This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Neanderthal subsistence strategies in Southeastern France between the plains ofthe Rhone Valley and the mid-mountains of the Massif Central (MIS 7 to MIS 3)

Camille Daujeard a,*, Paul Fernandes b, Jean-Luc Guadelli b, Marie-Hélène Moncel a, Carmen Santagata b,Jean-Paul Raynal b

aDépartement de Préhistoire, Muséum National d’Histoire Naturelle, UMR 7194, 1 rue René Panhard, 75013 Paris, FrancebUniversité Bordeaux 1, PACEA-IPGQ, UMR 5199, Bâtiment B18, Avenue des facultés, 33405 Talence, France

a r t i c l e i n f o

Article history:Available online 24 February 2011

a b s t r a c t

New investigations of finds frommiddle Rhone Valley sites (Le Figuier, Abri du Maras, Baume Flandin andRanc Pointu 2) and from the mountainous southeastern area of the Massif Central (Sainte-Anne I,especially the unpublished data for unit J2) allowed assembly of new data on both chronological andenvironmental grounds for human occupation and consequently on Neanderthal subsistence strategiesin this area.

The southeastern Massif Central region makes it possible to examine Neanderthal occupation modes inan area linking medium altitude territories with the plains of the Rhone River corridor. This work aims todetermine site occupation events and territory management strategies for these two environments. Thevariability of the site occupation types in the middle Rhone Valley supports the hypothesis of humangroups who anticipated their land use strategies, and suggests a particular type of circulating model forthese areas. The evidence suggests that the human presence consists of brief stopping-places, short-termregular camps (interspersed with some occupations by carnivores), or long-term residential camps. Thetypes of occupation observed in the two mountain sites of Velay (Sainte-Anne I cave and Baume-Valléerock-shelter) a priori do not indicate occupations specific to an environment at this altitude. On thecontrary, they suggest the existence of short-term regular camps, probably seasonally utilised andexploited when good seasons favoured the expansion of accessible territories. The sites in the plains ofthe Rhone Valley corridor indicate a greater variety of occupation types, but still within the framework ofstrong seasonal mobility.

� 2011 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction

Methodological and taphonomical advances, combined witharchaeological data from numerous European sites, allow one tothink beyond the simple hunting versus scavenging question forneanderthalers. Instead, questions can be raised about the Nean-derthal subsistence modes. The debate about behavioural differ-ences between Neanderthals and anatomically modern humans ispresently focused on strategies for territorial exploitation (Boyle,2000; Conard and Prindiville, 2000; Moncel, 2003; Szmidt, 2003,2009; Fiore et al., 2004; Valensi and Psathi, 2004; Burke, 2006;

Grove, 2009; Daujeard and Moncel, 2010; Rendu, 2010; amongothers). Did these two hominid groups adopt differing subsistencestrategies? Did the Neanderthals use more-or-less permanentresidential sites, and exploit intensively their immediatesurroundings (circulating model), while modern humans adopteda logistic behaviour showing amore organised spatial and temporalorganisation, centred around their base camp (radiating model;Mortensen, 1972; Binford, 1982). New data shed light on thisimportant debate. In Eastern Europe, Neanderthal occupationwithin continental mountains is found in the Alps, the Pyreneesand the Italian Piedmont mountain ranges. The sites are located inthe “mid-mountain” environment at altitudes between 500 and1500 m, where the topography, slope conditions and climaticconditions permitted suitable access and resource exploitation. Theexistence of these medium altitude sites raises the question abouttheir functionwithin these territories. Two possible hypotheses canbe envisioned:

* Corresponding author.E-mail addresses: camilledaujeard@hotmail.fr (C. Daujeard), nuage75@club-

internet.fr (P. Fernandes), jl.guadelli@pacea.u-bordeaux1.fr (J.-L. Guadelli),moncel@mnhn.fr (M.-H. Moncel), carmen_santagata@tin.it (C. Santagata),jp.raynal@pacea.u-bordeaux1.fr (J.-P. Raynal).

Contents lists available at ScienceDirect

Quaternary International

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

1040-6182/$ e see front matter � 2011 Elsevier Ltd and INQUA. All rights reserved.doi:10.1016/j.quaint.2011.01.047

Quaternary International 252 (2012) 32e47

Author's personal copy

- Neanderthal occupation of the altitudinal environment tookplace during periods of climatic warming, such as MarineIsotopic Stage (MIS) 3, leading to an expansion in exploitableterritories e behaviour similar to that observed on the plains(Koslowsky, 1994).

- Occupation of the mountain sites corresponds to particularneeds in the medium altitude zones and leads to original andparticular occupation strategies developed for such sites.

Are these hypotheses reflected in the material culture? Forexample, the hypothesis of a culture specific to the alpine regionhas been discarded in the light of recent research (Bächler, 1940;Bernard-Guelle, 2002; Tillet, 2002; Bona et al., 2007). Somemedium altitude sites have produced poor lithic assemblages thatare composed of finished imported tools associated with abundantremains of bears; however, these sites suggest human stays ofa very short duration (Cârciumaru et al., 2002).

Recent work has underlined a reasoned and diversified exploi-tation of mountainousmassifs. For example, the Mousterian sites ofVercors, on the edge of the French Alps (1000e1500 m), similar tothose in the Italian Piemont, indicate a large Neanderthal pop-ulation and suggest the presence of highly mobile human groupswhich were adapted to seasonal movements in search of ungulates(red deer, roe deer or mountain sheep) within a territory whichstretched betweenmassif and plains environments (Peresani, 1998;Peresani and Sartorelli, 1998; Bernard-Guelle, 2002, 2005;Arzarello, 2003; Arzarello and Peretto, 2005). The numerousadvantages available near Vercors (raw materials, diverse biotopes,differing landscapes) also led to repeated and regular occupations;especially short-term seasonal hunting camps complementarywith those located in the plains (Bernard-Guelle, 2002, 2005).

The rock-shelter of Pié Lombard (250 m), in the Alps-Maritimes,can be considered as a “first stop” at the foot of the pre-alpinemountains, in contrast to coastal occupations that suggest occupa-tions of larger territories. Petro-archaeological determinationsconfirm that the mid-mountain hinterland was frequented at alti-tudes above 1000 m (Porraz, 2005). In the French Pyrenees, theGrotte du Noisetier (825 m, MIS 3) shows evidence of a humanpresence that is not tied to faunal resources or lithics that are avail-able in the surrounding areas. Exclusive exploitation of big game,namely red deer, equally presents in both foothills and plains, alongwith rawmaterials similar to the ones known from the sites at loweraltitudes, do not justify the regular forays made by the Nean-derthalers at this altitude. Instead, the scenario ismore suggestive ofa temporary stopping site or short-term camp (Mourre et al., 2008).

In the southeast quarter of the Massif Central, the mid-moun-tains are connected to the Rhone Valley by a succession of plateaus.The hydrographic basins in this region (Ardèche, Allier and LoireRivers) are only separated by tiny interfluves and the topographicalrelief is not a barrier that prevents movement between the valleys.Themovement of human groups between the different zones of thisvast area during the course of the Pleistocene is a question whichremains to be tackled in depth. Proposed models mostly associatecultural identity with oro-hydrographical patterns. Recent excava-tions in this geographic zone offer evidence from a corpus of sitesthat is sufficient to begin discussions about humans occupyingspecific territories and the circulation of human groups.

The southeast margins of the Massif Central offer a high degreeof chronological and bio-stratigraphic resolution which givesopportunities to compare the subsistence, technical and rawmaterial procurement strategies between the Rhone Valley and themid-mountain region of the Velay-Vivarais (Raynal, 2007;Fernandes et al., 2008; Raynal et al., 2008).

Movement routes demonstrated by flint procurement in themountain sites and in the sites of the plains attest that humans

circulated more frequently within an area of around 50e60 km indiameter and that they exploited a multiplicity of regional lithicoutcrops. Northesouth movements are clearly demonstrated bya distinct presence of certain raw materials within the twoconsidered territories. The current data provided by flint procure-ment give no evidence for eastewest exchanges between theRhone Valley and the Velay. However, the territories covered allowone to envisage and test the hypothesis of a Central-Eastern paleo-ethnographic unit, at least when considering the different Mous-terian facies identified in the area since MIS 5. Were the rawmaterials fully representative of various Neanderthaler group’scultural identities? Perhaps this question is related to the conceptof modernity.

The sites of Sainte-Anne I (Haute-Loire, 737 m) and Baume-Vallée (Haute-Loire, 795 m) illustrate this type of Neanderthaloccupation in the Massif Central mountains. In this paper, thepreliminary and unpublished environmental, taphonomic andlithic data obtained for unit J2 of Sainte-Anne I, which is currentlybeing excavated, and those of unit J1 as well as that from Baume-Vallée are compared with the more numerous sites of the plainsand the low plateaus of the Rhone Valley. The variety of occupationsites along the Rhone corridor suggests a variation on the twosystems of mobility proposed by Binford (1982); in other wordsa model of the circulating type that was “planned and organised”and which linked residential base camps with secondary short-term camps that function as seasonal hunting camps or as inter-mediate steps on an extended journey (Daujeard, 2008; Daujeardand Moncel, 2010).

Is there a single mode of occupationwhich is specific to themid-mountains, located away from the Rhone Valley and that is distinctfrom the ones of the plateaus and plains bordering the RhoneValley? Is it possible to observe, thanks to the subsistence strate-gies, the existence of specialised sites in a mountain context? Orshould, on the contrary, the Massif Central be considered as a vastarea that was uniformly exploited by the Neanderthalers, nomatterwhat the altitude was?

2. Environmental and chronological context of the sites

The middle valley of the Rhone and its right bank tributariesform a strategic zone for studying the northesouth and eastewestmigratory passages of human groups. This valley certainly playeda major role in the peopling processes of regional river valleys andgorges (Fig. 1). The region has a pivotal role between the mid-mountains of the Massif Central and the mountainous areas ofVercors and the Alps. Its microclimates favoured the persistence ofthe human occupations throughout the Pleistocene. The middleRhone Valley is part of a well-distinguished geological zone oflimestone called Bas-Vivarais which runs along the Rhodaniancorridor. Rich in watercourses, gorges and calcareous plateaus, itcontains numerous shelters. Sites which contain Middle Palae-olithic occupation layers are well distributed along this part of theright bank of the Rhone Valley. FromValence to the Ardèche gorges,more than a dozen rock-shelters and caves have been recorded.Some, such as Payre or the sites of Soyons (Abri Moula and GrotteNéron), border the Rhone Valley, while numerous others are scat-tered along the tributaries of the Ardèche and the Chassezac Rivers.North of this limestone karst region is the igneous mountains of theMassif Central, where two high sites, Sainte-Anne I (737 m) andBaume-Vallée (795 m), have yielded several rich units of MiddlePalaeolithic stone artefacts and bone remains.

The cave of Sainte-Anne I is a small cavity (approximately50 m2) facing south at 737 m above sea level. It developed ona fracture in the tuff massif of the ancient volcano of Saint-Anne. Itlies 150 m above the Borne Valley which is a tributary of the left

C. Daujeard et al. / Quaternary International 252 (2012) 32e47 33

Author's personal copy

bank of the Loire. Several archaeo-stratigraphic deposits withMiddle Palaeolithic assemblages containing handaxes (J1, J2, J3, J4,from the most recent to the oldest), separated by roof and wallcollapse (E1, E2, E3), have been recognised in the porch area of thiscave. These deposits have been heavily modified by secondary frostaction. The cold fauna spectrum and the presence of a specific horse

clearly place these deposits withinMIS 6, which is confirmed by thepreliminary datings (Raynal, 2007; Fig. 2; Table 1).

The Baume-Vallée rock-shelter (also named Laborde rock-shelter, at Solignac-sur-Loire) faces southeast at the foot of a cliff795 m above sea level, on the left bank of the Ourzie River which isa tributary of the left bank of the Loire. In its lower part it contains

Fig. 1. Location of the main Middle Palaeolithic sites in southeastern France.

C. Daujeard et al. / Quaternary International 252 (2012) 32e4734

Author's personal copy

several archaeo-stratigraphic units which have been referred to theFerrassie type Charentian Mousterian. The stratigraphy shows thatsedimentation was the result of frost action, in particular solifluc-tion, which becomes increasingly active towards the top of theMousterian sequence and determines a secondary stratogenesis.Thermoluminescence (TL) and Electron Spin Resonance (ESR) datesgave an age around 80 ka (MIS 5a) (Raynal, 1988; Raynal et al.,2005).

The chronology of the Rhone Valley sites indicates a Neander-thal occupation from MIS 8/7 in unit F of Payre until MIS 3 for themost recent (Fig. 2; Table 1). At the beginning of the last interglacialthe number of sites increases (Moncel, 2003; Daujeard, 2008;Moncel et al., 2008a,b; Valladas et al., 2008). Among the agesmost recently determined for the Mousterian of the Ardèche region(between about 30 and 43 ka) are the units of Baume Néron

(Defleur et al., 1994), Ranc de l’Arc (Defleur et al., 1990), Abri Moula(Evin et al., 1985), Baume d’Oullins, Abri des Pêcheurs and Saint-Marcel (Fig. 2; Table 1).

With all the caution needed, given the geological and archaeo-logical contexts, the diversity of the macro-mammals speciesrecovered in the studied units provides a picture of the existingbiotopes and allows definition of broad types of palaeo-environ-ments in which Middle Palaeolithic humans acted:

(1) Rocky landscapes for adapted species;(2) Plateaus and wooded valley bottoms suitable for large forest

herbivores;(3) Cold and open spaces suitable for large herd-dwelling ungu-

lates and for the “arctic group” higher in mountains (Delpechet al., 1983; Griggo, 1995).

Fig. 2. Chronological framework of the Middle Palaeolithic sites on the right bank of the middle Rhone Valley, including southeastern Massif Central (modified from Moncel, 2008).

C. Daujeard et al. / Quaternary International 252 (2012) 32e47 35

Author's personal copyTa

ble

1To

pog

raphy

,chronolog

y,numbe

rof

remains,frag

men

tation

andbo

nedestructionindices

forev

erystudiedsamples.

Sites

Livingsp

ace

andorientation

Archae

olog

ical

laye

rsan

dleve

lsEx

cava

tedarea

/site

surface

Age

ofoc

cupations

NISP

NRf

NISP/NR

(%)

Isolated

teeth/N

ISP

NRl

Lithic

den

sity/m

2

(richestoc

cupationleve

l)

Bau

med’Oullinsb

Huge

rock

-shelter,

N-E

Leve

lsRaan

dRb

Cham

ber1z

3m

2MIS

3(endof

Würm

II,C

ombier,1

967;

Deb

ard,1

988;

leve

lR:31

,750

�58

3by

aminoacid

racemization;

Lafontet

al.,19

84)

ee

ee

132

44m

2

Figu

iercave

(Ardèche)

Largecave

porch

and

largean

dnarrow

cham

bers,S

Old

excava

tion

s(roo

m1an

dco

rridor)

TotalCham

ber1

(z10

0m

2)

MIS

4or

3(endof

Würm

II,

Com

bier,1

967)

139

177a

e0.55

988a

e

2007

(roo

ms2an

d3)

3m

2(z

8m

2)

MIS

4or

3(M

IS5?

forlowerleve

ls,

Mon

celet

al.,20

10)

358

4406

8.1

0.36

311

35m

2

BarassesII-Balazuc

(Ardèche)

Narrow

cham

ber,S

Leve

l1

z6m

2(z

40m

2)

MIS

4or

3(A

ncien

tW

ürm

,Com

bier

1968

;Crégu

t-Bon

nou

reet

al.,20

10)

4243

ae

0.36

00m

2

Leve

ls2e

3z

9m

2(z

40m

2)

1465

5297

27.7

0.14

845m

2

Bau

me-Vallée

(Hau

te-Loire)

Largerock

-shelter

(795

m),S-E

Lowersequ

ence,

units0,

1an

d2

z29

m2

MIS

5ato

MIS

4(A

ncien

tW

ürm

,Ray

nal,1

988;

dates

arou

nd80

ka(TLan

dES

R/U

eTh

),Ray

nal

etal.,20

05)

286

3824

7,5

e25

52e

Pêch

eurs

cave

(Ardèche)

Pit-cave

,Narrow

cham

ber,S-E

2005

z4m

2

(z15

e20

m2)

Endof

MIS

5to

3(A

ncien

tW

ürm

,dates

betw

een83

and39

kaBP(U

eTh

),Masao

udiet

al.,19

94;31

ka(C14

),Ev

inet

al.,19

85;Deb

ard,1

988;

(Mon

cel,20

06;Mon

celdir.,20

08c;

Mon

celet

al.,20

10;Crégu

t-Bon

nou

reet

al.,20

10)

207

235a

e0.12

202

22m

2

Bau

medes

Peyrards

(Vau

cluse)

Huge

rock

-shelter,

S-O

Upper

ensemble

(a,b,c,d)

Totalactual

area

MIS

4e

beginningof

MIS

3(deLu

mley,

1969

)22

9415

,049

15.2

0.20

ee

Maras

shelter

(Ardèche)

Largerock

-shelter,

S-E

Leve

l1e

10To

talarea

(z33

m2)

MIS

3(endof

Würm

II,C

ombier,

1967

;Deb

ard,1

988)

4982

75.9

0.35

3695

92m

2

Upper

laye

r(z

<30

0)z

11m

2En

dof

MIS

5to

MIS

4(Lev

el5:

90e70

kaBP(U

eTh

),(M

oncela

ndMichel,2

000;

Deb

ard,1

988;

Mon

cel,

2006

;Mon

celet

al.,20

10)

181

741

24.4

0.24

1989

100m

2

Lower

laye

r(z

>30

0)4m

2MIS

5e

(5e?

)(M

oncel,20

06;

Mon

celet

al.,20

10)

5092

95.4

0.1

238

36m

2

Ran

cPo

intu

2Narrow

cham

ber,

NeO

Leve

lc

4m

2En

dof

MIS

5(W

ürm

I,Com

bier,

1967

;Deb

ard,1

988)

33e

ce

990

30m

2

Saint-Marcelcave

(Ardèche)

Largecave

porch

,SUpper

leve

ls11

m2(z

70m

2)

EndMIS

5or

MIS

3(Lev

ele:

29,330

�65

0BP,

Evin

etal.,19

85;

Leve

lf:37

,850

�55

0BP,

Szmidtet

al.,20

10(C14

))

169

862

19.6

0.06

483m

2

Ensemble7

11m

2(z

70m

2)

EndMIS

5or

MIS

3(D

aujeard,2

008)

1962

17,569

11.2

0.07

3705

113m

2

Lower

leve

ls8m

2(z

70m

2)

EndMIS

5(D

ebard,1

988)

188

919

20.5

0.14

924

53m

2

Leve

lu

5m

2(z

70m

2)

MIS

5e(Eem

ian,C

régu

t-Bon

nou

re,in

Defl

euret

al.,20

01,C

régu

t-Bon

nou

reet

al.,20

10)

568

3006

18.9

0.13

214

43m

2

Bau

meFlan

din

(Ardèche)

Med

ium-sized

and

narrow

cham

bers,O

Terrace-leve

l3(200

5)2m

2MIS

5e(Eem

ian,G

agnière

etal.,19

58;

Com

bier,1

967;

Mon

celet

al.,20

08b,

2010

;Crégu

t-Bon

nou

reet

al.,20

10)

104

116a

e0.15

136

68m

2

Cham

ber1

(195

4e19

57)

Totalarea

(z25

m2)

168

180a

e0.6

1071

ae

Sainte-A

nneI

(Hau

te-Loire)

Med

ium-sized

cave

porch

(737

m)S

J1,

J2(200

6e20

09)

Totalarea

(z70

m2),

Totalarea

(z40

m2)

MIS

6(dates

arou

nd17

0e21

0ka

(ESR

/UeTh

),Ray

nal

etal.,20

07)

346,

340

6365

,62

904.2,

5.4

e,

0,45

2377

,17

2840

m2,

58m

2

Payre-F(A

rdèche)

Middle-sized

cham

ber

Leve

lFa

25m

2(z

40m

2)

MIS

7,23

2�15

ka(TL),2

35�18

ka(ESR

/UeTh

),(M

asao

udie

tal.,19

97;Mon

cel

dir.,20

08c;

Valladas

etal.,20

08)

1307

24,541

5.3

0.20

2489

100m

2,(16

6m

3)

Leve

lFb

z15

m2(z

40m

2)

211

3115

6.8

0.23

804

54m

2,(16

1m

3)

Leve

lFce

d25

m2(z

40m

2)

753

8868

8.5

0.20

1104

41m

2(74m

3)

Findet

25m

2(z

40m

2)

187

3384

5.5

0.29

ee

Total

25m

2(z

40m

2)

2459

39,909

6.2

0.20

ee

NISP;

Numbe

rof

Iden

tified

Specim

ens;

NRf;Numbe

rof

faunal

Rem

ains;

NRl;Numbe

rof

lithic

Rem

ains;

take

nfrom

Dau

jeardan

dMon

cel,20

10withnew

dataforSa

inte-A

nneIan

dBau

me-Vallée.

aSa

mpleswithou

t<5cm

frag

men

ts.

bFa

unal

datafrom

Com

bier,1

967.

cUniden

tified

frag

men

tsnon

-cou

nted.

C. Daujeard et al. / Quaternary International 252 (2012) 32e4736

Author's personal copy

3. Material and methods

Apart from the Baume d’Oullins series (Combier, 1967) and thosefrom Baume-Vallée (Guadelli, Fiore and Tagliacozzo, in Fiore et al.,2005; Raynal et al., 2005) and Sainte-Anne I (J1) (Guadelli, Fiore andTagliacozzo, in Raynal, 2007), all the faunal data presented here arederived from the analysis and first-hand data of one author (CD) (seeDaujeard, 2008; Daujeard and Moncel, 2010, for taphonomical andzooarchaeological methods). The focus is on the unpublished datafrom unit J2 of Sainte-Anne I, where excavations are still in process.

For each faunal assemblage, analysis distinguished between thenumber of total remains (NR), number of identified specimens(NISP) andminimumnumber of individuals (MNI). Only taxonswithan NISP (without isolated teeth)>30 were included in the count forminimum number of elements (MNE) (Binford, 1984; Lyman,1994).Considering the scarcityof the identifiedarticular portionsat Sainte-Anne I, theMNE could not be calculated. Anatomical parts have beenrepresented by their NISP according to six categories (from Wilson,1989, modified): head (skull and mandibles), axial skeleton (verte-brae and ribs), fore quarters (scapula, humerus, radio-ulna andcarpal), hind-quarters (pelvis, femur, patella, tibia and tarsal) andfore, hind and indeterminate foot (metapodials and phalanxes). Therepresentativeness of the bone samples was evaluated according tothe number of remains, the area of excavation, the impact ofdifferential preservation (MAU percentages of each anatomicalelement compared with its bone bulk density; Grayson, 1989;Lyman, 1994) and the anatomical elements yielding similar MNIvalues (Delpech and Villa, 1993; Daujeard, 2008; Table 1). With theexceptionof the carnivores of Payre-Fandof themicro-mammals, allbonematerial was studied fromevery distinct archaeological unit inall sites. The type of damage to each specimen (identifiable andrecorded bones >5 cm) was recorded systematically, except fordental remains (apart from cases of ingestion), unreadable remainsand bone pieces less than 5 cm long; the latter category only beingused for fragmentation studies (Villa and Mahieu, 1991) andcarbonization analyses (Costamagno et al., 1998; Théry-Parisot,2001; Théry-Parisot et al., 2005).

The lithic assemblages were studied by reconstructing thereduction sequences for each large class of raw material type. Thesourcing of raw material used geological surveys along with macro-scopic and microscopic analyses. The technical processes identifiedhave been described according to classical approaches (Geneste,1988; Boëda, 1994a,b; Boëda et al., 1990; Geneste et al., 1997). Thepurpose was to record the knapping, shaping and retouching modesaccording to the technical features of the lithic products and todetermine various technical activities according to the types of rawmaterials being flaked or tools being manufactured. At Sainte-Anne Ifor example, the volcanic lithics and quartz were analysed together,because of the similar technical processes identified in their modifi-cation. The analyses of the reduction sequence were conductedaccording to the various techno-economic phases that underline thestrategies of humans in managing their environmental resources(Geneste, 1985; Boëda, 1986, 1994b; Jaubert and Bismuth, 1996;Mourre, 1996, 1997; Geneste et al., 1997), their emphasis on sometechnological aspects (Jaubert and Mourre, 1996) and/or the coexis-tence of different flaking techniques (Jaubert and Farizy, 1995).

4. Data

4.1. Variability of the Neanderthal occupations on the right bank ofthe middle Rhone Valley

While it is sometimes difficult to determine whether site occu-pation took place continuously in semi-permanent camps or inter-mittently in short-term regular camps, some faunal and lithic

criteria provideanestimateof thedurationofoccupationevents (seeTable 14 in Daujeard and Moncel, 2010). Firstly, alternating accu-mulations by carnivores and humans is one of the criteria that tendto be associated with short duration occupation, especially ifdamage attributable to both these predator groups is present on thesame bone elements, indicating the pene-contemporaneity ofconsumption. A quantitative estimate of human activity indices(human bone modifications, hearths and fire use, artefacts) is alsoinformative about the type of occupation, whether residential,seasonal or transient. The completeness of the butchery sequenceand of the lithic reduction sequences also indicates the type ofoccupation event. Finally, the raw material (both faunal and lithic)procurement activities and the toolkitmanagement strategies allowan interpretation of functional variability between sites to be made(i.e. Kuhn, 1992; Conard and Prindiville, 2000; Costamagno et al.,2006). In correlating the information gathered from the faunal andlithic assemblages in theMiddle Palaeolithic units of the sites on theright bank of the middle Rhone Valley, three categories can bedistinguished by their occupation duration (Daujeard, 2008;Daujeard and Moncel, 2010; Fig. 3). The Neanderthal occupationswere either very brief stays (Balazuc, Abri des Pêcheurs), regularhunting camps of short duration that alternated with periods ofcarnivore occupation (Payre-F, Baume Flandin, Baume d’Oullins,Ranc Pointu 2, Le Figuier), or theyweremore permanent residentialsites (Saint-Marcel, Abri du Maras) (Moncel, 2008; Moncel et al.,2008a, 2010; Rivals et al., 2009). Regionally, the large habitats ofBaume Néron and Abri Moula VIII (Soyons, Ardèche) could also beassociatedwith long-termoccupation related to specialised huntingof reindeer herds (Defleur et al., 1994, 2001). The site of Orgnac 3(MIS 9e8, Ardèche;Moncel and Combier,1992; Forsten andMoigne,1998; Moncel et al., 2005) belongs to the category of the regularseasonal camps for hunting aurochs and horse herds, interspersedwith visits by carnivores at the bottom of the sequence. The briefstopping-places in restricted and steep-sloping habitats was alsoidentified several times in the Hortus cave (pit locus; de Lumley,1972a,b; Boulbes, 2003). The data point towards bivouacs, duringwhich the abundant ibex carcasses were apparently preferred bycarnivores but shunned by hominids in favour of other rarer specieswhich were either killed or scavenged (specifically cervids).

Apart from varied tool proportions and numbers of boneretouchers, the variability observed in the occupationmodes cannotbe due simply to cultural traditions expressed by technologicalbehaviour (knapping modes and tool types). For similar modes, thelithic assemblages reflectdifferent reduction sequencesanddifferentresulting products (Fig. 3). The data relating to raw material prove-nance have a firm link with the type of occupation and its duration.Activities associatedwith theartefacts suggest, in themain, localflintgathering as well as some exploitation of certain types of stones ata secondary level. In many sites the local flint is suitable for mostknapping purposes and its procurement was preferentially centredon the southern plateauwhere flint abounds whereas quartz, basalt,quartzite and limestonewere collected in the lower valleys. Unit F inPayre, for example, which is considered to be a short-term regularcamp (Daujeard andMoncel, 2010), exhibits major flint exploitationof deposits less than 20 km to the south (Fernandes and Moncel, inMoncel, 2008). Local rocks (including basalt, quartz and quartzite)were brought to the site not only as tools but also as large flakes,which implies recurring and far-sighted planning behaviour(Moncel, 2008). The large dimensions of some of these tools may belinked to the processing of large to very large ungulates (such ashorse, bison, rhinoceros or proboscideans). Where flint is rare, forexample at the Abri des Pêcheurs which appears to be a brief stop-ping-place rather than a place of long occupation, other stone typeswere occasionally used (e.g. quartz), and flint was brought fromdistant sources in the form of unretouched or retouched flakes

C. Daujeard et al. / Quaternary International 252 (2012) 32e47 37

Author's personal copy

(Fernandes et al., 2010). Resharpening of tools is rare in theseassemblages and thus it is not possible to correlate an intensiveexploitationofflintwithdistantorigins as in thecaseof Pradelles andsimilar sites in southwestern France (Costamagno et al., 2006).

The variability observed in the types of occupation sites in thisregionof theRhoneValley, their seasonalrecurrenceandtheirstability(whatever their duration) within the long sequences of the MiddlePalaeolithic and the regional mobility range (50e60 km according tothe raw materials), allow association of the mode of mobility witha logistical system of territorial organisation e in the sense of func-tional and planning aspects (Fig. 3). However, uncertainty about thegroup size (probably family groups at Payre according to the humanremains, cf. Daujeard and Moncel, 2010), and also about the sexualdivision in the tasks of the inhabitants during these past periodsmakes it difficult to determine accurately if the population move-ments correspond to a system such as the logistical type described byBinford (1982). This uncertainty, added to the fact that in all the sitesthe lithics and fauna were procured locally (except for some exoge-neousmaterials) aswell as the fact that the occupations appear ratherresidential, allows attribution of this land use pattern to another typeof mobility being a far-sighted circulating model combined witha global and cultural geographic approach in which the micro-func-tional details are linked to the individual aspects of each site (cf.Binford,1982; Kelly,1995). This territorial management strategy linksboth non-specialised occupations with planned land use.

4.2. Sainte-Anne I and Baume-Vallée: two mid-mountain sites

4.2.1. Palaeo-environmental dataAt Sainte-Anne I cave, unit J2 yields the same ungulate species as

unit J1 (Guadelli, in Raynal et al., 2005, 2008; Raynal, 2007). The

faunal group is dominated by the reindeer, the horse and the ibex.An extinct species, the woolly rhinoceros, is also present (Table 2).From a palaeo-environmental viewpoint, the most important partof the spectrum is occupied by the open arctic fauna and themountain fauna groups, together suggesting the existence of harshand severe climatic conditions, such as those which would havedominated during MIS 6. The presence of the red deer, a dominantspecies among the forest fauna group (Delpech et al., 1983) may besurprising in a cold and steppe environment. This is due to the factthat the compartmentalised relief may have isolated some forested,leafy refuge zones, where red deer could continue to exist despitevery severe climatic conditions. This possibility is supported by theubiquitous character of this Cervid and also by palynological anal-yses which identified a very weak ratio of afforestation in manysites where the red deer is present. Carnivores are quite rare (%NISP), and comprise the fox, the wolf, the lynx and the extinct cavelion (Table 2). A small coprolite (attributed to a fox or a wolf) hasbeen recovered from unit J2.

At Baume-Vallée (Guadelli, Fiore and Tagliacozzo, in Fiore et al.,2005; Raynal et al., 2005), the horse (Equus caballus germanicus) isthe prevalent species, followed by the Cervidae (Cervus elaphus andRangifer tarandus), ibex (Capra ibex), the Bovinae (Bos or Bison) andother equids (Equus hyndruntinus), while the remainder of theassemblage is composed of bird species (Gala et al., 2005) andindeterminate carnivore fossils.

4.2.2. Taphonomical dataIn the two upper units of Sainte-Anne I cave, the extreme

fragmentation and poor surface preservation of faunal remains areessentially due to periglacial processes (Guadelli, 2008) whichwerefurther exacerbated by the hydric conditions in the cave. In J1, post-

Fig. 3. The primary herbivore species and the main knapping reduction sequences in the studied Middle Palaeolithic sites of the Massif Central and middle Rhone valley, accordingto age and type of occupation.

C. Daujeard et al. / Quaternary International 252 (2012) 32e4738

Author's personal copy

depositional alterations restrict the zooarchaeological interpreta-tions. However, some remarks have beenmade (for J1 see Fiore andTagliacozzo, in Raynal, 2007). Some remains show striations andfractures, indicating human butchery and bone modification, whileonly two indeterminate remains show carnivore tooth traces.

In J2, the ratio of identification is very low, while the bonedestruction is very high (Table 3). The abundance of isolated teeth isa consequence of differential preservation. Desquamations, frost-fractures, concretions and polish are the natural bone alterationsmost often identified, making almost a quarter of the materialindecipherable (Table 3). Nevertheless, the degree of differentialpreservation is limited by the presence of immature teeth, the highpercentage of spongy bones and nor by the very good surfacepreservation on a large half portion of the material.

Only seven bones are whole (exclusively short foot bones).Fragmentation that occurred at the time of deposition (green bonefractures) and post-depositional (dry bone factures) occur in equalproportions. Among the green bone fractures, traces of percussion(notching, flaking, etc) are more abundant than bite marks(Table 4). With respect to the study of traces left by carnivores andhumans on the bone surfaces, unidentifiable bones and isolatedteeth have been excluded. Cutmarks are more frequent thancarnivore tooth-marks (Fig. 4; Table 5). The proportions varyaccording to the species, the reindeer being the focus of humanactivities such as skinning, dismembering, defleshing, scraping of

the metapodia and marrow extraction (Fig. 5; Table 6). Theextraction of marrow from the long bones, mandibles and thephalanges e parts which preserve their fat for the longest time andwhich contain the highest proportion of non-saturated fatty acids,indicates an optimal exploitation of carcasses.

Concerning the location of the butchery marks on the ungulatescarcasses, the shafts of the long bones are the most affected, whilethe carnivore marks are prevalent on the epiphysis (Table 7). Suchpartitioning of damages suggests that carcasses were primarilyconsumed by humans in the cave, while carnivores gainedsecondary scavenging access to kills (Blumenshine, 1986, 1988; cf.Guadelli, in Raynal et al., 2008). This hypothesis is supported by thesuperposition of tooth-marks over butchery striations on one of thesix bones which have both cutmarks and tooth-marks together. Notrace of fire has been identified on the bone material, but thepresence of fragments of burnt flint in the deposits attests toa controlled use of fire. Black colouration on the bones, caused bymetallic oxides and organo-mineral compounds may have obliter-ated traces of burning on the bone material. Six bone retouchershave been recorded (0.4% of coordinated bone remains). Threebones are from very large-sized ungulates and three others arefrom medium-sized ungulates.

The data indicate that the majority of the bones were accumu-lated by human activity. The proportions of butchery marks andcarnivore tooth-marks indicate that humans had primary access tothe carcasses. Intensive marrow extraction is evident on a largeproportion of the material. An almost complete butchery sequencehas been identified for reindeer, but only secondary steps of thecarcasses treatment were processed into the site for horse andother taxa.

Of course carnivores have certainly played their part in intro-ducing bones into the cave, but modification by them seemsminimal. Only a few bone remains can be identified as those ofcarnivores (% NISP; Table 2). Until now only one coprolite has beenrecovered from J2 but poor preservation conditions may be thecause of the low recovery rate. While the data do not point to anintense use of the cave as a carnivore’s den, they at least suggesttheir recurrent presence there.

At Baume-Vallée, periglacial processes have also caused signif-icant surface abrasion and fragmentation to the faunal assemblage(Guadelli, 2008). The only complete bones are small ones fromjoints and the phalanges. Articular surfaces of long bones are veryrare, but isolated teeth are abundant. In spite of the poor state ofpreservation, butchery processes have been identified (marrowextraction, defleshing) as well as the use of certain pieces asretouching tools (25; 0.6% of coordinated bone remains). Carnivoremodification is very rare and the majority of the fresh bone fractureis attributable to human activity. Very few burnt bones have beenrecorded (Fiore and Tagliacozzo, in Fiore et al., 2005; Raynal et al.,2005) but the presence of numerous charcoals and burnt flint

Table 2Relative abundance for the macro-mammals at Sainte-Anne 1e J2 (NISP; Number ofIdentified Specimens; NR; Number of total Remains; MNI; Minimum Number ofIndividuals per taxon; Indet.; Indeterminated). For the unidentified ungulates thatmay be present at Sainte-Anne I, they are grouped in four size classes (from Bunn,1986, modified); Small size ungulates (<100 kg; sizes 1 and 2); roe deer, boar,chamois, fallow deer, ibex; Middle size ungulates (100e300 kg); red deer, reindeer;Large size ungulates (300e1000 kg); horse, bovines; Very large size ungulates(>1000 kg); rhinocerotids, proboscideans.

Taxa (SA1-J2) NISP %NISP MNI %MNI

Canis lupus 4 1.7 1 4.5Vulpes vulpes 4 1.7 1 4.5Panthera spelaea 2 0.9 1 4.5Lynx spelaea 1 0.4 1 4.5Carnivores indet. 6 2.6 0 0.0Carnivores 17 7.2 4 18.2

Coelodonta antiquitatis 2 0.9 1 4.5Equus caballus cf. piveteaui 72 30.6 6 27.3Equus hydruntinus 1 0.4 1 4.5Perissodactyla 75 31.9 8 36.4

Rangifer tarandus 119 50.6 6 27.3Cervus elaphus 5 2.1 1 4.5Cervidae indet. 4 1.7 0 0.0Bovinae 1 0.4 1 4.5Capra ibex 14 6.0 2 9.1Artiodactyla 143 60.9 10 45.5

Total NISP 235 100.0 22 100.0Small size ungulates 6Middle size ungulates 899Large size ungulates 118Very large size ungulates 1Ungulates indet. 17Large size rodents 1Remains indet. 237Total NR (coordinated) 1514Total NR (with sieved fragments) 6290

Table 4Percentage of bone remains with green bone fractures (NRg), with percussionnotches (NRp) or with carnivore marks (NRc), and with dry bone fractures (NRd)(NR; mandibles, limbs bones and shaft fragments >5 cm without complete bones).

SA1-J2 NRp NRc NRg NRd NR

NR 165 48 814 688 1358% 20.3 5.9 59.9 50.7 100

Table 3Bone preservation and fragmentation rates and percentages of remains with the most severe natural alterations (NISPa; included anatomical identified remains).

SA1-J2 NISPa/NR Isolated teeth/NISPa Spongy bone Complete bones Concretions Polish Desquamation Gelifracts Unreadable remains

NR 340 152 135 7 132 105 304 323 377% 5.4 44.7 29.7 3.7 9.7 7.7 22.4 23.8 24.9

C. Daujeard et al. / Quaternary International 252 (2012) 32e47 39

Author's personal copy

confirms the use of fire in the site. Microwear analyses of approx-imately 300 pieces of flint from unit 1 demonstrate that the toolswere used for cutting, thinning and scraping resources includingsoft animal materials, herbaceous plants or soft wood (Lemorini, inRaynal et al. 2005).

4.2.3. Acquisition strategies for animal resourcesIn unit J1 of Sainte-Anne I, horse remains are dominant (52%

NISP), followed by reindeer (20% NISP), while deer and ibex aremore rarely seen. All the horse cranial elements recovered belongto adult individuals and consist mainly of long bones as well as anatlas and a scapula. The majority of the bones recovered frommedium-sized ungulates are long bones and elements of thescapula, coxal and axial skeleton all being equally represented. Thispattern of skeletal remains suggests that entire carcasses ofmedium-sized animals were brought to the site for processing butonly the largest parts of horse carcasses (Fiore and Tagliacozzo, inRaynal, 2007).

In unit J2, the main species of the herbivore spectrum is thereindeer whose remains are the most common (almost 50% of theNISP) followed by equids which count the same number of indi-viduals (Table 2). Ibex, woolly rhinoceros, bovines and other cervidscomplete the spectrum. In spite of the fact that there is one primaryherbivore species, the faunal acquisition strategy probably exploi-ted all available environments around the sites including steppeplateaus and rocky landscapes. All age classes are represented forreindeer and horse remains with the majority of these animals

being young adults (Table 8). Some deciduous teeth (NR¼ 4) indi-cate autumn killing periods. Ibex remains represent at least twoindividuals, a mature adult and an older adult. For bovines, at leastone individual is represented. Deer and the hydruntine horse areeach represented by one mature individual. It is not possible to givedefinitive age assessments for the other taxa.

Concerning the skeletal profiles, cranial remains are dominantonly because of the quantity of isolated teeth, especially for thehorse (Fig. 6). The almost-total absence of the axial skeleton in alltaxa could be due to the difficulties of taxonomic attribution. Thispart is well illustrated among the categories of medium-sizedungulates. The fore- and hind-quarters are equally represented forhorse and reindeer. Metapodials are very rare for the horse, but arethe dominant element for reindeer. The under-representation ofthis element for the horse could be a function of the difficulty inidentifying the fragments of long bones of the large ungulates. Atthis stage of the study, it is difficult to infer from the skeletalprofiles how the carcasses that were transported were used. Someconsiderations similar to those already published for the assem-blage of unit J1 (Fiore and Tagliacozzo, in Raynal, 2007) areproposed here. The rarity of some elements of the horse (axialskeletal and foot bones) indicates the introduction into the site ofcertain favoured portions, especially the head, hind and forequarters, while the presence of almost all carcass elements of thereindeer suggests that entire carcasses were more often trans-ported back to the site.

At Baume-Vallée hunting focused mainly on cervids and equidsduring the first period of human occupation, while equids becomethe dominant species hunted later. The ibex, which is in thirdposition numerically, is present throughout. All age classes of thesethree species are represented, from juveniles to the oldest. In themost recent phases the majority of horse remains represent youngor mature adults. As at Saint-Anne I, skeletal profiles indicateprovisioning using entire carcasses of medium-sized ungulates, butonly portions of the equine carcasses (Fiore et al., 2005).

Fig. 4. Small sized ungulate shaft fragment showing tooth-marks in the J2 unit of Sainte-Anne I (P30, 191; scale¼ 1 cm).

Table 5Percentages of bone remains with carnivore marks (NRc) or anthropic cutmarks(NRa) (NR; Number of observable remains >5 cm without isolated teeth).

SA1-J2 NRc NRa NR

NR 48 117 981% 4.9 11.9 100

Fig. 5. Reindeer femur shaft fragment showing defleshing cutmarks in the J2 unit of Sainte-Anne I (P25, 845; scale¼ 1 cm).

C. Daujeard et al. / Quaternary International 252 (2012) 32e4740

Author's personal copy

4.2.4. Animal procurement and treatment strategiesIn both sites, the presence of all parts of the skeleton processed

during primary and complete butchery sequences, at least for oneof the main species, does more support the hypothesis of a habitatsite than of a specialised site for the processing of carcasses, as hasbeen recently demonstrated at Pradelles for example, in the South-West of France (Costamagno et al., 2006; Meignen et al., 2006). Thediversity of the ungulate range, age categories and the presence ofupper limb bones among the remains of the principle taxa opposesthe hypothesis of a specialised killsite, such as the site of Mauran(260 m, Haute-Garonne; Farizy et al., 1994; Rendu, 2010). On thecontrary, the diversity of the ungulate spectrum argues for anopportunistic style of hunting such as has been observed in the siteof Portel-Ouest (410 m, Ariège, “Petites-Pyrénées”, MIS 3), close toMauran (Gardeisen, 1999).

Apart from the exception of the lesser role played by carnivoresat Sainte-Anne I and Baume-Vallée, the type of ungulate exploita-tion resembles that seen at Portel. Similarities are due not only tothe diversity of ungulates which were exploited, but also to thedominant presence of young and mature adults, to the primaryaccess to either whole carcasses or meat-bearing upper fore- andhind limbs leading to a conclusion of site usage concentrating onintensive and systematic in situ butchery.

At Sainte-Anne I, seasonal indicators such as deciduous teethsuggest an autumn habitation period. Occupation of the sites at thiselevation has probably been sporadic, since winter snow-cover onthe surrounding plateaus would certainly limit or even completelyprevent all activities. The rarity of carnivore marks on bones fromseasonal sites can be explained by human occupations very closelyspaced in time, or by a late arrival of humans just prior to thecommencement of severe winter conditions that limited theactivity of scavenging animals. The use of fire at Sainte-Anne I andBaume-Vallée is confirmed by the presence of burnt flint. At themoment the absence of burnt bones from these sites is still unex-plained, possibly due to preservation problems including peri-glacial post-depositional processes or difficulties in attributing thecause of black staining on bone surfaces. In both sites, boneretouchers are present in similar proportions.

All the elements argue for the utilisation of the sites of Baume-Vallée and Sainte-Anne I as seasonal hunting camps, frequentlyinhabited during the milder periods of the year.

4.2.5. Lithic raw material procurement and treatment strategiesAt Sainte-Anne I, quartz and volcanic and certain types of local

flint, exhibit complete reduction sequences indicating that theseabundant local lithic materials were flaked within the site. Incontrast, other lithics from semi-local and exogenous sources show

incomplete reduction sequences and, in some way, were preparedelsewhere before appearing in the site. Easy access to local rawmaterials permitted the production of a comparatively largenumber of tools. Comparison between units J2 and J1 shows nosignificant differences (Raynal, 2007), except that the number oftools made on volcanic pebbles varies between units J2 and themore recent J1 but this may be an excavation bias. On the otherhand, polyhedral cores are more numerous in J1 but the proportionof debitage remains unchanged. If the discoid reduction sequencefollows the same procedures, then the Levallois reduction sequenceshows recurrent centripetal flaking with the final preferential stageof debitage being a large Levallois flake. The techniques used areopportunistic and the strategies are adapted to themorphology andflaking characteristics of the stone used. However, chronologicallyspeaking, the assemblages are very close in time. From a techno-typological point of view, the lithics resemble the series recoveredfrom the Ardèche site of Payre, along the Rhone Valley where theraw materials were chosen for their proximity rather than for theirquality (Moncel, 2001; Raynal et al., 2005; Fernandes et al., 2008).

The limited use of the Levallois method is unlikely to bea chronological criterion, the occurrence of which has beenobserved throughout the Massif Central and Pyrénées (Jaubert,2001). The low proportion of bifaces is attributed to the type ofmaterial being utilised and so this kind of tool does not offer muchhelp in chronological terms (Moncel, 2001; Turq, 2001). At Baume-Vallée, in the Ferrassie type Mousterian occupations, flint wasmore-or-less the exclusive raw material, with minimal numbers ofquartz and volcanic lithics. Flint knapping techniques were pref-erentially the Levallois and Quina methods.

Table 8MNI for every age category of the main herbivores (for individual age determinationthe replacement pattern and wear stages have been used; see Grant, 1982 or Kleinand Cruz-Uribe, 1984; among others).

MNI (SA1-J2) Juvenile Youngadult

Matureadult

Oldadult

Very oldadult

Total

Equus caballus 2 2 1 1 6Equus hydruntinus 1 1Rangifer tarandus 2 1 1 1 1 6Cervus elaphus 1 1Capra ibex 1 1 2Bovinae 1 1

0 10 20 30 40 50 60 NISP

HEAD (withoutisolated teeth)

HEAD (withisolated teeth)

AXIAL

FOREQUARTER

FOREFOOT

HINDQUARTER

HINDFOOT

IND FOOTEquusRangifer

Fig. 6. Representation of the six anatomical portions for reindeer and horse in the J2unit of Sainte-Anne I (% NISP).

Table 6Percentages of anthropic and carnivore marks and green bone fractures for the twomain ungulates.

SA1-J2 NRc NRa NRg

Equus caballus 5.3 21.1 39.1Rangi fertarandus 10.4 28.4 78.2

Table 7Distribution of shaft zones with anthropic cut-marks (NZs) and/or tooth-marks(NZc) along the ungulates long bones (NZ; Number of observable Zones).

Epiphyses Diaphysisextremities

Mediandiaphysis

Total

NZa NZ % NZa NZ % NZa NZ % NZa NZ %1 18 5.6 32 154 20.8 70 611 11.5 103 783 13.2NZc NZ % NZc NZ % NZc NZ % NZc NZ %2 18 11.1 11 154 7.1 24 611 3.9 37 783 4.7

C. Daujeard et al. / Quaternary International 252 (2012) 32e47 41

Author's personal copy

Such petro-archaeological studies contribute to a better under-standing of the mode of exploitation of lithic raw materials byNeanderthalers. In the Massif Central, flint is not abundant fromprimary geological sources, being found only in lacustro-palustrinebasins. The most abundant lithic resources occur in old alluvialformations belonging to a vast drainage system along the Southernand Eastern margins of the Massif (Lozère and Ardèche) wherelimestones of the Lias and Dogger series contain abundant flintsources. In contrast, the middle Rhône Valley offers an abundantand varied lithic potential. At Sainte-Anne I, raw materialprocurement took place at more than thirty different localitiesclose to the site in the Puy-en-Velay basin as well as at secondaryand sub-primary colluvial and alluvial outcrops. The Nean-derthalers obviously had an excellent knowledge of the occurrenceof potential local resources.

The presence of some specific flint types suggests the use ofexploitation routes which crossed the borders of fluvial systems.The frequentation of the western highlands of the massif and of thevalleys of the Loire, the Dore and the Allier Rivers demonstratesthat there were some northesouth routes. The frequentation of thevalley of the Allier is suggested by evidence, limited though it is, offlint from Madriat, Arlanc and La Comté. One piece indicatesa journey to (or from), or trade with, the hills of the Bas-Beaujolaisat Alix in Rhône, close to the northern Rhône Valley. The persis-tence of a number of procurement circuits e local, regional andmore remote is also attested in the Ferrassie type Mousterianindustry of Unit 1 at Baume-Vallée, dating to approximately MIS 5.The same flint types are represented in this site and the procure-ment territories are similar in extent. Behavioural differencesdetermine the choice of the most utilised raw material which wascollected from secondary Jurassic flint outcrops. Perhaps thisindicates a different kind of mobility for these Mousterian groups.Raw material procurement in these two mid-mountain sites iswidespread, similar in size to what is observed for the site of Abrides Pêcheurs in Ardèche on the eastern margin of the massif.Conversely, the flint procurement observed at Payre and Saint-Bauzile is less widespread and focussed on flint-rich limestoneformations along the Rhône corridor. Presently, using petro-archaeological analysis, no circulation of materials between theRhône Valley and Velay within the Massif Central has beendiscovered. The sites of Baume-Vallée and Abri du Maras havesimilar material cultures which remain to be studied and comparedwithin the region to gain a more detailed understanding of regionalexploitation patterns. The territorial perspective provided by ananalysis of the lithic raw materials shows only one aspect of thepaleo-ethnographic activities of very mobile Neanderthal groups.

4.3. Regional land use and occupation types for mid-mountain sitesof the southeastern Massif Central

A great diversity of both local materials and those from somedozens of kilometres afield is evident in the exploitation of lithicresources during theMiddle Palaeolithic in the southeastern MassifCentral (Figs. 7 and 8). Some rarer pieces indicate extensive travelor trade, of 100 km ormore. Materials imported from outcrops withworkshops illustrate the complex economics of Neanderthalsociety while the presence of exogenous stones abandoned on thistype of site suggests knowledge of an even wider world. Despitethis suggested complexity, the importation of distant materialsdoes not reflect planning strategies comparable to those observedelsewhere (Bourguignon et al., 2006).

Large-scale regional mobility is suggested beginning from thetime of MIS 6 and continuing until the end of the Middle Palae-olithic. At Sainte-Anne I as well as Baume-Vallée, human groupscollected flint from more than 30 km afield and from 12 different

outcrops. The low volume of material brought from a long distanceand the multiplicity of the outcrops used suggests that stoneprocurement took place during other subsistence activities (Fig. 8)rather than as a result of a systematic collection strategy. As localprocurement strategies are largely dominant at these sites, theyoccupy favourable and preferred locations and around them andwithin a perfectly well-known environment, seasonal activitieswere organised (Raynal, 2007).

Modern climatic data (Fillod, 1985) show that the different lithicsources are located within the most welcoming areas during thesummer, including the basin of Puy-en-Velay, and the Allier andLoire Valleys. The flint sources of Malzieu and Naussac are locatedinwinter snowy areas and thus virtually inaccessible in that season.The fauna associated with the stone industries generally indicatesa global cold climate it seems likely that human occupations tookplace only during favourable seasons (Raynal et al., 2005).

The two technological concepts, namely Levallois and discoidalflaking techniques, occur on both sites although applied to differentrocks in different proportions. However, the support obtained fromthe two techniques does not illuminate the question of why tworeduction sequences were used.

Within the two small shelters, both with favourable orienta-tions, at Baume-Vallée and Sainte-Anne I, there are several indi-cators suggesting short-stays as opposed to longer ones. These arethe frost-susceptible properties of the rock walls which erodequickly, climatic constraints, the manner in which the raw mate-rials were managed, the limited spatial organisation, the scarcity ofsigns of fire use and the presence of carnivore marks on bones.However, the scarcity of these features in the presence of completelithic and faunal reduction sequences does not suggest they wereused as localities where brief stops were made within a largeterritory. Indeed, in view of the criteria given in Table 1, the prop-erties of these two medium altitude sites suggest that they aretemporary camps occupied for short durations during seasonalhunting activities, such as at Le Portel-Ouest (Gardeisen, 1999),rather than brief stopping-places as suggested for Noisetier cave(Mourre et al., 2008). According to the current data, rather thanbeing specialised sites (kill sites or carcass processing locations),the sites were regularly used short-term camps in which wolvesand other carnivores also found occasional refuge and scavengedthe carcasses abandoned by humans (Fig. 3).

This type of residential pattern is also suggested by the occur-rence of a limited number of exhausted cores of local raw materialused to produce quantities of standardised flakes. If the duration ofoccupation events could be judged by a large number of artefactsproduced on local volcanic rocks, quartz and types of flint, theabsence from the stone tool assemblage of certain items from thereduction sequences, such as large-sized and retouched flakes,indicates the use of certain of these products away from the site.The conclusion must be that these objects were taken away fromthe site as the occupantsmoved through their territories. Some flinttypes can even be considered as ‘mobile’ or ‘portable’ productswhen flaking workshops appear on outcrops (as flint from LeMalzieu, Madriat, Laps or Arlanc for example), or even as territorialindicators (e.g. the flint from Bas-Beaujolais).

5. Discussion

5.1. Land use between mid-mountains and plains in southeasternMassif Central

Compared with the sites of the plains and low plateau of themiddle Rhone Valley, the higher sites of Sainte-Anne I and Baume-Vallée have traits similar to those of the short-term regular campsof Payre, Le Figuier and Baume Flandin. As in the plains examples,

C. Daujeard et al. / Quaternary International 252 (2012) 32e4742

Author's personal copy

the mid-mountain sites do not show specialised occupation activ-ities. With the exception of the open-air workshop sites, such asSaint-Pierre-Eynac (Haute-Loire), the human occupation at higheraltitudes took the form of temporary residential camps. Thediversity of flint outcrops used in the mountain territories suggestsextensive mobility among human groups, governed by seasonalcontrasts and the extent of the utilised territories. Local collectingof raw materials is also a trait of the sites in the middle RhoneValley. A comparative analysis of the two Middle Pleistocene sitesof Payre and Sainte-Anne I have already concluded that land use(mineral and animal) followed the same rules in both cases (seeFernandes et al., 2008; Raynal et al., 2008). However, the occupantsof Sainte-Anne I and Baume-Vallée located in areas relatively poorin primary flint resources needed to utilise varied lithic sources,including flint derived from areas further afield. The Ardèche, Loireand Allier hydrographic basins lie close together and there is nobarrier to human circulation between the Rhone corridor and smallriverine basins inside the Massif Central (Raynal et al., 2005). Thus,the hypothesis for the altitude settlements includes an enlargedpeopling of the mid-mountain area by people from the plains andlow plateaus of the middle Rhone Valley during favourable periodsof the year. The human groups using the two regions employedsimilar exploitation strategies, so there is no need to postulate thatthe population came from further afield. Another less-likelyhypothesis could be that there were perennial occupations of theMassif Central (at low or medium altitudes according seasons) by

human groups adapted to this environment, and whose exploita-tion strategies were the same as those of the occupants on theborders of the Rhone Valley. Further research on lithic rawmaterialand faunal exploitation, on the composition of the tool kits and onthe choices of the flaking modes may require a revision of thesepreliminary results. However, at present, there are no suggestionswithin the stone tool sequences that suggest a dramatic reversalwill be required.

5.2. Land use between mid-mountains and plains in othermountainous massifs

In comparison to these conclusions, the occupation of the Pyr-enees Mountains and foothills during the Middle Palaeolithicprovides a different and more diversified picture, in which bothspecialised and residential sites occur. There are short-term regularhunting camps as in the southeasternMassif Central, for example atLe Portel-Ouest (mid-mountain; Gardeisen, 1999); kill-sites atMauran (foothills); brief stops during journeys as at Grotte duNoisetier (mid-mountain; Mourre et al., 2008) and at Teixonerescave (plain; Moia, Spain; Rosell et al. 2010), on the other side of thePyrénées; hunting stop at Grotte de Gabasa (mid-mountain;Huesca, 780 m; OIS 3; Utrilla and Montes, 1989; Utrilla et al., 2006)and a long-term residential camp at Abric Romani (plain; Vaqueroet al., 2001). The massif of Vercors (Bernard-Guelle, 2002, 2005)shows a type of occupation closer to what can be observed in the

Fig. 7. Stone gathering area following flint typology of Sainte-Anne I (OIS 6) and their presently accessible sources. Faunal diversity and possible routes and links to other domains atthe ‘‘limiting outcrops’’.

C. Daujeard et al. / Quaternary International 252 (2012) 32e47 43

Author's personal copy

southeastern Massif Central, namely a system of territorial occu-pations in which economic exploitation of resources could beorganised around seasonal movements of the primary camps sit-uated in the valleys combined with the use of secondary mountaincamps (Bernard-Guelle, 2002). This system implies a “strongmobility resulting from the frequentation of varied and contrastingenvironments, and was timed by seasonal migrations of the biggame animals. The relative isolation explains the establishedstability in the behaviour and the technical traditions” (translatedfrom Bernard-Guelle, 2005).

However, some noteworthy differences should be highlighted.Firstly, the movement of materials between the two environments(mountains and neighbouring valleys) has been demonstrated forVercors, but the scale of movement is not at all the same. It mustalso bementioned that the locations of true base camps in the plainare unknown to date; (conversely see the long-term residentialcamps at Saint-Marcel or Abri du Maras for the middle Rhone

Valley). Finally, the existence of numerous specialised open-airworkshop sites for flint makes an exception of Vercors which,combined with its attractive location, may explain the humansettlement there. However, this work has highlighted the existenceof lithic procurement sites with workshops in the Cévennes and thelower Auvergne, wherever significant resources occur (such asMalzieu, Madriat, Laps, Arlanc, Saint-Pierre-Eynac, Le Mazet, etc).At this stage in the research, a comparative analysis such as thisdemands caution in its interpretations. The open-air lithic depositsof the Rhone Valley are still rare, but the sites of Lagorce and Saint-Bauzile show that this type of exploitation site did exist (Monceland Perrève, 1999; Durand et al., 2009).

5.3. What types of settlements occur in mountainous territory?

The data do not suggest that the peopling of themid-mountain isrelated to specific types of occupation as seen for the sites on the

Fig. 8. Minimum stone gathering area (msga) following flint typology of unit 1 of Baume-Vallée (OIS 5a) and their presently accessible sources. Possible routes and links to otherdomains outside the subsistence sphere. Other close Mousterian sites: Le Rond de Saint-Arcons (RSA) and Rochelimagne (R) (from Fernandes et al., 2008).

C. Daujeard et al. / Quaternary International 252 (2012) 32e4744

Author's personal copy

plains. The availability of seasonal hunting territories perhapsexplains the intermittent occupation of the rock-shelters and cavesregularly used as camps by small and mobile groups exploiting theavailable diverse biotopeswith theminimumof excursion time. TheNeanderthal occupation of Velay is similar to that of the Vercorsmassif, being tied to seasonal forays searching for certain ungulatespecies as they sought summer grazing at higher altitudes (deer, roedeer, wild boar or mountain sheep) (Bernard-Guelle, 2005).

6. Conclusion

Amultidisciplinary approach focused on procurement strategiesand processing of animal and lithic resources is a profitable way tostudy occupation types and the kinds of relationships between themid-mountain areas, their foothills and neighbouring plains duringthe Middle Palaeolithic in the southeast of the Massif Central andthe middle Rhone Valley.

This type of analysis must be approached cautiously sincenumerous factors must be considered including e an imbalance infield surveys, incomplete data from excavation, scarcity of open-airsites, absence of strict contemporaneity between sites, differentialstates of preservation and so on. Nevertheless, combined studies ofthe raw material, the technical choices and lithic productionsystems, strategies of animal exploitation can now be developed inthis area using new excavations as well as by re-examination ofpreviously collected data.

Proximity to and accessibility of territories defined by thehydrological network and slightly accentuated relief (catchmentsand riverine basins), along with a potential to share adjacentterritories as demonstrated by rawmaterial procurement strategies(Fig. 7), adds to the necessity for stringent questioning to obtainmore accurate palaeo-ethnographic understandings. In Velay, theexistence of mid-mountain temporary hunting camps, similar tothose observed in the plains of the middle Rhone Valley, suggestsoccupation modes and perhaps waves of peopling motivated torespond to environmental conditions and changes by identicalfactors.

On the basis of current data, the occupation of the mid-moun-tain Massif Central area between MIS 6 and 3 does not differ fromthat which has been established for other massifs such as Vercorsor the Italian Piedmont and suggests the expansion of huntingterritories into these higher altitudes during favourable seasonsrather than any occupation adapted to a specific mountainousenvironment.

These preliminary results require confirmation based upon newexcavations and a strengthened multidisciplinary approach. Thiswill be developed through a collective research project (PCREspaces et subsistance au Paléolithique moyen dans le sud duMassif Central) coordinated between two of the authors (J.-P.R. andM.-H.M.) and will be closely aligned with the research developedamong the PCR Réseau de lithothèques en Rhone-Alpes coordi-nated by one of the authors (P.F.). Materials from other shelters andopen-air sites of Velay and caves of the right bank of the RhoneValley will be re-examined while new excavations will makepositive contributions to this research; as in the case of Abri duMaras (Ardèche). In the near future these cooperative approacheswill help to draw a more realistic picture of Neanderthal subsis-tence strategies in the southeastern Massif Central and the middleRhone Valley.

Acknowledgements

We sincerely thank S. Gaudzinski and L. Kindler for organizingthe session “Hominin subsistence” at the 11th ICAZ and for allowingus to participate. Part of this work relies on excavation projects,

specific surveys and studies funded by the French Ministry ofCulture and Communication, Région Aquitaine and communities ofHaute-Loire (CG43 and CCPM) and Ardèche. Thanks to the Prehis-tory Museum of Orgnac l’Aven and Jean Combier for permission towork on certain collections from the Ardeche. We greatly thank thetwo anonymous reviewers for their constructive comments. TheEnglish translation was made by Sally Reynolds and editing byPeter Bindon.

References

Arzarello, M., 2003. Contributo allo studio del comportamento tecno-economicodell’uomo di Neandertal: l’industria litica della serie musteriana del riparoTagliente (Stallavena di Grezzena, VR, Italia). Ph.D.Thesis. Università degli Studidi Ferrara, Italia.

Arzarello, M., Peretto, C., 2005. Nouvelles données sur les caractéristiques et l’év-olution techno-économique de l’industrie moustérienne de Riparo Tagliente(Verona, Italie). In: Molines, N., Moncel, M.H., Monnier (Eds.), Données récentessur les modalités de peuplement et sur le cadre chronostratigraphique, géo-logique et paléogéographique des industries du Paléolithique ancien et moyenen Europe (Rennes, 22e25 septembre 2003). British Archaeological ReportsInternational Series, 1364.

Bächler, E., 1940. Das alpine Palaeolithikum der Schweiz im Wildkirchli, Drachen-loch und Wildenmannlisloch. Basel, Monographien Zur Ur-Und Fruhgeschichteder Schweiz. Verlag Birkhauser & Cie.

Bernard-Guelle, S., 2002. Modalités d’occupation et d’exploitation du milieumontagnard au Paléolithique moyen: l’exemple du Massif du Vercors (Préalpesdu nord). Bulletin de la Société Préhistorique Française 99 (4), 685e699.

Bernard-Guelle, S., 2005. Territoires et mobilité des groupes moustériens en Ver-cors: analyse et discussion. L’Anthropologie (Paris) 109, 799e814.

Binford, L.R., 1982. The archaeology of place. Journal of Anthropological Archae-ology 1, 5e33.

Binford, L.R., 1984. Faunal Remains from Klasies River Mouth. Academic Press, NewYork.

Blumenshine, R.J., 1986. Carcass consumption sequences and the archaeologicaldistinction of scavenging and hunting. Journal of Human Evolution 15,639e659.

Blumenshine, R.J., 1988. An experimental model of the timing of hominid andCarnivore influence on archaeological bone assemblages. Journal of Archaeo-logical Science 15, 483e502.

Boëda, E., 1986. Approche technologique du concept Levallois et évaluation de sonchamps d’application: étude de trois gisements saaliens et weichséliens de laFrance septentrionale. Thèse. Université de Paris, France.

Boëda, E., 1994a. Le débitage discoïde et le débitage Levallois récurrent centripète.Bulletin de la Société préhistorique Française 90e96, 392e404.

Boëda, E., 1994b. Le concept Levallois: variabilité des méthodes. CRA-Monographie9. CNRS, France.

Boëda, E., Geneste, J.-M., Meignen, M., 1990. Identification de chaînes opératoireslithiques du Paléolithique ancien et moyen. Paleo 2, 43e80.

Bona, F., Peresani, M., Tintori, A., 2007. Indices de fréquentation humaine dans lesgrottes à ours au Paléolithique moyen final. L’exemple de la Caverna Generosadans les Préalpes lombardes, Italie. L’Anthropologie 111, 290e320.

Boulbes, N., 2003. Les grands herbivores de la grotte moustérienne de l’Hortus(Valflaunès, Hérault): Etude paléontologique, archéozoologique, paléo-écologique. Thèse. Université de Perpignan, France.

Bourguignon, L., Delagnes, A., Meignen, L., 2006. Systèmes de production lithique,gestion des outillages et territoire au Paléolithique moyen: où se trouve lacomplexité? In: Astruc, L., Bon, F., Léa, V., Milcent, P.-Y., Philibert, S. (Eds.),Normes techniques et pratiques sociales. De la simplicité des outillages pré-etprotohistoriques. XXVIèmes Rencontres Internationales d’Archéologie etd’Histoire d’Antibes. APDCA, Sophia Antipolis, pp. 75e86.

Boyle, K.V., 2000. Reconstructing Middle Palaeolithic subsistence strategies in theSouth of France. International Journal of Osteoarchaeology 10, 336e356.

Bunn, H.T., 1986. Patterns of skeletal representation and hominids subsistenceactivities at Olduvai Gorge, Tanzania, and Koobi Fora, Kenya. Journal of HumanEvolution 15, 673e690.

Burke, A., 2006. Neanderthal settlement patterns in Crimea: a landscape approach.Journal of Anthropological Archaeology 25, 510e523.

Cârciumaru, M., Moncel, M.-H., Angelhicu, M., Cârciumaru, R., 2002. The Borosteni-Cioarei Cave (Carpathian Mountains, Romania). Middle archaeological finds andtechnological analysis of the lithic assemblages. Antiquity 76, 681e690.

Combier, J., 1967. Le Paléolithique de l’Ardèche dans son cadre paléoclimatique.Publication de l’Institut de Préhistoire de l’Université de Bordeaux, Delmas, pp. 4.

Conard, N., Prindiville, T.J., 2000. Middle Palaeolithic hunting economies in theRhineland. International Journal of Osteoarchaeology 10, 286e309.

Costamagno, S., Griggo, C., Mourre, V., 1998. Approche expérimentale d’un prob-lème taphonomique: utilisation de combustible osseux au Paléolithique. Pré-histoire Européenne 13, 167e194.

Costamagno, S., Meignen, L., Beauval, C., Vandermeersch, B., Maureille, B., 2006. LesPradelles (Marillac-le-Franc, France): a mousterian reindeer hunting camp?Journal of Anthropological Archaeology 25, 466e484.

C. Daujeard et al. / Quaternary International 252 (2012) 32e47 45

Author's personal copy

Crégut-Bonnoure, E., Boulbes, N., Daujeard, C., Fernandez, P., Valensi, P., 2010.Nouvelles données sur la grande faune de l’Eémien dans le Sud-Est de laFrance.Q6 congress, Montpellier 2008. Quaternaire 21 (1), 227e248.

Daujeard, C., 2008. Exploitation du milieu animal par les Néanderthaliens dans leSud-Est de la France. British Archaeological Reports International Series, S1867.

Daujeard, C., Moncel, M.-H., 2010. On Neanderthal subsistence strategies and landuse: a regional focus on the Rhone Valley area in southeastern France. Journal ofAnthropological Archaeology 29, 368e391.

Debard, E., 1988. Le Quaternaire du Bas-Vivarais d’après l’étude des remplissagesd’avens, de porches de grottes et d’abris sous roche. Dynamique sédimentaire,paléoclimatologie et chronologie. Documents des laboratoires de géologie deLyon. 103.

Defleur, A., Valladas, H., Radulescu, C., Combier, J., Arnold, M., 1990. Stratigraphie etdatation carbone 14, en spectrométrie de masse par accélérateur, du Mous-térien récent de l’abri du Ranc de l’Arc (Ardèche, France). Comptes Rendus del’Académie des Sciences de Paris 311, 719e724.

Defleur, A., Bez, J.-F., Crégut-Bonnoure, E., Desclaux, E., Fontugne, M., Jeannet, M.,Magnin, F., Talon, B., Thinon, M., Combier, J., 1994. Industries, biostratigraphie,restes humains et datation du gisement moustérien de la Baume Néron (Soyons,Ardèche). Comptes Rendus de l’Académie des Sciences de Paris 318, 1409e1414.

Defleur, A., Crégut-Bonnoure, E., Desclaux, E., Thinon, M., 2001. Présentation paléo-environnementale du remplissage de la Baume Moula-Guercy à Soyons(Ardèche): implications paléoclimatiques et chronologiques. L’Anthropologie105, 369e408.

Delpech, F., Donard, E., Gilbert, A., Guadelli, J.-L., Le Gall, O., Martini-Jacquin, A.,Paquereau, M.-M., Prat, F., Tournepiche J.-F., 1983. Contribution à la lecture despaléoclimats quaternaires d’après les données de la paléontologie en milieucontinental. Quelques exemples de flore et de faune d’Ongulés pris dans lePléistocène supérieur. Paris, éditions du CNRS (Cahiers du Quaternaire 34).

Delpech, F., Villa, P., 1993. Activités de chasse et de boucherie dans la grotte desEglises. In: Desse, J., Audoin-Rouzeau, F. (Eds.), Exploitation des animauxsauvages à travers le temps. XIIIèmes Rencontres Internationales d’Archéologieet d’Histoire d’Antibes. APDCA, Sophia Antipolis, pp. 79e102.

Durand, E., Franc, O., Pasty, J.-F., Alix, P., Moncel, M.-H., 2009. Le site perchémoustérien de la montagne d’Andance (Saint-Bauzile, Ardèche), premièreapproche archéologique. Ardèche Archéologie 26, 15e21.

Evin, J., Maréchal, J., Marien, G., 1985. Lyon natural radiocarbon measurements X.Radiocarbon (New Haven) 27, 386e454.

Farizy, C., David, F., Jaubert, J., 1994. Hommes et bisons du Paléolithique moyen àMauran (Haute-Garonne). Gallia Préhistoire. CNRS, Paris. XXXème supplément.

Fernandes, P., Raynal, J.-P., Moncel, M.-H., 2008. Middle Palaeolithic raw materialgathering territories and neandertal mobility in the southern massif central ofFrance: first results from a petro-archaeological study on flint. Journal ofArchaeological Science 35, 2357e2370.

Fernandes, P., Moncel, M.-H., Lhomme, G., 2010. Analyse des comportements faceaux ressources minérales de deux sites du Paléolithique moyen: Payre, Abri desPêcheurs (Ardèche, France). In: Congrès international de Florence, Approachesto the Study of Lithic Technology, Rivista di Scienze Preistoriche, LVII, pp. 31e42.

Fillod, A., 1985. Le climat de la Haute-Loire. Éditions de la Borne, Saint-Vidal,collection Patrimoine local 2.

Fiore, I., Gala, M., Tagliacozzo, A., 2004. Ecology and subsistence strategies in theEastern Italian Alps during the Middle Palaeolithic. International Journal ofOsteoarchaeology 14, 273e286.

Fiore, I., Raynal, J-P., Tagliacozzo, A., 2005. Archeozoologia e aspetti tafonomici delsito musteriano di Baume-Vallée (Massicio centrale francese). Atti 3� ConvegnoNazionale di Archeozoologia (Siracusa, 2000) a cura di I. Fiore, G. Malerba andS. Chilardi, Strudi di Paletnologia II, Collana del Bullettino di Paletnologia Ital-iana, Istituto Poligrafico e Zecca dello Stato, Roma, pp. 81e92.

Forsten, A., Moigne, A.-M., 1998. The horse from the Middle Pleistocene of Orgnac 3(Ardèche, France). Quaternaire 9, 315e323.

Gagnière, S., Gauthier, L., Hugues, C., 1958. Le Moustérien de la Baume Flandin àOrgnac-l’Aven (Ardèche). In: Congrès Préhistorique de France, XVème session,Poitiers-Angoulême, pp. 491e509.

Gala, M., Raynal, J.-P., Tagliacozzo, A., 2005. Bird remains from the Mousterian levelsof Baume-Vallée (Haute-Loire, France): preliminary results. Feathers, grit andsymbolism. Birds and humans in the ancient Old and New Worlds. Proceedingsof the 5th Meeting of the ICAZ Bird Working Group in Munich[26.7.e28.7.2004]. Documenta Archaeobiologiae 3, pp. 141e145.

Gardeisen, V., 1999. Middle Palaeolithic Subsistence in the West Cave of ‘‘Le Portel’’(Pyrénées, France). Journal of Archaeological Science 26, 1145e1158.

Geneste, J.-M., 1985. Analyse lithique d’industrie moustériennes du Périgord: uneapproche technologique du comportement des groupes humains au Paléo-lithique moyen. Thèse. Université de Bordeaux I, France.

Geneste, J.-M., 1988. Systèmes d’approvisionnement en matières premières auPaléolithique moyen et au Paléolithique supérieur en Aquitaine. In: Otte, M.(Ed.), L’Homme de Néandertal, La Mutation, pp. 61e70.

Geneste, J.-M., Jaubert, J., Lenoir, M., Meignen, L., Turq, A., 1997. Approche tech-nologique des Moustériens charentiens du sud-ouest de la France et du Lan-guedoc oriental. Paleo 9, 101e142.

Grant, A., 1982. The use of tooth wear as a guide to the age of domestic ungulates.In: Wilson, B., Grigson, C., Payne, S. (Eds.), Ageing and Sexing Animal Bonesfrom Archaeological Sites. British Archaeological Reports International Series,S109, pp. 91e108.

Grayson, D., 1989. Bone transport, bone destruction, and reverse utility curves.Journal of Archaeological Science 16, 643e652.

Griggo, C., 1995. Significations paléoenvironnementales des communautés animalespléistocènes reconnues dans l’abri Suard (Charente) et la grotte de Bois-Ragot(Vienne): essai de quantification de variables climatiques. Thèse. UniversitéBordeaux I, 1342.

Grove, M., 2009. Hunteregatherer movement patterns: causes and constraints.Journal of Anthropological Archaeology 28, 222e233.

Guadelli, J.-L., 2008. La gélifraction des restes fauniques. Expérimentation ettransfert au fossile. Annales de Paléontologie 94 (3), 121e165.

Jaubert, J., 2001. Un site moustérien de type Quina dans la vallée du Célé (Pailhès àEspagnac-Sainte-Eulalie, Lot). Gallia Préhistoire 43, 1e99.

Jaubert, J., Bismuth, T., 1996. Le Paléolithique moyen des Pyrénées centrales:esquisse d’un schéma chronologique et économique dans la perspective d’uneétude comparative avec les documents ibériques. In: Delporte, H., Clottes, J.(Eds.), 118� Congrès National des Sociétés historiques et scientifiques (Pau1993), CTHS, pp. 9e26.

Jaubert, J., Mourre, V., 1996. Coudoulous, le Rescoundoulous, Mauran: diversité desmatières premières et variabilité des schémas de production d’éclats. Qua-ternaria Nova VI, 313e341.

Jaubert, J., Farizy, C., 1995. Levallois débitage: exclusivity, absence or coexistence withother operative schemas (Garonne Basin, France). In: Dibble, H.L., Bar-Yosef, O.(Eds.), The Definition and Interpretation of Levallois Technology, Monograph inWorld Archaeology. Prehistory Press, 23, Madison Wisconsin, pp. 227e247.

Kelly, R.L., 1995. The Foraging Spectrum: Diversity in HuntereGatherer Lifeways.Smithsonian Institution Press, Washington D.C.

Klein, R.G., Cruz-Uribe, C., 1984. The Analysis of Animal Bones from ArchaeologicalSites. University of Chicago Press.

Kuhn, S., 1992. On planning and curated technologies in the Middle Paleolithic.Journal of Anthropological Research 48, 185e214.

Koslowsky, J., 1994. Le rythme climatique du Pléistocène supérieur et la présencehumaine dans les montagnes. Preistoria Alpina 28, 37e47.

Lafont, R., Périnet, G., Bazile, F., Icole, M., 1984. Racémisation d’acides aminésd’ossements fossiles du Paléolithique supérieur Languedocien. Comptes Rendusde l’Académie des Sciences de Paris 299, 447e450.

de Lumley, H., 1969. Le Paléolithique inférieur et moyen du Midi Méditerranéendans son cadre géologique, t. 1: Ligurie-Provence. Gallia Préhistoire CNRS, Vèmesupplément.

de Lumley, H.,1972a. La grotte de l’Hortus (Valflaunès, Hérault), situation, descriptionet historique des fouilles. In: de Lumley, H. (Ed.), La grotte moustérienne del’Hortus. Etudes Quaternaires e Géologie, Paléontologie, Préhistoire, pp. 9e16.

de Lumley, H., 1972b. La stratigraphie des dépôts quaternaires de la grotte del’Hortus. In: de Lumley, H. (Ed.), La grotte moustérienne de l’Hortus. EtudesQuaternaires e Géologie, Paléontologie, Préhistoire, pp. 57e100.

Lyman,R.L.,1994.VertebrateTaphonomy.CambridgeUniversityPress, Cambridge (MA).Masaoudi, H., FalguèresBahain, C., Yokohama, J.-J., Lhomme, G., 1994. Datation

d’ossements et de planchers stalagmitiques provenant de l’abri des Pêcheurs(Ardèche) par la méthode de déséquilibre des familles de l’Uranium (U/Th).Quaternaire 5, 79e83.

Masaoudi, H., Falguères, C., Bahain, J.-J., Moncel, M.-H., 1997. Datation du sitepaléolithique moyen de Payre (Ardèche): nouvelles données radiométriques(méthodes U/Th et ESR). Comptes rendus de l’Académie des Sciences, Paris 324(IIa), 149e156.

Meignen, L., Costamagno, S., Beauval, C. Bourguignon, L., Vandermeersch, B.,Maureille, B., 2006. Gestion des ressources lithiques au Paléolithique moyendans une halte de chasse spécialisée sur le Renne: Les Pradelles (Marillac-Le-Franc, Charente). In: Moncel, M.-H., Moigne, A.-M., Arzarello, M., Peretto, C.(Eds.), Proceedings of the XV World Congress UISPP (Lisbon, 4e9 September2006). Raw Material Supply Areas and Food Supply Areas. Integrated approachof the behaviours, vol. 5, Session WS23. British Archaeological Reports Inter-national Series S1725, pp. 127e139.

Moncel, M.-H., 2001. Middle Palaeolithic assemblages with blades in south-eastFrance. The question of the technical variability during Middle Palaeolithic andits meaning. Archaeology, Ethnology and Anthropology of Eurasia 2, 37e47.

Moncel, M.-H., 2003. L’exploitation de l’espace et la mobilité des groupes humainsau travers des assemblages lithiques à la fin du Pléistocène moyen et au débutdu Pléistocène supérieur. La moyenne vallée du Rhône entre Drôme et Ardèche.British Archaeological Reports International Series, S1184.

Moncel, M.-H., 2006. (dir.) La faune des campagnes de fouille 2005–2006. Projetd’étude de sites de la fin du Pléistocène moyen et du début du Pléistocènesupérieur en moyenne vallée du Rhône – Bilan des campagnes 2005e2006.Rapport S.R.A., D.R.A.C. Rhône-Alpes (inédit).

Moncel, M.-H., Combier, J., 1992. L’industrie lithique du site pléistocène moyend’Orgnac 3 (Ardèche). Gallia Préhistoire 34, 1e55.

Moncel, M., Perrève, V., 1999. Un atelier de taille moustérien: le Clos du Charnier àLagorce (Ardèche, France): quelques remarques sur les modes de débitage.L’Anthropologie 103, 471e484.

Moncel, M.-H., Michel, V., 2000. Première datation radiométrique par U–Th d’unniveau moustérien de l’abri du Maras (Ardèche, France). Bulletin de la SociétéPréhistorique Française 97, 371e375.

Moncel, M.-H., Moigne, A.-M., Combier, J., 2005. Pre-Neandertal behaviourduring isotopic stage 9 and the beginning of stage 8. New data concerningfauna and lithics in the different occupation levels of Orgnac 3 (Ardèche,South-East France): occupation types. Journal of Archaeological Science 32,1283e1301.

Moncel, M.-H., Crégut-Bonnoure, E., Daujeard, C., Lartigot, A.-S., Lebon, M.,Puaud, S., Boulbes, N., Croizet, N., 2008a. La Baume Flandin (commune d’Orgnac

C. Daujeard et al. / Quaternary International 252 (2012) 32e4746

Author's personal copy

l’Aven, Ardèche): nouvelles données sur ce gisement du Paléolithique moyen.Palevol 7, 315e325.

Moncel, M.-H., Brugal, J.-P., Prucca, A., Lhomme, G., 2008b. Mixed occupation duringthe Middle Palaeolithic: Case study of a small pit-cave-site of Les Pêcheurs(Ardèche, southeastern France). Journal of Anthropological Archaeology 27,382e398.

Moncel, M-H., dir., 2008c. Payre. Des occupations humaines de la moyenne valléedu Rhône en contexte de grotte et d’abri de la fin du Pléistocène moyen et dudébut du Pléistocène supérieur. Bilan des fouilles 1990e2002, Mémoire de laSociété Préhistorique Française, 46.

Moncel, M.-H., Daujeard, C., Crégut-Bonnoure, E., Boulbes, N., Puaud, S., Debard, E.,Bailon, S., Desclaux, E., Escudé, E., Roger, T., Dubar, M., 2010. Nouvelles donnéessur les occupations humaines du début du Pléistocène supérieur de la moyennevallée du Rhône (France). Les sites de l’Abri des Pêcheurs, de la Baume Flandin,de l’Abri du Maras et de la Grotte du Figuier (Ardèche). Quaternaire 21 (4),389e415.

Mortensen, P., 1972. Seasonnal camps and early villages in the Zagros. In: Ucko, P.,Tringham, R., Dimbleby, G.W. (Eds.), Man, Settlement and Urbanism. Duck-worth, London, pp. 293e297.

Mourre, V., 1996. Les industries en quartz au Paléolithique. Terminologie, méth-odologie et technologie. Paleo 8, 205e223.

Mourre, V., 1997. Industries en quartz: précision terminologique dans les domainesde la pétrographie et de la technologie. Préhistoire et Anthropologie Médi-terranéennes 6, 201e210.

Mourre, V., Costamagno, S., Bruxelles, L., Colonge, D., Cravinho, S., Laroulandie, V.,Maureille, B., Thiébaut, C., Viguier, J., 2008. Exploitation du milieu montagnarddans le moustérien final: La Grotte du Noisetier à Fréchet-Aure (PyérénéesCentrales Françaises). In: Proceedings of the XV World Congress UISPP (Lisbon,4e9 September 2006). Mountain environments in prehistoric europe: settle-ment and mobility strategies. British Archaeological Reports InternationalSeries.

Peresani, M., 1998. La variabilité du débitage Discoïde dans la Grotte de Fumane(Italie du nord). Paléo 10, 123e146.

Peresani, M., Sartorelli, A., 1998. The lithic assemblages at the Cave of Fumane.Newevidence of technological variability in the Middle palaeolithic of northern Italy.In: Proceedings of the XIII Congress of the International Union of Prehistoricand Protohistoric Sciences, Forli, Italy, pp. 269e278

Porraz, G., 2005. En marge du milieu alpin-dynamiques de formation des ensembleslithiques et modes d’occupation des territoires au Paléolithique moyen. Thèse.Université de Provence, France.

Raynal, J-P., 1988. Un site moustérien de moyenne montagne en haute vallée de laLoire: l’abri Laborde (Baume-Vallée) à Solignac-sur-Loire (Haute-Loire). In:Livret-guide de l’excursion de la séance décentralisée de la Société Pré-historique Française, Les peuplements paléolithiques du Massif central (Le Puy-en-Velay, 8e9 octobre 1988).

Raynal, J-P., Le Corre-le-Beux, M., Santagata, C., Fernandes, P., Guadelli, J-L., Fiore, I.,Tagliacozzo, A., Lemorini, C., Rhodes, E., Bertran, P., Kieffer, G., Vivent, D., 2005.Paléolithique moyen dans le Sud du Massif central: les données du Velay(Haute-Loire, France). In: Molines, N., Moncel, M.-H., Monnier, J.-L. (Eds.), Col-loque international: Données récentes sur les modalités de peuplement et sur lecadre chronostratigraphique, géologique et paléogéographique des industriesdu Paléolithique ancien et moyen en Europe (Rennes, 22e25 septembre 2003).British Archaeological Reports International Series, S1364, pp. 173e201.

Raynal, J.-P., dir., 2007. Saint-Anne 1 (Sinzelles, Polignac, Haute-Loire) e Le Paléo-lithique moyen de l’unité J1, Les dossiers de l’Archéo-Logis, CDERAD, 3.

Raynal, J.-P., Moncel, M.-H., Fernandes, P., Santagata, C., Guadelli, J.-L., Fernandez, P.,Patou-Mathis, M., 2008. Espace minéral et espace de subsistance au Paléo-lithique moyen dans le sud du Massif central en France: les sites de Sainte-AnneI (Haute-Loire) et de Payre (Ardèche). In: Moncel, M.-H., Moigne, A.M., Arzar-ello, M., Peretto, C. (Eds.), Proceedings of the XV World Congress UISPP (Lisbon,4e9 September 2006). Raw Material Supply Areas and Food Supply Areas.

Integrated approach of the behaviours, vol. 5, Session WS23. British Archaeo-logical Reports International Series, S1725, pp. 141e161.

Rendu, W., 2010. Rendu, Hunting behaviour and Neanderthal adaptability in the LatePleistocene site of Pech-de-l’Azé I. Journal of Archaeological Science 37, 1e13.

Rivals, F., Moncel, M.-H., Patou-Mathis, M., 2009. Seasonality and intra-site variationof Neanderthal occupations in the Middle Palaeolithic locality of Payre (Ardè-che, France) using dental wear analyses. Journal of Archaeological Science 36,1070e1078.

Rosell, J., Blasco, R., Rivals, F., Chacon, M.G., Menendez, L., Morales, J.I., Rodriguez-Hidalgo, A., Cebria, A., Carbonell, E., Serrat, D., 2010. A stop along the way: therole of Neanderthal groups at level III of Teixoneres cave (Moia, Barcelona,Spain). Quaternaire 21, 139e154.

Szmidt, C., 2003. The Mousterian in Mediterranean France e A regional, integrativeand comparative perspective, British Archaeological Reports InternationalSeries, S1147.

Szmidt, C., 2009. From regional patterns to behavioural interpretation: assessing theMiddle to Upper Palaeolithic transition in Mediterranean France. In: Camps, M.,Szmidt, C. (Eds.), The Mediterranean from 50,000 to 25,000 B.P.: Turning Pointsand New Directions. Oxbow Books, Oxford, pp. 231e254.

Szmidt, C., Moncel, M.-H., Daujeard, C., 2010. New data on the late Mousterian inMediterranean France: First radiocarbon (AMS) dates at Saint-Marcel Cave(Ardèche). Palevol 109, 185e199.

Théry-Parisot, I., 2001. Economie des combustibles au Paléolithique e expérimen-tation, taphonomie, anthracologie, Dossier de documentation archéologiqueCNRS 20.

Théry-Parisot, I., Costamagno, S., Brugal, J-P., Fosse, P., Guilbert, P., 2005. The use ofbone as fuel during the Palaeolithic, experimental study of bone combustibleproperties. In: Mulville, J., Outram, A.K. (Eds.), The Zooarchaeology of Fats, Oils,Milk and Dairying. Proceedings of the 9th Conference of the InternationalCouncil of Archaeozoology (Durham, 2002), pp. 50e59.

Tillet, T., 2002. Les grottes à ours et occupations néandertaliennes dans l’arc alpin etjurassien. In: Tillet, T., Binford, L.W. (Eds.), L’Ours et l’Homme, ERAUL 100, Liège,pp. 167e183.

Turq, A., 2001. Réflexions sur le biface dans quelques sites du Paléolithique ancien-moyen en grotte ou abri du Nord-Est du Bassin Aquitain. In: Cliquet, D. (Ed.), Lesindustries à outils bifaciaux du Paléolithique moyen d’Europe occidentale,ERAUL, 98, Liège, pp. 141e149.

Utrilla, P., Montes, P., 1989. La grotte moustérienne de Gabasa (Huesca, Espagne). In:Freeman, L., Patou, M. (Eds.), L’Homme de Néandertal. La Subsistance, vol. 6.ERAUL, Liège, pp. 145e153.

Utrilla, P., Montes, P., Gonzales-Samperiz, P., 2006. Est-ce que c’était l’Ebre unefrontière à 40e30 ka? In: Cabrera Valdes, V., Bernaldo de Quirós, F., MaílloFernández, J.M. (Eds.), En el centenario de la cueva de el Castillo: el ocaso de losneandertales. Centro asociado a la Universidad nacional de educación a dis-tancia en Cantabria, pp. 165e191.

Valensi, P., Psathi, E., 2004. Faunal exploitation during the Middle Palaeolithic insouth-eastern France and north-western Italy. International Journal of Osteo-archaeology 14, 256e272.

Valladas, H., Mercier, N., Ayliffe, L., Falguères, C., Bahain, J.-J., Dolo, J.-M., Froget, L.,Joron, J.-L., Masaoudi, H., Reyss, J.-L., Moncel, M.-H., 2008. Chronology of theMiddle Paleolithic sequence of Payre (Ardèche, France) based on radiometricdating methods. Quaternary Geochronology. Quaternary Science Review 3,377e389.

Vaquero, M., Vallverdu, J., Rosell, J., Pasto, I., Allué, E., 2001. Neandertal behavior atthe middle Palaeolithic site of Abric Romani, Capellades, Spain. Journal of FieldArchaeology 28, 93e113.

Villa, P., Mahieu, E., 1991. Breakage patterns of human long bones. Journal of HumanEvolution 21, 27e48.

Wilson, B., 1989. Fresh and old table refuse. The recognition and location ofdomestic activity at archaeological sites in the Upper Thames Valley, England.Archaeozoologia 3, 237e260.

C. Daujeard et al. / Quaternary International 252 (2012) 32e47 47