andJurmain, 2007). Physiologically mediated activity also has animportant role in the development of OA in young ages, althoughnone direct correlation has been found between intense strenuousactivity and its onset (Knüsel, 2002). In light of this to, the elevatedOA frequency can be explained by two reasons: i) the skeletal el-ements with this pathology belonged to older individuals thanthose used to generate the MNI and demographic profile of thesample (since the majority of the ages were estimated usinginnominate bones) or, more likely, ii) the OA may have been causedby intense physical activity from a young age.

Apart from the DJD only few pathological lesions were noted.New bone formation consistent with nonspecific inflammatoryperiostitis marks was observed on 5 of the 27 long bone shafts(18%). Furthermore, porotic lesions were noted on few bones: onecranium (1/1) presented cribra orbitalia, and three femoral necks(3/7), one subadult and two adults, exhibited cribra femoralis suchas the one described byMiquel-Feucht et al. (2001). No other marksrelated to infectious or metabolic diseases were observed, apartfrom a female adult cranium showing signs of pitting and erosiveactivity in the alveolar and palatine processes. A gummatous dis-ease may have led to these bone lesions -a detailed study of thiscranium is currently been carried out.

Given the rocky and mountainous topography of the area, thealmost complete absence of traumatic lesions is also noteworthy,despite the small sample size. In fact, only one trauma wasobserved, a perimortem lesion on the neurocranium of a youngwoman (Fig. 5). It extends from the right to the left parietal andexhibits sharp edges without evidence for remodeling. Despite thelack of the frontal bone and the frontal border of the right parietaldo not allow determining it with certainty, the lesion is compatiblewith a sharp force trauma. Only one adult maxilla and threemandibles (one from an adult and two from subadults) wereavailable to examine for oral pathologies. These present numerouspost-mortem and five ante-mortem tooth losses. The eleven teethwhich were available for assessment (one adult premolar, fouradult molars, and eight deciduous molars) showed no evidence ofcaries or enamel hypoplasia and were affected by medium-leveldental calculus. The two adult jawbones showed no evidence ofperiodontal disease.

Fig. 5. Evidence of perimortem trauma in one of the neurocrania probably belonging toa young woman (white arrows indicate traumatic lesion).

Table 4Carbon and nitrogen stable isotope ratios, collagen quality indicators, osteologicaland archaeological information for human and animal samples (y ¼ years old;Indet.¼ indeterminate). *Data for sample 113 which was originally processed to testcollagen preservation are included here, even though the bone was not used for MNIand did not necessarily belong to a separate individual.

Sample Bone d13C d15N C/N %C %N %Coll Sex/species Age group

101 Coxal �19.8 9.7 3.2 43.7 15.7 22.79 Female 18e25 y102 Coxal �20.1 9.5 3.2 41.9 15.0 21.45 Male 30e34 y103 Femur �20.1 9.3 3.2 44.2 16.1 20.88 Female Indet.104 Femur �20.1 9.3 3.2 41.7 15.0 19.61 Female Indet.105 Coxal �19.9 9.4 3.2 43.5 15.9 23.72 Male 35e40 y106 Femur �19.9 9.3 3.2 43.6 15.8 24.10 Female 17e20 y107 Femur �19.8 10.2 3.1 44.1 16.4 21.89 Male 18e25 y108 Femur �19.8 9.4 3.2 41.4 15.0 18.43 Female Indet.109 Coxal �20.0 9.4 3.2 43.4 15.8 16.12 Male 35e40 y110 Cranium �19.7 9.7 3.2 43.7 16.1 20.97 Indet. 6e8 y111 Ilium �20.3 8.9 3.2 42.5 15.5 18.98 Indet. 4e6 y112 Cranium �19.8 9.5 3.2 39.6 14.2 17.87 Female 20e30 y113 Rib* �19.9 9.2 3.3 44.1 15.7 3.95 Indet. 20e40 y901 Cranium �20.7 9.1 3.2 42.9 15.4 18.70 Pig 0e1 y902 Mandible �20.3 6.4 3.2 43.2 15.9 21.41 Ovicaprid >3 y

O. L�opez-Costas et al. / Journal of Archaeological Science 55 (2015) 209e218 215

The low number of pathological lesions should not be inter-preted directly as an indicator of the good overall health status ofthe population, since the relationships between both variables areof course complex. Firstly, it is important to consider that thestudied sample is limited and that individuals buried in the cavemay have been pre-selected, for their status or other reasons.Nevertheless, with the exception of DJD, there were few visiblesigns of stress (non-specific infections, trauma, porotic lesions,metabolic disorders). Two radically different scenarios can explainthese results. On the one hand, the examined remains may havebelonged to frailer individuals whose death happened soon afterbeing affected by disease and before any bony response occurred(Wood et al., 1992). Therefore, the stress markers are not visible inthe skeletons. On the other hand, the individuals could have had alow susceptibility to stress and were not seriously affected by fac-tors such as disease or malnutrition during their lifetime. Alterna-tively, they could have a high tolerance for stress and/or recoveredquickly from any diseases affecting them. Although these lattersituations are possible, they are not easily consistent with therelatively low life expectancy observed in the paleodemographicrecord.

Finally, the evidence of interpersonal violence in prehistorictimes has been recorded in some sites of Iberia (e.g. Aranda-Jim�enez et al., 2009; Etxeberria Gabilondo and Herrasti Maci�a,2007). Jim�enez-Brobeil et al. (2009) found a significant increasein cranial injuries during the Bronze Age in Southern Spain (Argarculture), in comparison with preceding and subsequent periods.The authors suggest that the appearance of militarist and hierar-chical societies during Bronze Age may have been responsible for arise in interpersonal violence. However, as Aranda-Jim�enez et al.(2009) have noted, the absence of sharp injuries does not allowinterpreting directly the trauma as the result of a violent act(Aranda-Jim�enez et al., 2009). To our knowledge, there are nosimilar studies with regard to the NW Iberia. Although amass graveafter a violent event may be a plausible explanation for some col-lective burial, the assemblage from Cova do Santo exhibited nounequivocal evidence for interpersonal violence, nor were anyarchaeological artefacts recovered that might suggest this.

3.5. Paleodiet

The bone collagen obtained was very well preserved, with anaverage yield of 19.4 ± 4.8 wt% (1 s.d.). The individual data andsummary statistics are given in Tables 4 and 5 and are plotted inFigs. 6 and 7. The Cova do Santo animal sample comprises oneovicaprid mandible, one ovicaprid tibia and one pig cranium(Table 2, Table 4 samples 901e902 and Fig. 6A). The two ovicapridbones belonged to adult specimens and it cannot be excluded thatthey belonged to the same individual. The ovicaprid tibia d13C datawas obtained through radiocarbon dating, so its isotopic value maynot be directly comparable. The presence of calcareous patinas onthe bone surface suggests ancient deposition, as also supported bythe Late Neolithic age of the ovicaprid tibia (Table 2). Still, the threed13C fall between �21.7‰ and �20.3‰ and therefore reflect over-whelmingly C3-ecosystem based diets with no measurable contri-butions of C4-plants (or marine resource consumption). The pig(sample 901) presents a relatively high d15N ratio (9.1‰), which isin the range for human samples. It belonged to a young specimen(less than one year old), and the nitrogen isotope signal may berelated to the consumption of mother's milk, which is commonlyknown as ‘suckling effect’ (Balasse and Tresset, 2002). This effectcomplicates the determination of terrestrial protein intake, but,with d13C values of �19.7‰ and below, there is no evidence of C4plant or marine resource consumption in the animals. The nitrogenisotope baseline of the herbivore diet is therefore poorly

characterized with data for only the mandible (sample 902;d15N ¼ 6.4‰) (Table 4).

Human diet was explored by analyzing a selection of thirteensamples; twelve of them belong to the bones used to estimate theMNI (Table 4). d13C ratios are between �20.3‰ and �19.7‰ with arange of 0.6‰ and the d15N values from 8.9‰ to 10.2‰with a rangeof 1.3‰. Arithmetic means and small standard deviations(�19.9 ± 0.2‰ for d13C and 9.4 ± 0.3‰ for d15N) indicate a highlyhomogeneous isotopic composition of the diet. The intra-population analysis showed no significant differences betweenthe sexes (Table 5) or different age groups (KeW test: d13C df 2p ¼ 0.68; d15N df 2 p ¼ 0.45). Therefore, the isotopic compositionsuggests that all individuals shared a similar diet (Fig. 6B).

The spacing between the single ovicaprid sample and the hu-man mean is 0.4‰ and 3.0‰, for d13C and d15N respectively. Thelack of a robust herbivore and faunal baseline limits the interpre-tation of these data with respect to the diet of the Cova do Santohumans. Assuming 3e5‰ as the typical size of a collagenecollagend15N trophic level offset (Bocherens and Drucker, 2003), these datasuggest a mixed diet of plant and animal products with moderateanimal protein intake. However, there are severe limitations todetermining the relative importance of non-plant proteinwhen thelocal baseline (stable isotopes in crops and animals) has not beenproperly assessed. Manuring practices in crops, non-local prove-nance of the livestock, local variations in the isotopic compositionof plants or different fodder practices between animal species couldall influence the d13C and d15N observed in humans (see Fraser et al.,2013; Makarewicz and Tuross, 2006; Zavodny et al., 2014). Amongothers, some manuring practices produce a shift in 15N values ofdomestic livestock (e.g. Fraser et al., 2013), while the inclusion of C4pathway plants in fodder have the same consequences in 13C (e.g.Zavodny et al., 2014). Current animal samples do not allow checkingthe existence of these changes on trophic baseline and their in-fluence in human isotopic results; and this fact should be consid-ered when interpreting human diet. Cova do Santo is located in therock face above the river Sil, which is wide and rich in fish at thispoint (Fig. 2). Riverine fish have varying isotopic compositions (e.g.Fuller et al., 2012) and in the absence of fish bone samples from theriver Sil, we cannot define endpoint members to evaluate theircontribution to diet. However, at least in comparison to the singleherbivore sample available the isotopic signal of the Cova de Santohumans seems consistent with a fully terrestrial diet and there is noindication that freshwater protein made an important contributionto everyday subsistence.

Table 5Statistical summary inter-sex comparison of the d13C and d15N from Cova do Santo.

Cova do Santo Male Female ManneWhitney U-test

n 4 6X ± SD(d13C‰)

�20.0 ± 0.1 �19.9 ± 0.1 U 14.0 p ¼ 0.76

X ± SD(d15N‰)

9.6 ± 0.4 9.4 ± 0.1 U 6.5 p ¼ 0.26

Fig. 7. Scatter plot of human d13C and d15N data from Iberian Chalcolithic and BronzeAge sites. Adults are represented by solid symbols, subadults by empty symbols, thesample from Cova Pastora and Avenc included only adult individuals.

O. L�opez-Costas et al. / Journal of Archaeological Science 55 (2015) 209e218216

Few stable isotope data are available for Chalcolithic and BronzeAge populations of the Iberian Peninsula (Figs. 1 and 7). For theIberian Atlantic region d15N values are similar to those presentedhere (~9e10‰). A skeleton found in the cave of La Garma(Ribamont�an al Monte, Cantabria; 1746e1493 cal BC) has thelowest d13C ratios of the Chalcolithic and Bronze Age periods(�21.0‰ for d13C and 9.5‰ for d15N, Arias Cabal, 2005). The latterstudy also provides d13C data for Chalcolithic (X�20.8 ± 0.9‰;n ¼ 10) and Bronze Age (X�21.0 ± 0.6‰; n ¼ 11) remains fromcaves located on the Atlantic coast of Northern Iberia. Although thedata were obtained during radiocarbon dating and may thereforenot be directly comparable, these values are in line with the lowd13C observed at La Garma (Arias Cabal, 2005). Despite the sites'location close to the sea, there is therefore no evidence of 13Cenrichment in these samples that would indicate a measurablecontribution of marine protein to the diet. It would nevertheless bedesirable to repeat the analysis in the future in order to obtain fullycomparable (palaeodietary) d13C and d15N data. The Chalcolithicadults of the cave of Bolores (Torres Vedras, Lisboa, Portugal;2910e2460 cal BC;�19.6 ± 0.1‰ and 9.4 ± 0.3‰, n¼ 3) (Fig. 7) alsoshow a diet based on terrestrial C3 plant protein, as does a BronzeAge subadult individual from the same site (1800 cal BC; �20.4‰and 7.5‰) (Lillios et al., 2010).

Data from Central Iberia and the Mediterranean coast displayhigher d13C averages. Samples from the cave burials of Cova Pastora(Alcoi, Alicante; Chalcolithic 2913e2635 cal BC, Bronze Age1909e1752 cal BC) and Avenc dels Dos Forats (Carcaixent, Valencia;Chalcolithic 2870e2570 cal BC) are slightly enriched in 13C (Fig. 7)(García Puchol et al., 2010; McClure et al., 2011), specifically in thetwo Chalcolithic individuals from Avenc whose d15N ratios arehigher than those of the Bronze Age samples. Two necropoleis fromSouth Central Iberia, Motilla de Azuer (Daimiel, Ciudad Real; BronzeAge 2200e1350 cal BC) and Castillejo de Bonete (Terrinches, Ciudad

Fig. 6. Scatter plot of carbon and nitrogen isotope ratios from Cova do Santo. (A)

Real; Chalcolithic-Bronze Age 2340e1920 cal BC), present thehighest d13C ratios (Fig. 7) (N�ajera Colino et al., 2010; Salazar-Garcíaet al., 2013). Both are located in the same general region (within lessthan 100 km of each other), and featured inhumations in single ordouble earth graves with the occasional presence of grave goods,characteristics which differ from the Atlantic cave burials. The datafromMotilla de Azuer are characterized by relatively homogeneousd13C (X�18.6 ± 0.3‰; �19.4‰ to �17.8‰; n ¼ 59) but large vari-ation in d15N (X 10.6 ± 1.4‰; 8.6‰e13.6‰) (N�ajera Colino et al.,2010). The Castillejo de Bonete data appear to fall into twogroups, 2 individuals with values similar to those of Cova do Santo(ind 12726e20.2‰ and 7.7‰; ind 12729e20.3‰ and 8.6‰) andthree with data more consistent with those from Motilla de Azuer(ind 12725e18.7‰ and 9.8‰; ind 12727e18.8‰ and 9.2‰; ind12728e18.6‰ and 11.2‰) (Salazar-García et al., 2013). The 13C-enriched values in these two datasets were left unexplained (N�ajeraColino et al., 2010) or attributed to the intake of marine resources(Salazar-García et al., 2013), despite the fact that the sites are sit-uated at considerable distance from the sea (270 and 200 km,respectively, see Fig. 1). The latter interpretation relies on theassumption that C4 plants (millets) are poorly represented in thearchaeological record from Bronze Age Spain (see Bux�o and Piqu�e,2008); however, as we have argued earlier, paleobotanic evidence

humans and animals, (B) humans divided into male, females and subadults.

O. L�opez-Costas et al. / Journal of Archaeological Science 55 (2015) 209e218 217

is sparse and affected by problems of differential preservation, suchas the fact that millets rarely survive in charred assemblages(R€osch, 1998). While the currently available isotope data thereforeprovide no evidence of millet in the human diet on the SpanishAtlantic coast, the possibility of early consumption of C4 plants inSouthern Central Iberia cannot be dismissed. However, more iso-topic studies, in bone collagen and apatite, the generation of robustfaunal baseline data as well as further archaeobotanical researchare needed to determine the importance of millets in the diet ofBronze Age Iberia.

4. Conclusions

The Cova do Santo cave contained a large number of humanremains distributed over the two chambers which are currentlyaccessible. The skeletal elements were found disarticulated, prob-ably because of their deposition as secondary burials, a rite similarto that observed at other cave sites from Northwest Spain butscarcely seen in other areas of Northern Spain and Portugal. The ageof the 14C-dated human bones is consistent with the Middle BronzeAge (1890e1600 cal BC), and it is likely that most, if not all, burialactivity occurred during this period. The MNI of the recoveredhuman remains was 14: six females, four males, three subadultsand one individual of indeterminate sex and age. Estimated ages atdeath suggest that none of the individuals lived beyond forty years.The bones showed few signs of pathologies, with the exception of ahigh prevalence of degenerative joint disease, which could berelated to intense physical labor at a young age. There was a lowincidence of trauma and no clear evidence of interpersonalviolence, a fact which has been found in other Iberian prehistoricsites.

The stable isotope data from Cova do Santo are in agreementwith those obtained from other Chalcolithic and Bronze Age caveburials on the Atlantic coast of Iberia. The results suggest a diet thatrelied exclusively on C3 plant-based protein, without any measur-able consumption of marine protein or C4 plants. This is in contrastto sites from South Central Iberia, which present slightly higherd13C in comparison, a fact that may be related to minor consump-tion of millet or other C4 plants (e.g. sorghum). Since the d13C ratiosare lower than those observed in other areas of Europe, for nowwetherefore conclude that there is currently no evidence to supportthe suggestion that millets became a staple crop during the BronzeAge in the Iberian Peninsula. The isotopic data from Cova do Santocertainly suggest that the studied population belonged to a phaseprior to the introduction of these spring or summer-grown crops(millets).

This work has provided new data on a poorly studied type ofburial and new insights into the prehistoric life of NW Iberia, as thisis the first detailed study of a bone assemblage from an area wherebone preservation is compromised by the acidic nature of mostsoils and sediments. Future work on the cave will hopefullyimprove the structural stability of the ceiling and enable to renewthe fieldwork, including the recovery of a larger sample of humanremains, which will further advance our understanding of health,diet and daily life in the Bronze Period of Atlantic Europe.

Acknowledgments

This research was conducted under the research projects“Antropoloxía dos restos �oseos humanos de Galicia”, GPC2014/009and R2014/001 funded by the Xunta de Galicia (GPC2014/009 andR2014/001). O. L�opez-Costas is funded by a research grant PlanGalego I2C mod. A with the project “Estudo da influencia da dieta eos cambios clim�aticos no impacto das enfermidades infecciosas(tuberculose e brucelose) en sociedades pret�eritas”. Thanks are also

extended to the director of the archaeological intervention, FidelM�endez Fern�andez, and to Tina Moriarty for her assistance duringstable isotope labwork. The authors wish to thank Miguel C. BotellaL�opez for his help in the paleopathological study.

References

Aleman, I., Botella, M.C., Ruiz, L., 1997. Determinaci�on del sexo en el esqueletopostcraneal. Estudio de una poblaci�on mediterr�anea actual. Arch. Esp. Morfol. 2,7e17.

Aranda-Jim�enez, G., Mont�on-Subías, S., Jim�enez-Brobeil, S.A., 2009. Conflicting ev-idence? Weapons and skeletons in the Bronze Age of south-east Iberia. Antiq83, 1038e1051.

Arias Cabal, P., 2005. Determinaciones de is�otopos estables en restos humanos de laregi�on Cant�abrica. Aportaci�on al estudio de la dieta de las poblaciones delMesolitico y el Neolitico. Munibe. Antropol. Arkeol. 57, 359e374.

Arias Cabal, P., Armendariz Guti�errez, �A., 1998. Aproximaci�on a la Edad del Bronceen la regi�on cant�abrica. In: F�abregas Valcarce, R. (Ed.), A Idade do Bronce enGalicia : novas perspectivas. Cadernos do Seminario de Sargadelos (Edici�os doCastro), Sada, A Coru~na, pp. 47e80.

Armendariz Guti�errez, �A., 1990. Las cuevas sepulcrales en el País Vasco. Munibe.Antropol. Arkeol. 42, 153e160.

Balasse, M., Tresset, A., 2002. Early weaning of Neolithic domestic cattle (Bercy,France) revealed by intra-tooth variation in nitrogen isotope ratios. J. Archaeol.Sci. 29, 853e859.

Bettencourt, A.M.S., 1999. A paisagem e o Homem na bacia do C�avado durante o II eo I mil�enio AC (PhD thesis). Universidade do Minho, Braga.

Bettencourt, A.M.S., 2001. O povoado da Santinha, Amares, Norte de Portugal, nosfinais da Idade do Bronze Braga. Col. Cadernos de Arqueología. Monografías,Braga, p. 12.

Bettencourt, A.M.S., 2008. Life an death in the Bronze Age of the NW of IberianPeninsula. In: Fahlander, F., Oestigaard, T. (Eds.), The Materiality of Death:Bodies, Burials, Beliefs, BAR International Series 1768, pp. 99e104. Oxford.

Bettencourt, A.M.S., 2011. Estruturas e pr�aticas funer�arias do Bronze Inicial e M�ediodo Noroeste Peninsular. In: Bueno, P., Gilman, A., Martín Morales, C., S�anchez-Palencia, F.J. (Eds.), Arqueología, Sociedad, Territorio y Paisaje. CSIC, Madrid,pp. 115e139.

Bocherens, H., Drucker, D., 2003. Trophic level isotopic enrichment of Carbon andNitrogen in bone collagen: case studies from recent and ancient terrestrialecosystems. Int. J. Osteoarchaeol 13, 46e53.

Botella, M.C., Aleman, I., Jim�enez, S., 1999. Los huesos humanos. Manipulaci�on yalteraciones. Bellaterra, Barcelona.

Britton, K., Muldner, G., Bell, M., 2008. Stable isotope evidence for salt-marshgrazing in the Bronze Age Severn Estuary, UK: implications for palaeodietaryanalysis at coastal sites. J. Archaeol. Sci. 35, 2111e2118.

Buikstra, J.E., Ubelaker, D.H., 1994. Standards for data collection from human skel-etal remains. In: Proceedings of a Seminar at the Field Museum of NaturalHistory. Arkansas Archeological Survey Research Series. Fayetteville, Ark.

Bux�o, R., Piqu�e, R., 2008. Arqueobot�anica: los usos de las plantas en la peninsulaIb�erica. Ariel prehistoria, Barcelona.

Carballo Arceo, X., 2006. o millo miudo arqueol�oxico, Lingua do corvo. Asociaci�onCultural de Meiro, Bueu, 4, 10e11.

Collins, M.J., Galley, P., 1998. Towards an optimal method of archaeological collagenextraction: the influence of pH and grinding. Anc. Biomol. 2, 209e222.

Etxeberria Gabilondo, F., Herrasti Maci�a, L., 2007. Los restos humanos del enter-ramiento de SJAPL: caracterizaci�on de la muestra, tafonomía, paleodemografíapaleopatología. In: Vegas Aramburu, J.I. (Ed.), San Juan ante Portam Latinam.Una inhumaci�on colectiva prehist�orica en el valle medio del Ebro. Fundaci�onJose Miguel Barandiar�an y Diputaci�on Foral de Navarra, pp. 159e282.

Fern�andez Rodríguez, C., Ramil Rego, P., 1995. Cat�alogo y revisi�on crítica de lascolecciones faunísticas del Museo Provincial de Lugo. Bol. Mus. Pro. Lugo 7,189e218.

Fern�andez Rodríguez, C., Villar Quinteiro, R., 2003. Prospecci�on y excavaci�on decuevas en la cuenca del Sil (Rubi�a, Ourense): La Pala da Vella. Brigantium 14,13e22.

Fern�andez Rodríguez, C., Villar Quinteiro, R., Varela, P., Rey, J.M., Elorza, M., 1996.Primeros datos cronol�ogicos y paleontol�ogicos del yacimiento de Pala da Vella(Biobra, Ourense). In: Ramil-Rego, P., Fern�andez Rodríguez, C., RodríguezGuiti�an, M. (Eds.), Biogeografía pleistocena-holocena de la Península Ib�erica.Consellería de Cultura, Santiago de Compostela, pp. 249e260.

Fraser, R.A., Bogaarda, A., Sch€aferb, M., Arbogastc, R., Heaton, T.H.E., 2013. Inte-grating botanical, faunal and human stable carbon and nitrogen isotope valuesto reconstruct land use and palaeodiet at LBK Vaihingen an der Enz, Baden-Württemberg. World Archaeol. 45, 492e517.

Fuller, B.T., Müldner, G., Van Neer, W., Ervynck, A., Richards, M.P., 2012. Carbon andnitrogen stable isotope ratio analysis of freshwater, brackish and marine fishfrom Belgian archaeological sites (1st and 2nd millennium AD). J. Anal. A. T.Spectrom. 27, 807e820.

García Puchol, O., Cotino Villa, F., Miret Estruch, C., Pascual Benito, J.L., Mcclure, S.B.,Molina Balaguer, L., Alapont Martín, L., Carri�on Marco, Y., Morales, J.V., BlascoSenabre, J., Culleton, B., 2010. Cavidades de uso funeario durante el Neoliticofinal/Calcolitico en el territorio valenciano: trabajos arqueologicos en Avenc

O. L�opez-Costas et al. / Journal of Archaeological Science 55 (2015) 209e218218

dels Dos Forats o Cova del Monedero (Carcaixent, Valencia) Archivo de Pre-historia Levantina, vol. XXVIII, pp. 139e206.

Hedges, R.E.M., 2003. On bone collagendapatite-carbonate isotopic relationships.Int. J. Osteoarchaeol 13, 66e79.

Hedges, R., Reynard, L., 2007. Nitrogen isotopes and the trophic level of humans inarchaeology. J. Archaeol. Sci. 34, 1240e1251.

James, T.K., Rahman, A., McGill, C.R., Trivedi, P.D., 2011. Biology and survival ofbroom corn millet (Panicum miliaceum) seed. N. Z. Plant Prot. 64, 142e148.

Jim�enez-Brobeil, S.A., du Souich, P., Al Oumaoui, I., 2009. Possible relationship ofcranial traumatic injuries with violence in the south-east Iberian Peninsulafrom the Neolithic to the Bronze Age. Am. J. Phys. Anthropol. 140, 465e475.

Knüsel, C., 2002. Bone adaptation and its relationship to physical activity in thepast. In: Cox, M., Mays, S. (Eds.), Human Osteology in Archaeology and ForensicScience. Cambridge University Press, Cambridge, pp. 381e401.

Larsen, C.S., 2002. Bioarchaeology: the lives and lifestyles of past people. J. Archaeol.Res. 10, 119e166.

Lightfoot, E., Liu, X.Y., Jones, M.K., 2013. Why move starchy cereals? A review of theisotopic evidence for prehistoric millet consumption across Eurasia. WorldArchaeol. 45, 574e623.

Lillios, K.T., Waterman, A.J., Artz, J.A., Josephs, R.L., 2010. The Neolithic-Early BronzeAge mortuary rockshelter of Bolores, Torres Vedras, Portugal. J. Field Archaeol.35, 19e39.

Longin, R., 1971. New method of collagen extraction for radiocarbon dating. Nature230, 241e242.

Lopez-Costas, O., 2008. O estudo antropol�oxico dos restos �oseos recuperados danecr�opole da Cova do Santo, Pardoll�an, Rubi�a, Ourense. Rev. Real Acad. Cienc.Galicia XXVII, 131e144.

L�opez-Costas, O., 2012. Antropología de los restos �oseos humanos de Galicia:estudio de la poblaci�on romano y medieval gallega (PhD thesis). University ofGranada, Granada, p. 555.

Lopez-Costas, O., Rissech, C., Trancho, G., Turbon, D., 2012. Postnatal ontogenesis ofthe tibia. Implications for age and sex estimation. Forensic Sci. Int. 214.

Makarewicz, C., Tuross, N., 2006. Foddering by Mongolian pastoralists is recorded inthe stable carbon (d13C) and nitrogen (d15N) isotopes of caprine dentinalcollagen. J. Archaeol. Sci. 33, 862e870.

M�arquez-Grant, N., Rissech, C., Lopez-Costas, O., Alem�an, I., Caro Dob�on, L., 2011.Spain/Espa~na. In: M�arquez-Grant, N., Fibiger, L. (Eds.), The Routledge Handbookof Archaeological Human Remains and Legislation: an International Guide toLaws and Practice in the Excavation, Study and Treatment of ArchaeologicalHuman Remains. Routledge, Taylor and Francis group, London, pp. 423e438.

McClure, S.B., García, O., Roca de Togores, C., Culleton, B.J., Kennett, D.J., 2011.Osteological and paleodietary investigation of burials from Cova de la Pastora,Alicante, Spain. J. Archaeol. Sci. 38, 420e428.

McKinley, J.I., 2004. Compiling a skeletal inventory: disarticulated and co-mingledremains. In: Brickley, M., McKinley, J.I. (Eds.), Guidelines to the Standards forRecording Human Remains, BABAO. Dept. of Archaeology, University ofSouthampton. Institute of Field Archaeologists, University of Reading, pp. 14e17.

M�endez Fern�andez, F., 2006. Prospecci�on con toma de mostras no xacemento Covado Santo (Pardoll�an, Rubi�a, Ourense). Proyecto de actuaci�on arqueol�ogica. Un-published. Original deposited in Consellería de Cultura e Deporte. Direcci�onXeral de Patrimonio. Xunta de Galicia.

M�endez Fern�andez, F., Martínez L�opez, M.C., 2009. Prospecci�on e toma de mostrasno xacemento Cova do Santo (Pardoll�an, Rubi�a, Ourense). Memoria t�ecnica dostraballos. Unpublished report. Original deposited in Consellería de Cultura eDeporte. Direcci�on Xeral de Patrimonio. Xunta de Galicia.

Miquel-Feucht, M.J., Polo-Cerd�a, M., Villalain-Blanco, J.D., 2001. El síndrome criboso:criba femoral vs. criba orbitaria. In: S�anchez, J.A. (Ed.), Actas del V CongresoNacional de Paleopatología, Asociaci�on Espa~nola de Paleopatología. Ayunta-miento de Alcal�a la Real y Universidad Complutense de Madrid, Alcal�a la Real,Ja�en, pp. 221e237.

N�ajera Colino, T., Molina Gonz�alez, F., Jim�enez-Brobeil, S.A., S�anchez Romero, M., AlOumaoui, I., Aranda Jim�enez, G., Delgado-Huertas, A., Laffranchi, Z., 2010. Lapoblaci�on infantil de la Motilla del Azuer: Un estudio bioarqueol�ogico. Com-plutum 21, 69e102.

O'Leary, M.H., 1988. Carbon isotopes in photosynthesis. Bioscience 38, 328e336.Onta~n�on Peredo, R., Armendariz Guti�errez, �A., 2005. Cuevas y megalitos: los con-

textos sepulcrales colectivos en la Prehistorica reciente cant�abrica. Munibe.Antropol. Arkeol. 57, 275e286.

P�erez Ortiz, L., Fern�andez Rodríguez, C., 2005. El conjunto cer�amico del NeolíticoFinal de Pala da Vella (Rubi�a, Ourense). In: Arias Cabal, P., Onta~n�on Peredo, R.,García-Monc�o Pi~neiro, C. (Eds.), Actas del III Congreso del Neolítico en laPenínsula Ib�erica. Santander, pp. 327e335.

Rachie, K.O., 1975. The Millets: Importance, Utilisation and Outlook. InternationalCrops Research Institute for the Semi-arid Tropics, Hyderabad.

Rissech, C., Malgosa, A., 2005. Ilium growth study: applicability in sex and agediagnosis. Forensic Sci. Int. 147, 165e174.

Rissech, C., Schaefer, M., Malgosa, A., 2008. Development of thefemurdImplications for age and sex determination. Forensic Sci. Int. 180, 1e9.

Roberts, C.A., 2009. Human Remains in Archaeology: a Handbook. Council forBritish Archaeology, York.

R€osch, M., 1998. The history of crops and crop weeds in south-western Germanyfrom the Neolithic period to modern times, as shown by archaeobotanical ev-idence. Veg. Hist. Archaeobot 7, 109e125.

Ruiz-G�alvez Priego, M.L., 1998. La Europa atl�antica en la Edad del Bronce: un viaje alas raíces de la Europa occidental. Crítica, D.L., Barcelona.

Salazar-García, D.C., Benítez de Lugo Enrich, L.B., �Alvarez García, H.J., BenitoS�anchez, M., 2013. Estudio diacr�onico de la dieta de los pobladores antiguos deTerrinches (Ciudad Real) a partir del an�alisis de is�otopos estables sobre restos�oseos humanos. Rev. Esp. Antropol. Fis. 34, 6e14.

Scheuer, L., Black, S., 2000. Developmental Juvenile Osteology. Academic Press, SanDiego, New York, Tokyo.

Schoeninger, M.J., DeNiro, M.J., 1984. Nitrogen and carbon isotopic composition ofbone collagen from marine and terrestrial animals. Geochim. Cosmochim. Acta48, 625e639.

Sherratt, A., 1980. Water, soil and seasonality in early cereal cultivation. WorldArchaeol. 11, 313e330.

Su�arez Otero, J., 2014. Vaso de la edad del Bronce de Pardoll�an (Valdeorras), Peza domes. Museo Arqueol�oxico Provincial de Ourense Xaneiro, pp. 1e4.

Tafuri, M.A., Craig, O.E., Canci, A., 2009. Stable isotope evidence for the consumptionof millet and other plants in Bronze Age Italy. Am. J. Phys. Anthropol. 139,146e153.

Teira Bri�on, A., 2010. Tierra, metal y semillas. Consideraciones de la agricultura de laEdad del Hierro en Galicia. In: Bettencourt, A.M.d.S., Caetano Alves, M.I.,Monteiro-Rodrigues, S. (Eds.), Alteraç~oes Ambientais e Interacç~ao Humana naFachada Atlantica Occidental. Universidade do Minho, Braga, pp. 133e148.

Tereso, J.P., 2012. Environmental Change, Agricultural Development and SocialTrends in NW Iberia from the Late Prehistory to the Late Antiquity (PhD thesis).Facultade de Ciencias do Porto, Porto.

Tereso, J.P., Ramil-Rego, P., �Alvarez Gonz�alez, Y., L�opez Gonz�alez, L., Almeida-da-Silva, R., 2013. Massive storage in As Laias/O Castelo (Ourense, NW Spain) fromthe Late Bronze Age/Iron Age transition to the Roman period: a palae-oethnobotanical approach. J. Archaeol. Sci. 40, 3865e3877.

Trancho, G.J., Robledo, B., L�opez-Bueis, I., Sanchez, J.A., 1997. Sexual determinationof the femur using discriminant functions. Analysis of a Spanish population ofknown sex and age. Am. J. Phys. Anthropol. 42, 181e185.

Vaquero Rodríguez, M., Alonso Fern�andez, S., Alonso Fern�andez, C., AmeijendaIglesias, A., Blain, H.-A., F�abregas Valcarce, R., G�omez Merino, G., LomberaHermida, A.d., L�opez García, J.M., Lorenzo Merino, C., Lozano Ruiz, M., RodríguezRell�an, C., Rosell Ard�evol, J., Serna Gonz�alez, M.R., 2009. Nuevas fechas radio-m�etricas para la Prehistoria del noroeste de la Península Ib�erica: la cueva deValdavara (Becerrea, Lugo). Trab. Prehis 66, 99e113.

Vaquero-Rodríguez, M., Alonso-Fern�andez, S., Ameijenda-Iglesias, A., 2011.Archaeological excavations in the Becerre�a sites (eastern Lugo): Valdavara caveand Valdavara 3. In: Lombera Hermida, A.d., F�abregas Valcarce, R. (Eds.), To theWest of Spanish Cantabria. The Paleolithic Settlement of Galicia, BAR Interna-tional Series 2283, pp. 133e143. Oxford.

Vazquez Varela, J.M., 1994. El cultivo del mijo, (Panicum miliaceum, L.), en la culturacastre~na del Noroeste de la península Ib�erica. Cuad. Estud. Gallegos XLI (106),65e73.

Vogel, J.C., van der Merwe, N.J., 1977. Isotopic evidence for early maize cultivation inNew-York State. Am. Antiq. 42, 238e242.

Weber, S.A., Fuller, D.Q., 2007. Millets and their role in early agriculture. Pragdhara18, 69e90.

Weiss, E., Jurmain, R., 2007. Osteoarthritis revisited: a contemporary review ofaetiology. Int. J. Osteoarchaeol 17, 437e450.

Wood, J.W., Milner, G.R., Harpending, H.C., Weiss, K.M., Cohen, M.N., Eisenberg, L.E.,Hutchinson, D.L., Jankauskas, R., �Cesnys, G., Katzenberg, M.A., Lukacs, J.R.,McGrath, J.W., Roth, E.A., Ubelaker, D.H., Wilkinson, R.G., 1992. The osteologicalparadox: problems of inferring prehistoric health from skeletal samples (andcomments and reply. Curr. Anthropol. 33, 343e370.

Zavodny, E., McClure, S.B., Culleton, B.J., Podrug, E., Kennett, D.J., 2014. Neolithicanimal management practices and stable isotope studies in the Adriatic. Envi-ron. Archaeol. 19, 184e195.