at SciVerse ScienceDirect

Journal of Archaeological Science 40 (2013) 3865e3877

Contents lists available

Journal of Archaeological Science

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

Massive storage in As Laias/O Castelo (Ourense, NW Spain) from theLate Bronze Age/Iron Age transition to the Roman period: apalaeoethnobotanical approach

João Pedro Tereso a,*, Pablo Ramil-Rego b, Yolanda Álvarez González c,1,Luis López González c,1, Rubim Almeida-da-Silva d

aCIBIO (Research Center In Biodiversity and Genetic Resources), Universidade do Porto, Faculdade de Ciências, Departamento de Biologia, Edifício FC4,Sala 1A, Rua do Campo Alegre, s/n, 4169-007 Porto, Portugalb IBADER (Instituto de Biodiversidade Agraria e Desenvolvemento Rural), GI-1934, Territorio-Biodiversidade, Lab Botánica & Bioxeografía,Campus Universitario s/n, E-27002 Lugo, Galicia, Spainc Terra-Arqueos SL, Julio Prieto Nespereira N� 40, 32002 Ourense, SpaindUniversity of Porto, Dept. of Biology & CIBIO, Edif. FC4, R. do Campo Alegre, s/n� , 4169-007 Porto, Portugal

a r t i c l e i n f o

Article history:Received 7 May 2012Received in revised form28 April 2013Accepted 4 May 2013

Keywords:StorageIron AgeRoman periodNorthwest Iberia

* Corresponding author. Tel.: þ351 967908314.E-mail addresses: jptereso@gmail.com (J.P. Te

(P. Ramil-Rego), rubim.almeida@fc.up.pt (R. Almeida-1 (www.terraarqueos.com).

0305-4403/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.jas.2013.05.007

a b s t r a c t

The highest platforms of As Laias, a settlement located in northwest Iberia, were used for storage duringthe Iron Age and until the turn of the Era. In this time-span, the whole hilltop e the croa e was a walledenclosure mostly used for storage and due to recurrent fires affecting the settlement’s terraces, abundantplant remains were preserved in overlying wattle and daub storage structures.

Triticum aestivum subsp. spelta was the predominant crop stored in As Laias and was often stored inspikelets. Panicum miliaceum and Vicia fabawere important crops too and the presence of Quercus acornstestifies for its storage together with cereals and legumes. Evidence suggests a twofold situation occurredsince, in some structures, multiple species were stored while others were possibly filled with a singlecrop.

The evidence of massive storage, the abundance of charred plant remains and the uniqueness of thestorage structures makes As Laias one of the most important Iron Age sites in the whole NW Iberia. Theinterpretation of the site, together with other archaeological contexts in the region is controversial andcan have major consequences for the interpretative models of the social and economic features of thelocal Iron Age. The site itself is sufficient to testify for the productivity of Iron Age agriculture, but furtherconsiderations must include remarks regarding the social significance of a massive storage place.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

As Laias, or O Castelo, is located near Ourense, in the Munici-pality of Cenlle (Fig. 1). It is positioned in a spur, 240 m high, in theright margin of the river Minho. This position offers a major visualcontrol over a great extension of the river course.

The site was known to Lopez Cuevillas in the 1920’s and wasexcavated by Chamoso Lamas in 1956 who considered the site to berelated to mining activities in Roman times due to the presence of

reso), ramil.rego@gmail.comda-Silva).

All rights reserved.

artifacts presumably used in such activities and due to the presenceof a nearby gold mine, which existence was confirmed recently(Álvarez González and López González, 2000).

But it was only in 1997 that the site was again excavated. Thearchaeological intervention was carried out in the context of theconstruction of the highway A-52 (Fig. 2) and was directed by Y.Alvarez González and L. López González. Regardless of the rele-vance and uniqueness of the site, it was decided that it would bealmost completely destroyed without a complete excavation of theaffected area. Moreover, despite the abundance of archaeobotanicalremains (one of the main features of this site), its complete studywas not included in the mitigation or compensation measures.

This excavation was centered mainly on the higher part of thesettlement and allowed to define a hilltop fortified areae the croaewhich assumed rare characteristics in the proto-history of the

Fig. 1. Location of As Laias in northwest Iberia.

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e38773866

northwest Iberia. That fortified areawas used for storage during theIron Age and the abundant storage structures found did not have, atthe time of the excavation, any parallel in the region. Many werefound with part of their contents (Álvarez González and LópezGonzález, 2000).

Fig. 2. Destruction of As Laias during the

Despite the stratigraphic problems inherent to the site’s char-acteristics and the fact that there were significant problems con-cerning the sampling strategy applied, the study of As Laias iscrucial to understand the Iron Age in all northwest Iberia. In thiscontext, this study intends to optimize problematic data in order to

construction of the highway A-52.

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e3877 3867

get a general perspective on the crops whichwere stored in As Laiasand to understand the storage strategies and their relevance on aregional level.

2. NW Iberia from the Late Bronze Age to the Romanconquest

In the Late Bronze Age (1200e500/400 cal BC) settlementpattern is significantly diverse. While some settlements are posi-tioned in strategic positions, in hills or mid-altitude spurs con-trolling important river basins, others are closer to the valleys(Bettencourt, 1999, 2009). Although stone walls, embankments orwooden palisades are found in some of these settlements, mon-umentalization is still rare. This periodmarks a turning point on thelandscape level. As populations became more sedentary and con-nectivity between settlements was enhanced, an unprecedentedphase of deforestation began (Bettencourt, 1999, 2009; MuñozSobrino et al., 1997; Ramil Rego et al., 1998).

Major changes occurred somewhere between 600 and400 cal BC, marking the beginning of the Iron Age. Hillfortse locallyknown as “castros”e became themain type of settlement. Most arelocated in defensible locations and some present stonewalls and/orother kind of monumental structures. Some of these hillforts werefounded in the Late Bronze Age but were enlarged and reorganizedin this period. Many were founded in the Iron Age, now with stoneas the main building material (Martins, 1996; Parcero Oubiña andCobas Fernández, 2004). The overall familiarity between settle-ments in a vast region in northern Portugal and Galicia (northwestSpain) led to generalization of the term “Castros culture”. Still, thisdesignation mistakenly homogenizes a far more complex anddiverse reality (González-Ruibal, 2003; Martins, 1993e1994).

In the 2nd century cal BC several settlements were abandonedand others grew significantly in size and adopted proto-urbanmodels. Many fortifications were renewed (Martins, 1996; PeñaSantos, 2005). The chronology of this territorial organization ap-pears to be related to the first Roman incursions in the region e

D. Iunius Brutus’ campaign of 138e137 BC, reached the river Minhowith little resistance. Some authors suggest this change was as a

Fig. 3. View of the upper areas of As Laias. Arrows show the entrance p

defensive reaction and a prevention regarding future incursions,implying, thus, that the Romans did not get effective control overthe area with this first expedition (Alarcão, 1992). Peña Santos(2005) admits otherwise; that the Atlantic region between theriver Douro andMinho became part of the Roman Empire after thatincursion.

Nevertheless, Roman presence in western Iberia dates back tothe beginning of the 2nd century BC (Fabião, 1992). Therefore, localcommunities were in contact with this foreign influence before anymilitary intervention in the northwest.

During the 1st century BC other expeditions occurred mostly torecognize and obtain mineral resources. C. Iulius Caesar reachedBrigantium (A Coruña) in 61e60 BC (Morais, 2004) and the Canta-brian wars which finished the conquest of the Iberian peninsula in29e19 BC, during Augustus’ reign, took place mostly outside the“Castros” culture area (Peña Santos, 2005). It is not clear, thus,when the region became effectively controlled by the Romans, butarchaeological data suggests that in the turn of the Era, althoughthe region was politically part of the Empire and the new Romanadministration was settled, the hillforts maintained with littledifferences their Iron Age ways of living (Parcero Oubiña and CobasFernández, 2004).

3. General characteristics and chronology of As Laias

As Laias is a settlement with over 10 ha. It can be divided inthree main areas: the croa (the enclosure in the hilltop), the abruptslope toward the river and a plane area in the bottom of the slope,near the river.

The hilltop was fully delimited by an approximately circularstone wall with a single entrance and this enclosure was isolatedfrom the rest of the settlement by an area with no constructions(Álvarez González et al., 2009) (Fig. 3). Within the wall, em-bankments were built between the abundant rocky outcrops. Theplatforms created by these embankments were used for storageand no habitations were found. Most storage facilities wererectangular structures (c. 1 m � 1.5 m) built above the groundwith walls made of small intertwined branches and clay e wattle

ath to the fortified croa where the storage structures were found.

Fig. 4. Perspective with two storage structures. Scale: 1 m

Fig. 5. Storage structure. Scale: 1 m

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e38773868

and daub; the bottom was constituted by two distinct levels, afirst level of clay above which wood boards (sometimes cork)were placed (Álvarez González and López González, 2000)(Figs. 4 and 5). The boards were made of deciduous oak (Quercus)wood and the wattle was made with branches from severalspecies: Arbutus unedo, Erica, Leguminosae (considering thelocation of the site it is probably Cytisus), Quercus suber and Salix(Carrión Marco, 2003, 2005). The diameter of those branches wasmainly comprised between 5 and 10 mm (the shrubby species)and 11e15 mm (Q. suber) (Carrión Marco, 2003). The cover of thestructures is not known. It is possible that some of the clay withplants imprints, or some of the abundant charcoal are notderived from the walls but rather from the collapse of a cover ofany kind.

Only two poorly preserved circular storage pits were found aswell as two small circular stone constructions (Álvarez Gonzálezand López González, 2000). The latter were also interpreted asstorage structures because of their small size and fine stone pave-ments. One of these structures provided few carpological remainsand no other archaeological remains were found to suggest otherinterpretation. Still, both structures provided few data, demandingcaution in their interpretation.

The abrupt sloppy area between the croa and the bottom of thehill was vaster than the croa itself. This sloppy area seems to havebeen intensively occupied and it was characterized by thedissemination of stone embankments which created small hori-zontal areas where huts and water canals were built. Most hutswere built with branches and clay e these, as well as, postholes,were identified. In the excavated area, only two stone houses werefound. Some of the houses had a central hearth. In this area, a greatamount of grindstones were found (Álvarez González and LópezGonzález, 2000).

In the lowest part of the settlement artifacts were recoveredalong the agricultural fields as well as a huge hydraulic infra-structure (Álvarez González and López González, 2000).

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e3877 3869

Radiocarbon dates from As Laias (see Table 1 and Fig. 6)demonstrate the settlement was founded during the Late BronzeAge, probably around 800 cal BC. A hiatus preceded the majoroccupation phase of the site, during the Iron Age (from 500/400 cal BC until the end of the 1st century cal BC). The Romanpresence is residual, but data demonstrates the site remainedoccupied until at least the 3rd century AD (Álvarez González andLópez González, 2000). The Roman coins found in a destructionlevel of the wall testify for such late chronology (Martínez Mira,2004e2005). According to Álvarez González et al. (2009), thesloppy area was thoroughly abandoned since the turn of the Era aspeople moved to the lower platform, near the river.

The top enclosure e the croa e seems to have been used inten-sively for storage during the Iron Age. Most radiocarbon dates pointto the 5the2nd centuries cal BC, although earlier dates of the 8the5th centuries cal BC and later dates of 1st century cal BC e 1stcentury cal AD and even of the 2nde3rd centuries cal AD, extendconsiderably the time span of this elevation’s occupation, beingunclear if it was used for storage in the last phase (Álvarez Gonzálezand López González, 2000; Carrión Marco, 2003). In what thedefensive wall is concern, Álvarez González and López González(2000) suggest it was built c. 2nde1st century cal BC and that pre-viously another kind of structure, perhaps a palisade existed. Thissupposition is based on the fact that the inner area was previouslyused exclusively for storage, suggesting some kind of constructionother than a stonewallwould delimit such space. Still, a radiocarbondate froma structuredirectly belowapart of thewall (CSIC-1274, seeTable 1) questions this first interpretation, thus being possible thatthe wall was built in the beginning of the Iron Age.

Although the data available is sufficient to build this generalscenario, some difficulties arise when trying to understand in detailthe dynamics within the settlement along this time-span of over1000 years. This happens due to the structure of the settlement, i.e.the successive platforms supported by stone embankments. In thewell preserved platforms, several occupation levels were found.However, when these embankments were destroyed after theabandonment of the site, some layers were mixed.

In addition, the croa, having been used for storage, did notprovide many artifacts. In fact, these are rare, making it difficult toinfer chronologies which can only be attended by radiocarbondates. Still, the existence of multiples terraces, each with a specificstratigraphy thus with a specific story, made it difficult (most of the

Table 1Radiocarbon dates from As Laias. Original data from CarriónMarco (2003), recalibrated uscharcoal. Legend: O e outside the upper enclosure; I e inside the upper enclosure.

Sample Lab. Ref. Context

LAI 11-97-48 CSIC-1271 O Beneath the wall of a compartment, below LAIseparated by a hiatus

LAI 13-97-24 CSIC-1272 O Under the floor of a houseLAI 23-97-47 CSIC-1277 I Below the wall of a small circular stone structuLAI 14-97-19 CSIC-1273 O Occupation level of a quadrangular houseLAI 20-97-17 CSIC-1274 O Destruction level of a wattle and daub storage sLAI 11-97-47 CSIC-1270 O Beneath the wall of a compartment, above LAI

separated by a hiatusLAI 27-97-33 CSIC-1396 I Boards from a wattle and daub storage structurLAI 23-97-28 CSIC-1276 I Boards from a wattle and daub storage structurLAI Z21-97-30 CSIC-1275 O Path toward the entrance of the upper enclosurLAI 25-97-18 CSIC-1394 O Next to the wall of the upper enclosure, bellowLAI 1-97-49 CSIC-1269 O Filling of a terraceLAI 27-97-55 CSIC-1397 I Destruction level of a wattle and daub storage sLAI 29-97-225 CSIC-1399 I Destruction level of a wattle and daub storage sLAI 29-97-259 CSIC-1401 I Associated with a wattle and daub storage struLAI 25-97-30 CSIC-1395 O Next to the wall of the upper enclosure, aboveLAI 29-97-45 CSIC-1402 I Destruction level of a wattle and daub storage sLAI 29-97-107 CSIC-1400 I Boards from a wattle and daub storage structurLAI 33-97-5 CSIC-1398 I Wood post

times, impossible) to deduce the chronology of the storage facilitieswhich were not directly dated. It was not possible to get radio-carbon dates from all the structures, since thesewere too abundant.In the better preserved platforms it was possible to verify thatstorage facilities were built over each other in at least foursequential phases. Fires must have been frequent since charredplant remains appear abundantly in all phases.

4. Materials and methods

This study will focus on the storage structures from the walledenclosure on the top of the hill e the croa e and one structureoutside the walled area.

As was previously described, storage facilities were built inplatforms supported by stone embankments but the destruction ofthese embankments mixed some of the archaeological levels aswell as the storage structures’ content. This made it difficult toknow the exact number of storage facilities that existed in theexcavated area. For the same reasons, the reliable contexts for anarchaeobotanical study were limited. As was stated before, radio-carbon dating was necessary to know each context’s chronologyand special care was needed when selecting contexts to be studiedsince most structures were partially displaced, due to thedestruction of the embankments. Some undated structures placednear others with radiocarbon dates, in well preserved platformswere considered reliable.

Over 100 samples were collected and processed in the field bythe archaeological team in 1997. Still the sampling strategy was notappropriate and some relevant information is lacking.

The sediment was sieved using threemeshes, of 0.8mm, 0.4mmand 0.2 mm and then floated. Unfortunately, there is no informa-tion concerning the original volume of the structures’ content.Additionally, the collecting of samples inside each structure wasdone randomly making it difficult to understand the distribution ofthe different species inside the structure and, in some cases, thereal proportion between species.

Still, although these factors pose some limitations to thisarchaeobotanical study, the uniqueness of this site makes thisapproach indispensable.

Six storage facilities (STO) were selected to be studied. Five ofwhich were positioned within the wall and one was partiallycovered by it (see Fig. 7):

ing Calib. Rev 6.0.1, IntCal09 calibration curve (Reimer et al., 2009). Dates obtained on

14C dates (BP) Calibrated 1s Calibrated 2s

11-97-47, 2631 � 51 841e771 cal BC 912e593 cal BC

2604 � 35 809e777 cal BC 835e597 cal BCre 2592 � 40 810e674 cal BC 831e554 cal BC

2530 � 35 789e564 cal BC 797e539 cal BCtructure 2435 � 32 724e413 cal BC 751e405 cal BC11-97-48, 2398 � 39 519e401 cal BC 748e392 cal BC

e 2370 � 27 503e394 cal BC 519e390 cal BCe 2325 � 39 413e262 cal BC 515e231 cal BCe 2280 � 40 398e234 cal BC 403e208 cal BCLAI 25-97-30 2269 � 26 393e236 cal BC 397e210 cal BC

2254 � 39 389e215 cal BC 396e205 cal BCtructure 2223 � 26 364e210 cal BC 382e204 cal BCtructure 2208 � 26 358e206 cal BC 369e201 cal BCcture 2188 � 26 354e199 cal BC 361e177 cal BCLAI 25-97-18 2083 � 26 156e53 cal BC 180e41 cal BCtructure 2033 � 26 88 cal BCe4 cal AD 155 cal BCe49 cal ADe 1884 � 26 72e136 cal AD 67e215 cal AD

1845 � 26 130e214 cal AD 86e238 cal AD

Fig. 6. Radiocarbon dates from As Laias, probability plot (original data on Table 1).

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e38773870

STO.1 (LAI C20-17) e Partially destroyed storage facility, in partcovered by the defensive wall. In fact, the stratigraphic relationwith the wall justifies the selection of this sample. Branches andclay testify for the typology of the structure and a radiocarbon date

Fig. 7. General plan of As Laias indicating the

points out to a transition phase between the Late Bronze Age andthe Iron Age. Unfortunately, the date is not precise, covering a greattime span (CSIC-1274: 751e405 cal BC, at 2s).

STO.2 (LAI C27-30/32/33) e Two soil samples were collected inthis structure. One (LAI C27-30) corresponds to a dark sedimentdispersed inside the structure and the other (LAI C27-32) wascollected in the northern limit of the structure, it was a reddishcompact layer. These two levels covered wooden boards (LAI C27-33) in the base of the structure. A sample from these boards pro-vided a radiocarbon date from the beginning of the Iron Age (CSIC-1396: 519e390 cal BC, at 2s).

STO.3 (LAI C29-121/172/225) e Two soil samples from succes-sive levels of this structure were collected (LAI C29-121 above C29-172). The last sample (LAI C29-225) corresponds to cork pieces inthe base of the structure, which were radiocarbon dated testifyingits Iron Age chronology (CSIC-1399: 369e201 cal BC, at 2s).

STO.4 (LAI C29-96) e A small sample from a structure placed inthe surroundings of the previous one (STO.3). It is not dated, but theproximity and the fact that it was at a similar altimetry than LAIC29-172 suggests a similar chronology.

STO.5 (LAI C29-45/66/81) e Four soil samples from successivelevels of the same structure. LAI C29-45 is the top level, C29-81 isthe second level, C29-66 the third and C29-157 the base with seedsand wooden boards. The first level represents the abandonment ofthe structure and it has a radiocarbon date from the turn of the Era,most probably from the second half of the 1st century BC (CSIC-1402: 155 cal BCe49 cal AD, at 2s).

position of the studied structures (STO.).

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e3877 3871

STO.6 (LAI C29-244)e A small sample from a structure placed inthe surroundings of the previous one (STO.5) in the exact samealtimetry as LAI C29-81. It is not dated, but it has probably the samechronology as STO.5.

These samples cover the chronology of the site from the LateBronze Age/Iron Age transition until the turn of the Era. Only thelast phase, the 2nde3rd centuries AD, is not included. Preliminaryidentifications were done in 99 samples and the structures selectedfor this study include all crops found until now in the wholesamples.

The laboratory work has taken place in the IBADER facilities inLugo, by Pablo Ramil Rego and Luis Gómez Orellana. The largersamples were sub-sampled with the spoon method and identifi-cationwas carried out in the stereoscopemicroscope, by comparingthe archaeobotanical material with atlases and the referencecollection. For the identification of cereal remains, the criteria ofHillman et al. (1996), Buxó (1997) and Jacomet (2006) were used.

A special word must be said regarding the identification ofwheat remains and faba bean.

Although sometimes problematic, the distinction betweenthe morphological types Triticum spelta and Triticum aestivum/durum, and between T. spelta and Triticum dicoccum was possible.Grains from T. spelta are bigger, slimmer and flatter than those ofT. aestivum/durum. They are evenly curved in the dorsal side andconcave or flattish-concave in the ventral side; the extremitiesare rarely pointed and most of the times blunted. Theyfrequently correspond to the typical grains of T. spelta asdescribed by S. Jacomet (2006), although some can be difficult todistinguish from T. dicoccum, whenever the grains are slightlypointed near the embryo. Still, chaff and the presence of spike-lets with grains, thoroughly confirmed the identification ofT. spelta.

Chaff was identified as T. spelta on the basis of two main fea-tures: the wide and thin section of the lower part of the glumes (aspointed out by Jacomet, 2006) and the presence of strong venationin the glumes (Buxó, 1997; Hillman et al., 1996; Jacomet, 2006). Inaddition most spikelet forks were massive and had part of the nexthigher rachis segment.

Table 2Results from the samples in As Laias. Legend: Cells in grey e contain aggregated grains; ceweight (g) in the sample; t e estimated weight (g) per 100 g. .

In what naked wheat is concern, the sparse and fragmentedrachis remains did not allow the distinction between species.Grains clearly of this morphological type were found in a singlesample and were identified on the basis of the following features:the grains are oval and they present the maximum width in thelower half or near the middle of the grain (Buxó, 1997). Generallyboth extremities are blunt. Contrasting with T. spelta grains of thesame sample, grains from T. aestivum/durum are not slim and theyare flattish-convex in the ventral side, having a characteristicswollen aspect.

Seeds from Vicia faba are usually short with oval-globular shape.Several designations have been used to name such remains (themost common are Vicia faba var. minor, Vicia faba var. minuta andVicia faba var. equina) (Dopazo Martinez, 1996; Pinto da Silva, 1988;Tereso, 2009), some synonymous, other not. Still, the differentia-tion between distinct varieties is not possible to achieve throughthe morphology of the seeds and, given the current knowledge, it isabusive to assume any correspondence between archaeologicalmaterial and actual varieties. Thus faba beans will be designatedsimply as V. faba.

5. Results

Results from the carpological study are presented in Table 2.Cereals are the main crops found inside the structures.

These include Avena (oat), H. vulgare var. nudum (naked barley),H. vulgare subsp. vulgare (hulled barley) (Fig. 8), Panicum mil-iaceum (common millet), T. aestivum/durum (naked wheat) andT. aestivum subsp. spelta (spelt) (Fig. 9). Besides the cereals, Pisumsativum (pea), V. faba (faba bean) and acorns from Quercus werealso found.

It is interesting to notice that H. vulgare var. nudum andT. aestivum/durum were only retrieved in the STO.1, the oldeststorage structure dated until now in As Laias. The radiocarbon datepoints out to the transition between the Late Bronze Age and theIron Age. This storage structure provided the more diverse content,including all the crops found in As Laias e six different cereals andtwo pulses (Table 2). Still, H. vulgare subsp. vulgare and V. faba are

lls with outer contour markedemain types of remains in the sample; T e estimated

Fig. 8. Grain of Hordeum vulgare subsp. vulgare.

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e38773872

clearly the predominant crops in STO.1, making this earliest struc-ture the more original of all of those presented here. On the otherhand, the similarity between the contents of Iron Age structuresand those from the turn of the Era suggest some stability andcontinuity may have existed in consistence with the general in-terpretations for these time-periods (vide supra).

T. spelta is the main crop in Iron Age structures and those fromthe turn of the Era, STO.2, STO.3, STO.5 and STO.6, followed byP. miliaceum in STO.2 and STO.5, and followed by V. faba in STO.3. InSTO.4 grains from P. miliaceum are the more abundant remains.Overall, spelt and common millet are the more frequent crops in

Fig. 9. Remains of Triticum aestivum subsp. spelta. Spikelet (above) and grain (below).

the analyzed structures but Avena, H. vulgare subsp. vulgare are alsopresent throughout the time span.

Spelt chaff appears in almost all structures. The exception isSTO.4 but a single small sample was retrieved there. The presenceand abundance of such remains is an important feature to help usinterpret storage strategies in As Laias (vide infra).

Garden crops are more frequent in the earliest structures. Fababean is even the more relevant crop in the STO.1 and the second inSTO.3. Peas are always rare. Regarding the presence of wild fruits,acorns from Quercus are the only remains. They were stored inSTO.2.

6. Discussion

6.1. Crops in As Laias: regional overview

As pointed out before, T. spelta is the main crop in most of thestorage facilities studied until now. However, the most relevant factis its presence in STO.1, dated from a transitional moment betweenthe Late Bronze Age and the Iron Age. Unfortunately, the date is notprecise, covering a great time span (CSIC-1274 e 751e405 cal BC, at2s) (Table 1 and Fig. 6). The position of STO.1 beneath the defensivewall suggests this to be earlier than a phase of major changes in thesettlement and the general perspective obtained from the availableradiocarbon dates and archaeological data suggests there is a hiatusin the site’s occupation, between the Late Bronze Age and thebeginning of the Iron Age (Álvarez González and López González,2000).

The presence of spelt grains and chaff in all levels of As Laiasconfirms the idea that this specie was cultivated way before theRoman period, as suggested by Buxó (1997, Buxó et al., 1997). Innorthwest Iberia it was already retrieved in the Iron Age levels ofCastrovite (A Estrada) (Rey Castiñeira et al., 2011) and Crasto dePalheiros (Murça) (Figueiral, 2008). The presence of spelt in theChalcolithic levels of Crasto de Palheiros is controversial because itis in contrast with the general data for the dissemination of thiscrop throughout Europe (Akeret, 2005). Significant disturbanceswere documented in the site, some originated by its Iron Ageoccupation, others are much more recent. In fact, most radiocarbondates from the Chalcolithic levels of the platform where spelt wasretrieved fall into the time intervals of the regional Iron Age, RomanPeriod and even Medieval Ages and were considered anomalous bythe leading archaeologist (Sanches, 2008). For now, remains of AsLaias are the oldest well contextualized spelt remains in thenorthwest, being among the oldest in all the Iberian Peninsula.

Overall, the relevance of hulled wheats in northwest Iberia sincethe end of the Late Bronze Age and during the Iron Age and Romanperiod is suggested by their thorough presence in other settle-ments, such as Penalba (Campolameiro) (Aira Rodriguez et al.,1990), Castrovite (A Estrada) (Rey Castiñeira et al., 2011), Castrelinde San Juan de Paluezas (Borrenes) (Lopez Merino et al., 2010),Crasto de Palheiros (Murça) (Figueiral, 2008) and Terronha dePinhovelo (Macedo de Cavaleiros) (Tereso, 2009), among others.Archaeobotanical data even suggest that hulled wheats becamecrops as relevant, or possibly more relevant, than naked wheats innorthwest Iberia during the Iron Age, in contrast with other Iberianregions (Tereso, 2012). In this sense, it is interesting to notice thatnaked wheats were only recovered in the oldest storage structure(STO.1), from the Late Bronze Age/Iron Age transition, but eventhere in less amounts than spelt wheat.

In the same way, the presence of Avena in STO.1 is interesting.Oat is considered to have been cultivated in the Iberian Peninsulaonly since the 5th century BC (Buxó, 1997; Buxó and Piqué, 2008).Oat grains in earlier contexts, since the Neolithic, are considered tobelong to wild species. Their thorough presence in As Laias

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e3877 3873

confirms oat was an important crop since, at least, the 5th centuryBC although the great time span of the available radiocarbon datedoes not exclude an earlier chronology.

Also during the Bronze Age, but in an earlier stage, began thecultivation of P. miliaceum. The more ancient remains in north-west Iberia are those of Sola (Braga), the only ones that date backto the Middle Bronze Age (Bettencourt, 1999; Bettencourt et al.,2007). Since the Late Bronze Age it became a staple crop andits increasing relevance throughout the Bronze Age may berelated with changes toward greater territorialization and sed-entarization (Bettencourt, 1999; Tereso, 2012). Its relevance in AsLaias is perfectly in consonance with the regional scenario nowavailable.

Barleys are important crops in As Laias as well, particularlyH. vulgare subsp. vulgare, a regular crop in proto-historic andRoman times in the region (Oliveira, 2000; Ramil Rego, 1993;Ramil Rego et al., 1996; Ramil Rego and Fernández Rodríguez,1999). H. vulgare var. nudum (naked barley) was retrieved onlyin the oldest sample e STO.1 e and even there it is residual.Having been an important crop in the Neolithic, naked barley lostits relevance in all Iberia since the Chalcolithic and the Bronze Age(Buxó and Piqué, 2008). The reasons for the decrease in thecultivation of naked barley are not yet understood (Buxó, 1997),neither their presence in few Late Bronze Age or Iron Age settle-ments, As Laias included. In northwest Iberia, naked barley wasdetected in the Late Bronze Age levels of Torroso (Mos) (DopazoMartínez et al., 1996) and the Iron Age levels of Cortegada (Sil-leda) (Arnanz and Chamorro, 1990) and Alto do Castro (Cuntis)(Cobas Fernández and Parcero Oubiña, 2006; Dopazo Martínezet al., 1996; Parcero Oubiña, 2000).

Regarding the pulses, they were stored in As Laias but mostcome from the oldest storage structures studied. It would benecessary to date more structures in order to know if such patternis not a result of any kind of investigation bias. It is not surprisingthat V. faba is the main pulse in As Laias, since it is the morefrequent garden crop in Proto-historic settlements in the region(Oliveira, 2000; Ramil Rego, 1993; Ramil Rego et al., 1996).

6.2. Storage strategies in As Laias

Even in the restricted studied storage facilities from the As Laias’croa, we can detect differences between structures. It is clear thatmost structures were used for more than one crop but it is possiblethat others were not. In STO.6 spelt is almost the only species andthe presence of a few grains of P. miliaceum are more likely theresult of previous uses of the same structure. The same may havehappened in STO.4, although that is not so clear.

The coexistence of several crops in the same structure is still themore frequent case in As Laias and it is regular in many othercontexts from diverse chronological and geographic realities (Buxóand Piqué, 2008). But it is not clear if the crops weremixed togetherin the structures, perishable materials could have been used toseparate different crops. Y. Carrión Marco (2003) mentions thepresence of a small wooden bowl and evidence of interweavedplant material, possibly some kind of basketry, in As Laias.

The fact is that the structures have crops that need differentprocessing (e.g. faba bean and cereals; millet and spelt withspikelets included).

Still, the available information is dubious. All structures pro-vided, in fact, more than one species and in most of them therewere at least two crops in significant amounts. On the other hand,in small sized samples (Table 2) there is usually a clear prepon-derance of a crop and secondary crops appear in very smallamounts, suggesting some kind of organization could have existedinside the structures.

Proper sampling could have provided relevant informationregarding the distribution of crops in the structures. Unfortunately,this question remains open.

A comparisonwith other sites in the region is difficult since clearstorage facilities are rarely found in Late Bronze Age or Iron Agehillforts (Parcero Oubiña and Cobas Fernández, 2004) and thosewith abundant crops are even rarer. Penalba (Aira Rodriguez et al.,1990; Dopazo Martínez et al., 1996), S. João de Rei (Bettencourt,1999; Oliveira, 2000), Crastoeiro (Pinto da Silva, 2001; Dinis,2001) and Castrovite (Carballo Arceo, 1998; Rey Castiñeira et al.,2011) are exceptions. In the latter site, abundant radiocarbondates from several levels suggest the area was consecutively usedfor storage between the 4th and the 1st centuries cal BC, not beingclear whether crops were stored separately or not (Rey Castiñeiraet al., 2011).

In the Late Bronze Age hillfort of Penalba abundant plant re-mains were found in huts, associated to the ceramic containerswhere they were stored (Aira Rodriguez et al., 1990; DopazoMartínez et al., 1996). It is not possible to know if such facilitieswere used for long term storage. Despite the almost absence ofreferences to the provenience of the plant remains the homoge-neity of the samples which were studied suggests different specieswere stored separately (Dopazo Martinez, 1996).

In the Iron Age storage-pits at S. João de Rei (Bettencourt, 1999;Oliveira, 2000) and Crastoeiro (Dinis,1993e94, 2001; Pinto da Silva,2001) several crops and even Quercus acorns were found stored inthe same structures. The storage pit of S. João de Rei, provided aradiocarbon date covering the 4th and 3rd centuries cal BC(Bettencourt, 1999). Triticum grains are the main plant remains,followed at a great distance by Panicum/Setaria grains. Other re-mains are occasional (Oliveira, 2000). Thus it is possible that onlyone crop, possibly two, was stored in the last use of the structure.The only storage pit of Crastoeiro that has been published in detailprovided two radiocarbon dates ranging from the beginning of the4th to the middle of the 2nd centuries cal BC (Dinis, 2001). Nakedwheat, emmer (T. dicoccum) and millet (P. miliaceum) are pre-dominant, but two grains of oat (Avena) and two grape pips (Vitis)were also found (Pinto da Silva, 2001).

It is likely that the deposition of crops in storage structuresdepended on the timing of its use, since experiments with storagepits demonstrated that sealing is crucial to allow preservation(Miret i Mestre, 2008). The abundance of storage facilities in AsLaias suggests long-term storage, although short term storage fordaily uses could have existed too. Moreover, the abundance of chaffand whole spikelets with grains demonstrates that spelt wasfrequently stored unprocessed, hulled, which also suggests long-term storage. In fact, it is well known that storage of whole spike-lets prevents deterioration by fungus and insects (Buxó and Piqué,2008).

Assuming that long-term storage of T. spelta occurred in As Laiasit is likely that the same happened with other species since speltwas stored with other crops and, as mentioned before, sealing iscrucial to assure preservation. The timing of the opening should besimilar for all species found inside each structure.

The storage of collected wild fruits, namely acorns, is not uniqueto As Laias in northwest Iberia. It was discovered in Penalba, storedin vessels (Aira Rodriguez et al., 1990; Dopazo Martinez, 1996).

As we have seen, in the several moments of destruction/aban-donment fire allowed the preservation of great amounts of plantremains in As Laias. This demonstrated that in the several episodesof fires some structures were almost emptywhile others had plentyof material (STO.1 provided 94.3 g of plant material per 100 g ofsediment). For this remark we count with the preliminary results ofsamples from other storage facilities, not dated, which providedalmost no crops or even no crops at all.

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e38773874

The existence of several episodes of fire is itself somethingwhich needs to be explained. These covered the whole time-spanhere studied, i.e. since the Late Bronze Age/Iron Age Transitionuntil the Transition of the Era and were not restricted to the storagearea. Fire events were recorded throughout the excavated area andtheir interpretation is not unequivocal. The fact that it is impossible,for now, to know whether the fire events that occurred in thehabitation areas are in anyway related with those that wererecorded in the upper enclosure, makes their interpretation moredifficult.

As pointed out before, most huts in As Laias were made fromperishable materials and these could easily make accidental firesget catastrophic proportions. This seems to be the most straight-forward interpretation for the recurrent fire events. Still, whilemany other hypotheses could be proposed we point out other two:

- Fire could have been used to eliminate grain pests or clean thestorage structures. Nevertheless, it probably would berestricted to the storage facilities and only by accident wouldspread through all the upper enclosure and outside the walledarea.

- Considering the possibility of a defensive purpose in the stor-age strategy in As Laias, we cannot rule out the possibility thatthe fires were the result of conflicts. Fires, in this case, wouldalso be accidental since they destroyed some of the cropswhich could have been the main purpose of the confrontation.

6.3. Massive storage at As Laias in a regional perspective

In northwest Iberia, during the Late Bronze Age, storage wasdone in pits, following a prehistoric tendency (Bettencourt, 1999;Ayán Vila and Parcero Oubiña, 2009). Storage-pits vary in size andshape and different materials could help promote further isolationto avoid humidity. In Sola (Braga), cork was found in the bottom of apit (Bettencourt, 2000) and in the Iron Age levels of Castro Grandede Neixón (Boiro) the walls of some pits seem to have been coveredwith clay (Ayán Vila and Parcero Oubiña, 2009).

However, caution is needed since other uses for the pits couldhave existed. Criado Boado et al. (1997) and A.M.S Bettencourt(2008, 2009b), for instance, suggest pits could have beencommonly used for burials throughout the Bronze Age. Poor

Fig. 10. Examples of Iron Age storage pits that provided plant remains: section view fromsection view (right, bottom) from Crastoeiro (adapted from Dinis, 2001). Scale: 50 cm.

preservation of bones, justifies the existence of few clear evidenceof burials in pits in the region. Fraga do Zorro (Ourense, Spain)(Fábregas Valcarce, 2001) is an exception.

The use of storage-pits continues throughout the Iron Age, astestified by already mentioned the cases of Castro Grande deNeixón (Ayán Vila and Parcero Oubiña, 2009), S. João de Rei(Bettencourt, 1999) and Crastoeiro (Dinis, 2001; Dinis andBettencourt, 2009) (Fig. 10), and in the Roman Period, even innew kind of settlements such as the vicus of Agro de Ouzande(Silleda) (Ayán Vila and Parcero Oubiña, 2009). The same happenedin other regions in northern Iberia (Gracia Alonso, 2009). Still, it isnot clear whether underground storage continued to be the pri-mordial strategy for long-term preservation of plant food innorthwest Iberia. There are not many undisputed storage facilitiesin Iron Age hillforts. Still, throughout the region there was a reor-ganization of the settlements, mostly since the 2nd century BC,leading to the definition of family compounds within walls, in thisperiod. These are constituted by several buildings and a centralterrace. It is usually assumed that one of the compartments wasused for storage (Almeida, 1984; Silva, 1986; Parcero Oubiña andCobas Fernández, 2004). This hypothesis still lacks strong sup-porting evidence, but assuming it to be true it would mean that, bythis time, there was a dissemination of above-ground storage indomestic contexts.

Before the first divulgation of As Laias (Álvarez González andLópez González, 2000) no storage structures like those wattle anddaub structures of As Laias were documented in the region. Asstated before, most storage facilities in As Laias were rectangularwattle and daub structures, with wooden boards also covered withclay, in the bottom.

Recently, data from Castrovite was reevaluated and, aftercomparing with As Laias, storage structures of the same kind wereproposed for that site, substituting earlier suggestions of storagehaving occurred in ceramic vessels (Rey Castiñeira et al., 2011). Still,the evidence is fragmentary and posing this hypothesis was onlypossible in the light of the well preserved As Laias’ structures. Thiscase suggests that such kind of structures could have existed inother sites but their fragments are being interpreted in other ways.In a small test pit (2 m � 2 m) at Bovinho (Macedo de Cavaleiros) agreat amount of burnt clay with marks of branches were found inassociation with abundant crops. That clay was interpreted asdebris of a floor (Mendes, 2006) but we must consider the

S. João de Rei (left) (adapted from Bettencourt, 2000b) and plan view (right, top) and

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e3877 3875

hypothesis that it was part of wattle and daub storage facility un-detected in such small excavation area.

As pointed out before, it is possible that most family compoundsinside the hillforts had its own storage area. Considering this hy-pothesis, we can only conjecture about the scale and time-span ofthat storage. Still, the storage on a familial level in a multifunctionaldomestic compound does not exclude the possible existence oflarge-scale and long-term storage in specific areas of the settle-ments or in other settlements with different functions on a regionallevel. As Laias, together with other particular sites, can help us get adifferent perspective.

The croa of As Laias was an enclosure mostly used for storage.This is not a unique case in northwest Iberia but it clearly standsout due to the magnitude of the area and the kind of storagestructures that were used. Overall, a varied assemblage of cropswas stored. Having been massively used for storage one canquestion whether it resulted only from agricultural activities ofthe people who inhabited the settlement or if people from othersettlements stored their crops there. The concentration of cropsin a single place may be interpreted in several ways. One hy-pothesis is that it happened for safety reasons, since crops storedtogether would be easier to defend. Still, there are other possibleinterpretations. Unknown political or symbolic reasons couldlead to such concentrations of crops in a single place. This leadsto a major question, that is, who owned the crops. Such greatnumber of storage facilities, concentrated in one place, could leadto a simple explanation of the place as a communal storage area.But this is not clear. For sure many families contributed to fill thestructures and such resources needed to be managed, a task wedo not know was communal or concentrated in specific membersof the community.

Related to this issue, next to the only entrance of the croa, be-tween two bastions, a set of forty undecorated vessels were un-covered in the pavement level. These are in clear contrast with thelack of archaeological artifacts within walls. The position of thosevessels suggests they fell from some kind of shelf placed in the wall.Most presented similar shapes but varied sizes which led ÁlvarezGonzález and López González (2000) to suggest these could havebeen used as some kind of volumetric units of measure related tothe storage facilities.

Other Iron Age settlements can be found near As Laias, namelySan Cibran de Las (San Amaro) (Álvarez González et al., 2009;Arizaga Castro et al., 2006; López Cuevillas, 1922) and San Trocado(San Amaro) (Fariña Busto and Xusto Rodríguez, 1988; ArizagaCastro et al., 2006). Despite the lack of radiocarbon dates it ispossible that these may have been at least partially contempora-neous of As Laias, considering the long term occupation of thissettlement. The chance that these settlements and As Laias wererelated in any way, due to their proximity and their potentialcoexistence, allows another possible interpretation: As Laias mayhave been a storage area for a larger community than that whichinhabited its rocky slope and alluvial area. On the other hand, in anarea important for its mining activities (Álvarez González andLópez González, 2000), specialized settlements could have exis-ted. Future work in the region should address this question.

Other Iron Age sites with similar characteristics - massive andcentralized storage of crops e are Castrovite (Rey Castiñeira et al.,2011) and Castro Grande de Neixón (Ayán Vila and ParceroOubiña, 2009). In Castrovite several Iron Age levels of fire pre-served abundant plant remains, possibly stored in structuressimilar to those of As Laias (Rey Castiñeira et al., 2011). In CastroGrande de Neixón a central storage area with several storage pits,was excavated in the settlements’ highest platform. Some pre-sented remains of burned clay which probably constituted the in-ner facing of the pits (Ayán Vila and Parcero Oubiña, 2009). It is

possible that the storage pits found in the top platforms of Cras-toeiro (Dinis, 2001; Dinis and Bettencourt, 2009) and S. João de Rei(Bettencourt, 1999) are part of large storage areas as thosementioned above.

Ayán Vila and Parcero Oubiña (2009) interpret As Laias, Cas-trovite and Castro Grande de Neixón as places for communal stor-age, examples of storage in central places which concentrated alsothe role of redistribution and trade. The interpretation of the ves-sels in the entrance of the croa in As Laias as volumetric units ofmeasure, suggested by Álvarez González and López González(2000), would fit to this model. This would suggest the existenceof some kind of hierarchy (Ayán Vila and Parcero Oubiña, 2009).Still, this is not the only interpretation possible. Álvarez Gonzálezet al., 2009 suggest each storage structure in As Laias could havebelonged to a family. This position assumes that the space wherecrops were stored e within walls e was communal but the struc-tures were not.

These two interpretations are opposite and are both theoreti-cally possible. The problem of interpreting the storage area of AsLaias in terms of how it reflects any kind of social organization isthat such interpretation is mostly dependent on how we under-stand the social structure of those societies in the first place. Sinceregarding this subject the empirical data available for northwestIberian Iron Age has led to opposite interpretations (see a revisionin Parcero Oubiña and Cobas Fernández, 2004) it is only naturalthat As Laias itself originates different interpretative possibilities.

Independently of these questions, sites such as As Laias confirmthat local communities obtained surplus from their fields, at leastsince the beginning of the Iron Age and possibly since the LateBronze Age. It does not agree with some interpretative models forthe Iron Age that underestimate the role of crops to emphasizeanimal husbandry as the only staple food provider, in the light ofwhat Strabowrote (see critics to these models in Bettencourt, 1999;Ramil Rego and Fernández Rodríguez, 1999).

7. Conclusion

In As Laias/O Castelo evidence of massive storage was found.This allowed the definition of its croa as a storage enclosure. Cropswere diverse and cereals were predominant, namely T. aestivumsubsp. spelta, P. miliaceum, Avena and H. vulgare subsp. vulgare.V. faba was also important. Several crops were stored in the samestorage facility but it is likely that some of these structures had asingle species.

This site is determinant to the understanding of Iron Age innorthwest Iberia not only because of the abundance of storage fa-cilities with plant materials, contrasting with the informationavailable for the rest of the region, but also because of the pecu-liarity of the structures used for storage. These are now leading tothe reinterpretation of other archaeological data available for theregion.

Still, the implications of As Laias may go well beyond the char-acteristics of its storage facilities. This site provided early dates forspelt and oat, allowing us to suggest the introduction of thesespecies occurred in a transitional moment between the Late BronzeAge and the Iron Age. The preponderance of spelt since the Iron Ageuntil the turn of the Era, testifies for the importance of hulledwheats in the region, in contrast with other Iberian regions (Buxóand Piqué, 2008). This species appears together with abundantmillet (P. miliaceum), hulled barley and oat comprising an assem-blage of cereals well adapted to variable and harsh environmentalconditions.

The interpretation of As Laias, together with other sites withstorage areas in northwest Iberia, can have some implications inthe understanding of social practices and social organization of the

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e38773876

Iron Age people. Since the top enclosure of As Laias was usedmostly for massive storage, it was likely to aggregate crops andwild fruits (acorns) to be consumed by several families. Was this acommunal storage place? Was the stored food communal orproperty of specific members of the community? Was its distri-bution controlled by specific members of the community? Theanswer to these questions depends on the interpretation of thesocial organization of these communities but, at the same time,can have great implications on defining that interpretation. Thisdeadlock creates severe difficulties to the interpretation of As Laiasand its storage area and presents itself as a challenge for futureinvestigations in the region.

Acknowledgments

João tereso was supported by a PhD grant (SFRH/BD/41628/2007) from Fundação para a Ciência e Tecnologia. The authors wishto thank Rita Gaspar, Bruno Marcos and Ana Rita Ferreira for theirhelp with cartographic elements and anonymous referee for valu-able suggestions.

References

Aira Rodriguez, M., Ramil Rego, P., Alvarez Nuñez, A., 1990. Estudio paleocarpológicorealizado en el Castro de Penalba (Campolameiro, Pontevedra. España). Bot.Complutensis 16, 81e89.

Akeret, Ö., 2005. Plant remains from a Bell Beaker site in Switzerland, and thebeginnings of Triticum spelta (spelt) cultivation in Europe. Veg. Hist. Archae-obot. 14, 279e286.

Alarcão, J., 1992. A evolução da cultura castreja. Conimbriga 31, 39e71.Almeida, C.A.F., 1984. A casa castreja. Memorias de Historia Antigua 6, 35e42.Álvarez González, Y., López González, L., 2000. La secuencia cultural del asen-

tamiento de Laias: evolución espacial y funcional del poblado. In: Jorge, V.(Ed.), Actas do 3� congresso de Arqueología Peninsular. ADECAP, Vila Real,pp. 523e532.

Álvarez González, Y., López González, L., López Marcos, M., 2009. La Ocupación delEspacio Común y Privado en la Citania de San Cibran de Lás. In: CongressoTransfronteiriço de Arqueologia. Grupo Cultural Aqva Flaviae, Montalegre,pp. 195e208.

Arizaga Castro, A., Fábrega Álvarez, P., Ayán Vila, J., Rodríguez Paz, A., 2006.A apropiación simbólica da cultura material castrexa na paisaxe cultural dosChaos de Amoeiro (Ourense, Galicia). Cuadernos de Estudios Gallegos 53,87e129.

Arnanz, A., Chamorro, J., 1990. Estudio de frutos e semillas procedentes de cuatroCastros Gallegos. Problemas de interpretación de los resultados. In: Encontrossobre Paleoeconomia e Paleoambiente.

Ayán Vila, J., Parcero Oubiña, C., 2009. Almacenamiento, unidades domésticas ycomunidades en el noroeste preromano. In: García Huerta, R., RodríguezGonzález, D. (Eds.), Sistemas de almacenamiento entre los pueblos pre-rromanos peninsulares. Ediciones de la Universidad de Castilla-La Mancha,Huesca, pp. 367e422.

Bettencourt, A.M.S., 1999. A paisagem e o Homem na bacia do Cávado durante o II eo I milénio AC. Universidade do Minho, Braga.

Bettencourt, A.M.S., 2000. O povoado da Idade do Bronze da Sola, Braga, Norte dePortugal. Universidade do Minho/Instituto de Ciência Sociais, Braga.

Bettencourt, A.M.S., 2000b. Estações da Idade do Bronze e inícios da Idade do Ferroda Bacia do Cávado (Norte de Portugal). Universidade do Minho/Instituto deCiência Sociais, Braga.

Bettencourt, A.M.S., 2008. Life and death in the Bronze Age of the NW of IberianPeninsula. In: Fahlander, F., Oestigaard, T. (Eds.), The Materiality of Death:Bodies, Burials, Beliefs. British Archaeological Reports, Oxford, pp. 99e104.

Bettencourt, A.M.S., 2009. A Pré-história do Minho. Do Neolítico à Idade do Bronze.In: Pereira, P. (Ed.), Minho. Traços de Identidade. Universidade do Minho, Braga,pp. 70e120.

Bettencourt, A.M.S., 2009b. Práticas funerárias da Idade do Bronze de Trás-os-Montes e da Galiza Oriental. In: Congresso Transfronteiriço de Arqueologia.Grupo Cultural Aqva Flaviae, Montalegre, pp. 11e23.

Bettencourt, A.M.S., Dinis, A., Figueiral, I., Rodrigues, A., Cruz, C., Silva, I.,Azevedo, M., Barbosa, R., 2007. A ocupação do território e a exploração derecursos durante a Pré-história Recente do Noroeste de Portugal. In: Jorge, S.,Bettencourt, A.M.S., Figueiral, I. (Eds.), A Concepção das paisagens e dos espaçosna Arqueologia da Península Ibérica. Universidade do Algarve, Faro, pp. 149e164. Actas do IV Congresso de Arqueologia Peninsular.

Buxó, R., 1997. Arqueología de las Plantas. Crítica, Barcelona.Buxó, R., Alonso, N., Canal, D., Echave, C., González, I., 1997. Archaeobotanical re-

mains of hulled and naked cereals in the Iberian Peninsula. Veget. Hist.Archaeobot 6, 15e23.

Buxó, R., Piqué, R., 2008. Arqueobotánica. Los usos de las plantas en la penínsulaIbérica. Ariel, Barcelona.

Carballo Arceo, L., 1998. A agricultura en Castrovite (Orazo, A Estrada) durante aIdade do Ferro. Miscelánea Histórica e Cultural 1, 9e25.

Carrión Marco, Y., 2005. Dendrología y arqueología: las huellas del clima y de laexplotación humana de la madera. In: Molera, J., Farjas, J., Roura, P., Pradel, T.(Eds.), Avances en Arqueometría. Actas del VI Congreso Ibérico de Arqueome-tría. Universitat de Girona, pp. 273e282.

Carrión Marco, Y., 2003. Afinidades y diferencias de las secuencias antracológicas enlas vertientes mediterránea y atlántica de la Península Ibérica. Faculdat deGeografia i História, Universitat de Valéncia.

Cobas Fernández, I., Parcero Oubiña, C., 2006. Alto do Castro (Cuntis, Pontevedra).Síntesis de resultados y estudio de materiales, campaña 1993. CSIC e Xunta deGalicia.

Criado Boado, F., Santos Estévez, M., Parcero Oubiña, C., 1997. Settlement patterns,land-use and sacred landscapes in Galician late prehistory and ancient history:essay on an archaeology of sacred geographies. In: Criado Boado, F., Parcero, C.(Eds.), 1997. Landscape, Archaeology, Heritage, vol. 2. Tapa, pp. 11e18.

Dinis, A., 1993e1994. Contribuição para o estudo da Idade do Ferro em Basto: oCastro do Crastoeiro. In: Cadernos de Arqueologia, Serie 2, vol. 10e11,pp. 261e278.

Dinis, A., 2001. O povoado da Idade do Ferro do Crastoeiro (Mondim de Basto, Nortede Portugal). Universidade do Minho, Instituto de Ciências Sociais, Braga.

Dinis, A., Bettencourt, A., 2009. A Arte Atlântica do Crastoeiro (norte de Portugal):contextos e significados. Gallaecia 28, 41e47.

Dopazo Martinez, A., 1996. La dieta vegetal del Noroeste Ibérico durante el Hol-oceno. In: Una aproximación a través del análisis paleocarpológico. Universidadde Santiago de Compostela.

Dopazo Martínez, A., Fernández Rodríguez, C., Ramil-Rego, P., 1996. Arqueometríaaplicada a yacimientos Galaico-romanos del NW Penínsular: valoración de laactividad agrícola y ganadera. In: Ramil-Rego, P., Fernández Rodríguez, C.,Rodríguez Guitían, M. (Eds.), Biogeografia Pleistocena - Holocena de la Penín-sula Ibérica. Xunta de Galicia, Santiago de Compostela, pp. 317e332.

Fabião, C., 1992. O Passado Proto-Histórico e Romano. In: Mattoso, J. (Ed.),História de Portugal. Antes de Portugal, vol. 1. Círculo de Leitores, Lisboa,pp. 77e299.

Fábregas Valcarce, R., 2001. Los petroglifos y su contexto: un ejemplo de la Galiciameridional. Instituto de Estudios Vigueses, Vigo.

Fariña Busto, F., Xusto Rodríguez, M., 1988. Coto do San Trocado (San Amaro-Pun-xín-Ourense). In: VVAA (Ed.), Arqueoloxía-informes 2. Campaña, Junta deGalicia.

Figueiral, I., 2008. O Crasto de Palheiros (Murça, NE Portugal): a exploração dosrecursos vegetais durante o III/inícios do II� milénio AC e entre o I� milénio AC eo séc. II� DC. In: Sanches, M. (Ed.), O Crasto de Palheiros. Fragada do Castro.Murça e Portugal. Município de Murça, Murça, pp. 79e108.

González-Ruibal, A., 2003. Arqueología del Primer Milenio a.C. en el Noroeste de laPenínsula Ibérica. Complutense University of Madrid, Madrid.

Gracia Alonso, F., 2009. Producción y almacenamiento de excedentes agrícolas en elNE peninsular entre los siglos VII y II a.C. Análisis crítico. In: García Huerta, R.,Rodríguez González, D. (Eds.), Sistemas de almacenamiento entre los pueblosprerromanos peninsulares. Ediciones de la Universidad de Castilla-La Mancha,Cuenca, pp. 9e71.

Hillman, G., Mason, S., de Moulins, D., Nesbitt, M., 1996. Identification of archaeo-logical remains of wheat: the 1992 London workshop. Circaea 12, 195e210.

Jacomet, S., 2006. Identification of Cereal Remains from Archaeological Sites, 2aedição ed. Archaeobotany Lab, IPAS, Basel University.

López Cuevillas, F., 1922. O castro A Cibdade en San Ciprián de Lás, Ourense.Lopez Merino, L., Peña-Chocarro, L., Ruiz Alonso, M., Lopez Saez, J., Sanchez

Palencia, F., 2010. Beyond nature: the management of a productive culturallandscape in Las Medulas area (El Bierzo, Leon, Spain) during pre-Roman andRoman times. Plant Biosystems 144, 909e923.

Martínez Mira, I., 2004e2005. Tesorillos del s. III d.C. en la península ibérica (III),Lvcentvm 23/24, pp. 207e236.

Martins, M., 1993e1994. Continuidade e mudança no I milénio a.C., no NoroestePortuguês. In: Os diferentes cenários de representação do discurso arqueoló-gico, Cad. Arqueol, vol. 10e11, pp. 41e64.

Martins, M., 1996. Povoamento e habitat no Noroeste português durante o 1�

milénio a.C. In: Alarcão, J. (Ed.), De Ulisses a Viriato. O primeiro milénio a.C.Ministério da Cultura, Instituto Português de Museus, Museu Nacional deArqueologia, Lisboa, pp. 118e133.

Mendes, C., 2006. Povoado Mineiro do Bovinho, Freguesia de Edroso, Resultadospreliminares da campanha 2/2005, Cad. Terras Quentes 3, 53e59.

Miret i Mestre, J., 2008. L’experimentació sobre sitges tradicionals. Aportacions del’arqueologia i de l’agronomia. Rev. d’Arqueologia de Ponent 18, 217e240.

Morais, R., 2004. Autarcia e Comércio em Bracara Augusta no período Alto-Imperial:contribuição para o estudo económico da cidade. Universidade do Minho,Braga.

Muñoz Sobrino, C., Ramil-Rego, P., Rodríguez Guitián, M., 1997. Upland vegetation inthe north-west Iberian peninsula after the last glaciation: forest history anddeforestation dynamics. Veget. Hist. Archaeobot. 6, 215e233.

Oliveira, M., 2000. O registo paleocarpológico do NO peninsular entre o III� e o I�

milénios a.C. Contributo para o estudo da alimentação pré e proto-histórica.Universidade do Minho, Braga.

Parcero Oubiña, C., 2000. Síntesis de los trabajos de excavación en el yacimientocastreño de Alto do Castro (Cuntis, Pontevedra). Brigantium 12, 161e174.

J.P. Tereso et al. / Journal of Archaeological Science 40 (2013) 3865e3877 3877

Parcero Oubiña, C., Cobas Fernández, I., 2004. Iron Age archaeology of the north-west Iberian peninsula. e-Keltoi 6, 1e72.

Peña Santos, A., 2005. Breve síntesis de la prehistoria e arqueología do EixoAtlántico. In: Hidalgo Cuñarro, J. (Ed.), Rutas Arqueolóxicas do Eixo Atlántico/Roteiro Arqueológico do Eixo Atlântico. Gráficas Planeta, S.L., Vigo/Porto/Brux-elles, pp. 3e42.

Pinto da Silva, A., 1988. A paleoetnobotânica na arqueologia portuguesa. Resultadosdesde 1931 a 1987. In: Queiroga, F., Sousa, I., Oliveira, C. (Eds.), Palaeocologia eArqueologia. Câmara Municipal de Vila Nova de Famalicão, Vila Nova deFamalicão, pp. 13e29.

Pinto da Silva, A., 2001. Relatório do estudo carpológico. In: Dinis, A. (Ed.),O povoado da Idade do Ferro do Crastoeiro (Mondim de Basto, Norte dePortugal). Universidade do Minho, Instituto de Ciências Sociais, Braga.

Ramil Rego, P., 1993. Paleoethnobotánica de yacimientos arqueológicos holocenosde Galicia (N.O. Cantábrico). Munibe 45, 165e174.

Ramil Rego, P., Dopazo Martinez, A., Fernández Rodríguez, C., 1996. Cambios en lasestrategias de explotación de los recursos vegetales en el Norte de la PenínsulaIbérica. Férvedes 3, 169e187.

Ramil Rego, P., Fernández Rodríguez, C., 1999. La explotación de los recursos ali-menticios en el noroeste ibérico. In: García Quintela, M. (Ed.), 1999. Mitología ymitos de la Hispania prerromana, vol. III. Akal, pp. 296e342.

Ramil Rego, P., Muñoz-Sobrino, C., Rodríguez-Guitián, M., Gómez-Orellana, L., 1998.Differences in the vegetation of the North Iberian Peninsula during the last16,000 years. Plant Ecol. 138, 41e62.

Reimer, P., Baillie, M., Bard, E., Bayliss, A., Beck, J., Blackwell, P., Bronk Ramsey, C.,Buck, C., Burr, G., Edwards, R., Friedrich, M., Grootes, P., Guilderson, T., Hajdas, I.,Heaton, T., Hogg, A., Hughen, K., Kaiser, K., Kromer, B., McCormac, F.,Manning, S., Reimer, R., Richards, D., Southon, J., Talamo, S., Turney, C., van derPlicht, J., Weyhenmeyer, C., 2009. IntCal09 and Marine09 radiocarbon Agecalibration curves, 0e50,000 years cal BP. Radiocarbon 51, 1111e1150.

Rey Castiñeira, J.,Martín Seijo,M., Teira Brión, A., AbadVidal, E., Calo Ramos, N., CarballoArceo, L., Comendador Rey, B., Picón Platas, I., Varela Montes, A., 2011. CastroBYTE.Un modelo para a xestión da información arqueolóxica. Gallaecia 30, 67e106.

Sanches, M.J., 2008. O Crasto de Palheiros. Fragada do Crasto. Murça e Portugal.Munícipio de Murça, Murça.

Silva, A.C.d., 1986. A Cultura Castreja do Noroeste de Portugal. Museu Arqueológicoda Citánia de Sanfins, Paços de Ferreira.

Tereso, J., 2009. Plant macrofossils from the Roman settlement of Terronha dePinhovelo, northwest Iberia. Veget. Hist. Archaeobot. 18, 489e501.

Tereso, J., 2012. Environmental Change, Agricultural Development and Social Trendsin NW Iberia from the Late Prehistory to the Late Antiquity. Biology Depart-ment, Faculty of Sciences, University of Porto, Porto.