- �

______________________________________

** Université Chrétienne «Dimitrie Cantemir», Faculté d’Histoire, Bucarest, RO** Université «Lucian Blaga», Faculté d’Histoire et pour l’Étude du Patrimoine, Sibiu, RO

Atti Soc. Preist. Protost. Friuli-V.G., Trieste, XVI, 2006-200� (2008): �-2�

SUMMARY – Stones and mammoths. The Upper Palaeolithic ivory tools of Romania: recent data. Recent ap-proaches to ancient artefacts collections enable a detailed discussion on the rare evidence of use of ivory in the Romanian Upper Palaeolithic. The four objects presented in this paper come from Moldavia (RO). The first is a fragment of a mammoth tusk discovered in the open-air Gravettian site of Lespezi (layer IV), in the Bistrita Valley (Bacau county), in 1963, radiocarbon-dated to 18.110±300 uncal BP (Bln-806). The main characteristic of the tool is the presence of debitage traces along its proximal edge. They demonstrate the utilisation of notching and groov-ing techniques and probably transverse sawing by fiber. This is supposed to represent the oldest indicator of such a technique, which is common to the Mesolithic and Early Neolithic periods in this part of Europe. The second tool is a fragmented polisher (?) recovered from layer V of the open-air Gravettian site of Cotu Miculinti, in the Prut Val-ley (Botosani county), in 1981, radiocarbon-dated to 19.460±220 uncal BP (GrN-12662). M. Brudiu published it as obtained from a roe-deer antler. Following my analysis, it was clear that it was made from ivory. Three other tools, only two of which so far published, are the first ivory spear points ever known from Romania. They were recovered very recently from the Epigravettian layer II of the site of Piatra-neamţ - Poiana Ciresului (neamţ county), radio-carbon-dated between 19 and 16 ka BP. Given the extreme rarity of ivory tools from the Upper Palaeolithic sites of Romania it is probable that these specimens come from neighbouring territories of Central and Eastern Europe (Czech Republic, Ukraine, Moldavia or Russia), where the manufacture and use of such tools was common during this period.

RiASSUnto – Pietre e mammuth. Nuovi dati sugli strumenti in avorio del Paleolitico Superiore della Romania. L’analisi recente di alcune vecchie collezioni di reperti, ha favorito una discussione particolareggiata dell’impiego assai raro dell’avorio durante il Paleolitico Superiore in Romania. I quattro oggetti presentati in questo lavoro provengono dalla Moldavia (RO). Il primo è un frammento di zanna di mammuth rinvenuto, nel 1963, nello strato IV del sito Epigravettiano all’aperto di Lespezi, nella Valle del Bistrita (dipartimento di Bacau), datato radiomen-tricamente a 18110±300 uncal BP (Bln-806). Le caratteristiche più importanti riguardano le tracce di lavorazione per percussione diretta presenti all’estremità prossimale, mentre le usure presenti presso il peduncolo potrebbero derivare dallo scorrimento di fibre vegetali. Il secondo strumento è probabilmente un lisciatoio raccolto, nel 1981, nello strato V del sito Gravettiano all’aperto di Cotu Miculinti, nela Valle del Prut (distretto di Botosani), radiodatato

CORNELIU BELDIMAN∗ et DIANA-MARIA SZTANCS∗∗

PiERRES Et MAMMoUtHS.LES iVoiRES oUVRÉS AU PALÉoLitHiQUE SUPÉRiEUR

En RoUMAniE: DonnÉES RÉCEntES

In memoriam doctissimae Alexandra Bolomey

8 -

a 19460±220 uncal BP (GrN-12662). M. Brudiu ha pubblicato quest’oggetto come confezionato in palco di capriolo; un’analisi dettagliata ha dimostrato la sua manifattura in avorio. Altri tre strumenti, due soli dei quali sono stati sinora pubblicati, costituiscono i primi esemplari di punte di zagaglia in avorio rinvenuti in Romania. Questi sono stati raccolti molto recentemente nello strato II, Epigravettiano, della stazione di Piatra-neamţ, Poiana Ciresului, nel distretto di neamţ, datato radiometricamente ad un periodo compreso fra 19000 e 16000 uncal BP. Data la notevole rarità di reperti in avorio nei siti del Paleolitico Superiore della Romania, è probabile che essi provengano da territori vicini (Repubblica Ceca, Ucraina, Moldavia o Russia), dove la confezione e l’impiego di strumenti di questo tipo è comune durante il periodo in esame.

Keywords: Romania, Eastern Gravettian/Epigravettian, bone and antler industry, ivory, technology.

intRoDUCtion

Les plus anciens témoignages de l’utilisation de l’ivoire de mammouth datent du Paléo-lithique moyen. Ces éléments squelettiques à destination de matériaux de construction étai-ent insérés dans des structures d’habitat ou destinés isolément à la confection des outils et d’armes. Au Paléolithique supérieur, les défenses de mammouth ont constitué une des mat-ières premières d’origine animale de prédilection dans les cultures répandues entre la façade atlantique et l’Oural. De ce point de vue, son rôle important dans l’industrie complexe des matières dures animales pour la fabrication des outils, armes, objets d’art et de parure est déjà très bien connue (MACGREGOR, 1985; CLOTTES, 1990; CHIRICA et BORZIAC, 1995; BELDIMAN, 1996; CHRISTENSEN, 1999).

LES AntECEDEntS

En ce qui concerne les territoires actuels de la Roumanie, les plus anciens indices sig-nalés dans la littérature de l’utilisation de l’ivoire de mammouth datent du Paléolithique Moyen. À l’occasion des fouilles menées dans les années ‘60-‘80 par le regretté Dr. Alexan-dru Păunescu dans le site de terrasse de Ripiceni - lieu-dit ‘Izvor’, dép. de Botosani, placé sur la rive droite de la rivière Prut, on a mis au jour dans les niveaux Moustériens I, IV et V plusieurs structures d’habitat de type paravents, aménagés à l’aide de pierres calcaires, de bois et de défenses de mammouth entières ou de tronçons prélevés par percussion/entaillage (Păunescu, 1989; 1993: 31-35, 71-75, 85-91, 194; 1999: 168-299). Une autre découverte, cette fois fortuite, datant de Paléolithique moyen aussi, a été faite dans le site de Holboca, dép. de Iaşi, d’ou provient des restes squelettiques postcraniens, deux défenses entières et quelques outils en silex (ZAHARIA et al., 19�0: 198d; CHIRICA et TANASACHI, 1984-1985, vol. I: 1�1; vol. II: 488; BELDIMAN, 1996: 326; Păunescu, 1999: 30�-309).

LES ARtEFACtS DU PALEoLitHiQUE SUPERiEUR

En dépit des recherches archéologiques qui se sont déroulées systématiquement depuis un demi-siècle dans plusieurs grands sites de plein air du Paléolithique Supérieur en Rou-manie, surtout dans sa partie orientale (Ripiceni - ‘Izvor’, Mitoc - ‘Malul Galben’ et ‘Valea

- 9

Izvorului’, Crasnaleuca - ‘Staniste’, Cotu Miculinti - ‘Garla Mare’ etc.), les découvertes qui attestent l’utilisation des artefacts en ivoire de mammouth dans cette partie de l’Europe se sont avérées très rares.

Il faut d’abord préciser que dans deux sites gravettiens de plein air de Moldavie rou-maine ont été récupérés des objets de ce genre (tronçons de défense et une défense entière de mammouth) mais sans aucune trace décelable d’intervention technique. Dans le niveau Gravettien I b du site de Ripiceni - ‘Izvor’, dép. de Botosani, est signalé un tronçon de défense de mammouth en état avancé de fossilisation; d’après Alexandru Păunescu, il s’agit probablement d’une pièce retrouvée par les préhistoriques dans une couche moustérienne ou un gisement fossilifère (Păunescu, 1993: 158-159, 185; BeldiMan, 1996: 326).

Dans le niveau Gravettien III - l’unité d’habitation 51 du site de Mitoc - ‘Malul Galben’, dép. de Botosani, Vasile Chirica mentionne la découverte d’une défense de mammouth en-tière, probablement retrouvée aussi par les préhistoriques dans une couche Moustérienne ou un gisement fossilifère (chirica, 1989: 51; chirica et BorZiac, 1995; Păunescu, 1999: 149, 155).

Actuellement on connaît seulement cinq objets Paléolithiques en ivoire, datés de Gravet-tien oriental et de l’Epigravettien (fig. 1). Appart les pièces de Piatra neamţ-Poiana Ciresului, qui ont été découvert il y a deux-trois années, ils sont jusqu’à récemment restés presque inaperçus à la recherche préhistorique de Roumanie; par exemple, ils n’ont été discutés ni dans l’essai de synthèse rédigé par Vasile Chirica et Ilie A. Borziac et présenté au colloque de Ravello de 1992 (CHIRICA et BORZIAC, 1995), ni dans la synthèse récente des mêmes au-teurs (BORZIAC et al., 2006), ni dans les ouvrages de Codrin-Valentin CHIRICA (1996a; 1996b; 2004).

Dans ce contexte, l’ouvrage propose une analyse actualisée des données disponibles sur ces encore rarissimes artefacts Paléolithiques en ivoire de mammouth. Pour autres approches du sujet voir: BELDIMAN (1996; 1999a; 1999b; 2000; 2003; 2004a; 2004b; 2005a; 2005b; 2006; 200�).

Le fragment de défense de LespeziLe premier objet en ivoire connu est un fragment de défense découvert à l’occasion

des fouilles effectuées en 1963 par le Dr. Maria Bitiri-Ciortescu (Institut d’Archéologie de Bucarest) et le Dr. Viorel Capitanu (Musée Départemental de Bacau) dans le site Gravettien de Lespezi - ‘Lutarie’, comm. de Garleni, dép. de Bacau, en Moldavie Roumaine (fig. 1, la carte et n° 1). Les résultats de ces fouilles ont été publiés en 19�2 et 1991 (BITIRI et caPitanu, 19�2; Bitiri-ciortescu et al., 1991; Păunescu, 1998: 299-315). Il s’agit d’un campement saisonnier de chasseurs d’herbivores (bovidés, renne, cheval) situé sur la vallée de la rivière Bistrita. L’objet est conservé actuellement dans les collections du Musée National d’Histoire de la Roumanie, Bucarest, n° inv. 1564�5.

L’artefact est issu du niveau IV. Il a été signalé par la regrettée paléontologue Alexandra BOLOMEy (1991). Il a ensuite été l’objet de plusieurs articles publiés dans des revues rou-maines, étant présenté et discuté d’une manière exhaustive (BELDIMAN, 1996; 1999a; 1999b; 2000; 2003; 2005; 200�).

Dans le même site, Alexandra Bolomey avait signalé aussi quelques éléments isolés du squelette postcranien de mammouth (Elephas primigenius): quatre de niveaux V-IV et une de

10 -

Fig. 1 - Objets Paléolithiques en ivoire de Roumanie: répartition des découvertes. 1: fragment de défense de mam-mouth de Lespezi (d’après BEldImAN, 1996: 327, fig. 1); 2: lissoir de Cotu Miculinti; 3: pointes de sagaies de Piatra-neamţ - Poiana Ciresului (d’après CARCIUmARU et al., 2006c: 35, fig. 11/a).

- 11

niveau VI (BOLOMEy, 1991: 2�3, 282). En effet, ces indices sont insuffisants pour envisager la procuration de l’ivoire par la chasse locale.

Le niveau IV ne dispose pas de datations absolues; étant donné que le niveau sous-ja-cent (V) a livré la date 14C de 18.110±300 uncal BP (Bln-806) et celui superposé (III) une date 14C de 18.020±350 uncal BP (Bln-808), on peut supposer que la datation du niveau de provenance du fragment de défense se place autour de 18.000 uncal BP (Bitiri-ciortescu et al., 1991: 282-295; BELDIMAN, 1996: 328; Păunescu, 1998: 310-313). Le niveau IV est at-tribué à la deuxième phase du Gravettien de la vallée de Bistrita (Păunescu, 1998: 313). Les dimensions de l’objet sont les suivantes: longueur: 16,5 cm; largeur/épaisseur maximum: 6,3/4,5 cm.

Le fragment a la forme générale d’un parallélépipède irrégulier et il a subi l’action de délitage/délamination et de corrosion partielle des surfaces. On ne garde plus aucune portion des surfaces anatomiques. À son extrémité proximale sont conservés et clairement décelables les stigmates de débitage par percussion directe (entaillage) (fig. 2).

En même temps, autour du pédicule, on peut observer, mais de manière discontinue à cause de l’altération des surfaces par corrosion, la présence de portions de rainures courbes, larges de 2-3 mm, peu profondes et faiblement marquées, décelables à l’œil nu et à la loupe binoculaire (fig. 3). Nous avons interprété hypothétiquement ces stigmates comme les té-moins susceptibles de mettre en lumière un débitage par usure qui succède à un débitage par entaillage. Selon les opinions bien connues de François Poplin et d’autres spécialistes français, il s’agit de sciage transversal à la ficelle ou abrasion linéaire (POPLIN, 19�4; BIL-LAMBOZ, 19��: 101-102) (fig. 9, n° 4). La partie distale semble amincie aussi par entaillage. L’objet n’a pas subi des autres actions de façonnage.

En ce qui concerne la fonctionnalité de l’artefact, on a peut-être affaire à un fragment de matière première d’origine animale en cours de transformation, en vue d’obtenir un artefact semblable à la partie active des haches. À cet égard, il faut rappeler les multiples analogies ret-rouvées dans les sites de chasseurs gravettiens de grands herbivores du centre-est et de l’Est de l’Europe: République Tchèque (Predmost: Feustel, 19�3: 166-16�, pl. LVIII, n° 1-2; pl. LXIX, n° 3; pl. LXXI; luMley, 1984: 118); Ukraine (Kostienki I, Poljakov et Eliseevich; SEMENOV, 1985: 148-150, fig. �3, n° 1-2 et fig. �4, n°s 1-6; BOSINSKI, 1990: 118); Russie (Borshevo, Timonovka: koZłowski, 1992: �8, 1�8, fig. 8� c-d et fig. 12� f-g); République de Moldavie (Costesti: BORZIAC, 1994: 28, 35 et fig. 5, n° 10) (fig. 8-9). Une autre analogie, plus importante, provient d’Autriche; le site Epigravettien de Grubgraben, localisé dans le bassin moyen du Danube et daté après 20.000 BP, a livré trois fragments d’ivoire de mammouth, dont un est en cours de transformation (débité et façonné sommairement par entaillage) et morphologique-ment très proche de celui de Lespezi (LOGAN, 1990: �2-�3, 84 et fig. 6, n° 1) (fig. 8, n° 5).

Il est important de signaler, outre la présence rare d’un objet travaillé en ivoire au Paléo-lithique supérieur de Roumanie, l’attestation précoce probable d’un procédé de débitage so-phistiqué, le sciage transversal complet à la ficelle ou l’abrasion linéaire. Ce procédé a été largement appliqué au travail des matières osseuses dès le Mésolithique et le Néolithique ancien, ainsi que l’attestent aussi les matériaux connus en Roumanie (BELDIMAN, 1996: 332; 1999b; 200�). Cet aspect peut conférer à la découverte de Lespezi une importance particulière dans la perspective de l’évolution paléotechnologique dans le domaine de l’industrie des matières dures animales.

12 -

Fig. 2 - Fragment de défense de mammouth de Lespezi. 1-3: vues générales; 4-5: vues du bord portant des traces d’entaillage.

- 13

Fig. 3 - Fragment de défense de mammouth de Lespezi. 1-5: vuès de la partie et de l’extrémité proximale portant des traces d’entaillage et de sciage par abrasion linéaire.

14 -

Étant donné l’absence dans le site de provenance d’autres témoignages concernant le travail de l’ivoire, ainsi que la rareté des restes squelettiques de mammouth, on peut envis-ager une provenance extérieure, allogène, de l’objet de Lespezi, emporté probablement par les communautés de chasseurs de rennes de l’aire orientale ou centrale de l’Europe.

Le lissoir de Cotu MiculintiLe deuxième artefact Paléolithique en ivoire de Roumanie est un fragment mésio-distal

de lissoir (ou poignard ?). L’objet a été récupéré pendant les fouilles de 1981 conduites par le Dr Mihail Brudiu (Musée Départemental de Galati) dans niveau V de l’important site gravettien de chasseurs de rennes de Cotu Miculinti - ‘Garla Mare’, comm. Cotusca, dép. de Botosani, Moldavie roumaine, placé sur la rive droite de la rivière Prut (fig. 1, la carte et n° 2). Du même site provienent les plus nombreux et importants matériaux paléolithiques de Roumanie concernant le travail de l’os et du bois de renne (Brudiu, 19�9; 1980a; 1980b; 1983; 198�; 1994; 2006).

Dans plusieurs publications, l’auteur des recherches a précisé le fait que l’objet qui retient ici notre attention, toujours seulement inséré dans l’illustration mais jamais étudie en détail, est une ‘pointe de lance’ fabriquée en bois de renne (Brudiu, 1986; 198�; 1994) (fig. 4, n° 1).

En mai 2001, l’auteur principal a eu l’occasion d’examiner une partie de la collection publiée d’artefacts en matières osseuses de Cotu Miculinti, conservée au Musée Départemen-tal de Botosani. Ainsi on a puisse faire des observations qui enrichissent sensiblement les données concernant la typologie et la technologie des matières dures animales (os et spéciale-ment bois de renne) au Paléolithique supérieur de Moldavie.

À cette occasion, on a procédé aussi à l’examen de l’artefact découvert pendant les fouilles de 1981 et catalogué couramment comme ‘pointe de lance’ fabriquée en bois de renne. Par conséquent, on a puisse constater qu’il s’agissait, en fait, d’un objet en ivoire de mammouth. À cet égard, notre ouvrage présente toutes les données issues de l’examen direct.

Le niveau V du site, exploré en 1981 dans la Surface B, révèle une occupation relative-ment faible (Brudiu, 1986: �). L’objet a été récupéré dans le carré 5/3, à une profondeur de 2,60 m et publié comme pointe de sagaie en bois de renne (Brudiu, 198�: �5, 84 et fig. 12, n° 2). Dans le répertoire des artefacts en matières dures animales de Roumanie récemment élabore, il a l’indicatif CTM 9 (BELDIMAN, 1999b, vol. II: 39).

L’objet est conservé en état fragmentaire; il s’agit d’un fragment mésio-distal, avec une fracture ancienne au niveau de la partie mésiale et récente de la partie distale. Les surfaces sont corrodées et présentent des dépôts discontinus d’oxyde de manganèse (fig. 4, n°s 2-3 et fig. 5).

L’artefact a une silhouette étroite et allongée, de section biconvexe asymétrique au niveau de sa partie active (distale) et circulaire du reste. Il a été entièrement façonné par raclage axial. La face supérieure conserve une courbure proche de la courbure anatomique de la défense; cette observation permet l’appréciation de son diamètre initial autour de 5 cm. La fracturation accidentelle récente de la partie distale a mis clairement en évidence la texture spécifique stratifiée de l’ivoire - dentine (fig. 5, n°s 1-2 et 4). Les surfaces corrodées ont ef-facé les éventuelles traces d’utilisation (fig. 5, n° 3).

- 15

Les dimensions de l’objet sont les suivantes: longueur préservée: 8,9 cm; longueur ini-tiale présumée: 20 cm; longueur de la partie active initiale: 9 cm; diamètre maximum de la partie active: 1,4/1,3 cm; diamètre de la partie mésiale: 1,9 cm).

Les analogies de l’artefact en discussion avec d’autres objets fréquents dans le site de Cotu Miculinti ou dans d’autres sites gravettiens de Moldavie et/ou d’autres régions plus ou moins proches permettent de l’attribuer typologiquement et d’un point de vue fonctionnel dans la catégorie des lissoirs. Deux objets presque identiques, travaillés en os (fig. 6, n°s 1-2) et en bois de renne (fig. 6, n°s 3-4) ont été signalés dans le même site. Parmi les analogies avec des lissoirs en ivoire, on peut mentionner l’objet de Kostienki I Eliseevich (Feustel, 19�3: pl. LVIII, n° 2) (fig. 8, n° 3). Considérés généralement comme des outils réservés au prélèvement et pour la préparation des peaux, les lissoirs de ce genre sont aptes aussi pour l’utilisation comme armes (poignards). La section circulaire de la partie proximale rend dif-ficile leur fixation sur une hampe et, par conséquent, l’usage comme armature de sagaie.

Le niveau IV qui se superpose au niveau de provenance de l’objet est daté sur des bases radiométriques de 19.460±220 uncal BP (GrN-12662), ce qui permet d’estimer l’age du niveau V autour de 19.000 uncal BP (CHIRICA, 2001: 106).

L’importance des observations récentes sur le lissoir (re)découvert de Cotu Miculinti consiste avant tout dans l’identification définitive et indubitable du type et de la matière première, ce qui permet de parler du premier objet fini en ivoire connu jusqu’alors dans les cultures du Paléolithique supérieur en Roumanie. L’existence dans le même site d’un important lot d’artefacts en matières dures animales (os, bois de renne), finis, en cours de fabrication et déchets, atteste d’une manière indubitable la fabrication sur place de ces ar-tefacts, présents ici dans des proportions inconnues dans le reste du territoire à l’ouèst de la rivière Prut. En contraste avec cette situation, l’absence d’autres artefacts en ivoire sur place favorise l’hypothèse selon laquelle le lissoir semble être emporté des régions avoisinantes.

Les pointes de sagaies de Piatra Neamţ - Poiana CiresuluiTrois autres artefacts en ivoire proviennent des fouilles menées en 2004 et 2005 dans

le site placé sur la rive droite de la rivière Bistrita, dép. de neamţ (fig. 1, la carte et no. 3), étant poursuivies par un staff international dirigé par le Prof. Univ. Dr. Marin Carciumaru (Université «Valahia», Faculté de Sciences Humaines, Targoviste). A l’occasion du déca-page du niveau Epigravettien II, daté grosso modo sur les bases radiométriques aux environ 16.000-19.000 uncal BP ont été récupérés trois armatures en ivoires (pointes de sagaies) frag-mentaires et en état de conservation plus ou moins précaire. Heureusement les publications très récentes nous ont mis à la disposition des données relativement détaillées utilisables pour l’augmentation du répertoire des objets en ivoire du Paléolithique Supérieur de Roum-anie. Jusqu’à maintenant il n’a été possible pour nous d’examiner directement les objets, ni d’obtenir des images photo générales et de détail. Les descriptions qui suivent sont faites sur la base des données retrouvées dans les publications disponibles aussi bien que sur l’examen des dessins qui illustrent les mêmes sources (carciuMaru et al., 2003; 2004a; 2004b; 2005; 2006a; 2006b; 2006c; 200�a; 200�b).

Le premier objet (fig. �, n° 1), représente une pointe de sagaie du type à base sim-ple (HAHN, 1988). Il se trouve en état fragmentaire; les extrémités proximale et distale ont été cassées probablement pendant l’usage. Les conditions de conservation sont relativement

16 -

Fig. 4 - Lissoir de Cotu Miculinti. 1-3: vues générales et proposition de reconstitution de l’aspect initial (1 d’après BRUdIU, 1987: 75, 84 et fig. 12, n° 2).

- 1�

Fig. 5 - Lissoir de Cotu Miculinti. 1-2, 4: détails de la partie distale; 3: détail de la partie mésiale/face supérieure.

bonnes: on observe des portions exfoliées et partialement affectées par délitage et délami-nation, des fractures et fissures transversales. L’extrémité proximale a le contour concave asymétrique suite à la fracturation. La section de l’extrémité proximale est ovalaire. Les bordes sont rectilignes parallèles et légèrement convexes. Les parties mésiale et distale ont aussi des sections ovalaires. La fabrication a suivi plusieurs étapes d’une chaine opératoire relativement complexe. Les traces du débitage ont été effacées par les opérations ultérieures;

18 -

Fig. 6 - Lissoirs de Cotu Miculinti: 1-2 lissoir en os (2 d’après BRUdIU, 1987: 76, 84 et fig. 3, n° 3); 3-4 lissoir en bois de renne (3 d’après BRUdIU, 1987: 76, 84 et fig. 12, n° 1).

- 19

Fig. 7 - Pointes de sagaies de Piatra-neamţ - Poiana Ciresului: 1-2: vues générales (d’après CARCIUmARU et al., 2006c: 35, fig. 11/a).

20 -

Fig. 8 - Artefacts en ivoire du Gravettien de la partie Centrale et Orientale de l’Europe. 1: Predmost (d’après FEUS-TEl, 1973: pl. lXXI); 2: Kostienki I Eliseevich (d’après FEUSTEl, 1973: pl. lVIII, n° 1); 3: Kostienki I Eliseevich (d’après FEUSTEl, 1973: pl. lVIII, n° 2); 4: Predmost (d’après FEUSTEl, 1973: pl. lXIX, n° 3); 5: Grubgraben (d’après logAN, 1990: 73 et fig. VI-1).

- 21

Fig. 9 - Artefacts en ivoire du Gravettien central et est Europeén: 1 A-E: Kostienki I Eliseevich; 1 F-G: proposition de reconstitution du débitage de l’ivoire par percussion/entaillage (d’après SEmENoV, 1985: 150 et fig. 74, n°s 1-7); 2: Costesti (d’après BoRzIAC, 1994: 35 et fig. 5, n° 10); 3: Timonovka (d’après KozłowsKi, 1992: 78 et fig. 87, c-d); 4: identification spécifique et squelettique de la matière première et reconstitution du débitage par abrasion linéaire (d’après BEldImAN, 2000: 25 et fig. 5, n° 3).

22 -

on peut supposer le recours au double rainurage ou à l’entaillage en vue d’obtention d’une baguette rectangulaire ayant des sections carrées. Le façonnage a été appliqué très soigneuse-ment par raclage axial intense sur toute la longueur de la pointe, en but de régulariser sa surface et d’aménager la partie active pointue (distale). Les dimensions de l’armature sont les suivantes: longueur préservée: 21 cm; diamètre de l’extrémité proximale: 2,3/0,65 cm; diamètre de la partie proximale: 2,4/1 cm; diamètre de la partie mésiale: 2,4/1,1 cm; diamètre de la partie distale: 2/1 cm; diamètre de l’extrémité distale: 1/0,5 cm. La fracturation des extrémités a été produite probablement par impact durant l’usage (extrémité distale) et par la fixation dans la hampe (extrémité proximale) (carciuMaru et al., 2006b: 20, 35, fig. 11/a; 200�b: 15-16, 31, fig. 11/a).

Le deuxième artefact en ivoire de Poiana Ciresului (fig. �, n° 2) est une pointe de sa-gaie appartenant probablement au type à base raccourci (MONS, 1988). Il se présente en état fragmentaire; la partie distale a été cassée probablement pendant l’usage. Les conditions de conservation sont relativement précaires: on observe des larges portions exfoliées; la face inférieure est affectée longitudinalement sur sa longueur entière par délitage et délamination; on observe aussi des fractures et fissures transversales. L’extrémité proximale semble entière et a le contour convexe asymétrique (de façonnage) suite à l’aménagement par entaillage; ce-tte opération a produit sur la face supérieure des petites surfaces concaves irréguliers super-posées. La section de l’extrémité proximale est ovalaire. Les bordes sont légèrement courbes, le bord gauche convexe et celui droit concave sinueux. Les sections du fût sont probablement rectangulaires avec les marges arrondies ou ovalaires. Comme dans le cas avant, la fabrica-tion a suivi plusieurs étapes d’une chaine opératoire relativement complexe. Les traces du débitage ont été effacées par les opérations ultérieurs; on peut supposer le recours au double rainurage ou à l’entaillage en vue de l’obtention d’une baguette rectangulaire ayant des sec-tions carrées. Le façonnage a été appliqué très soigneusement par raclage axial intense sur toute la longueur de l’artefact, en but de régulariser sa surface et d’aménager la partie active pointue (distale). En plus on a l’entaillage de l’extrémité proximale en vue d’amincir cette partie de l’objet. Les dimensions de l’armature sont les suivantes: longueur préservée: 23 cm; diamètre de l’extrémité proximale: 2/1 cm; diamètre de la partie proximale: 2,3/1,3 cm; diamètre de la partie proximale et mésiale: 2,3/1,3 cm; diamètre de la partie distale: 2/1 cm. La fracturation de la partie distale a été produite probablement par impact durant l’usage (carciuMaru et al., 2006b: 20, 35, fig. 11/a; 200�b: 15-16, 31, fig. 11/a).

Les auteurs des recherches mentionnent aussi une troisième pointe en état très précaire dont on ne dispose pas de description ou illustration (carciuMaru et al., 2006b: 20; 200�b: 15-16).

Par l’amabilité du Prof. Univ. Dr. Marin Carciumaru, en août 200�, à l’occasion d’une visite sur le site de Poiana Ciresului on a puisse examiner une quatrième pièce, très proba-blement en ivoire aussi (encore inédite); il s’agit d’une baguette quasi-rectangulaire mince (longueur approx. 20 cm; largeur maximum approx. 2 cm; épaisseur approx. 0,3-0,5 cm) en état de conservation précaire.

Les armatures en ivoire de Piatra neamţ - Poiana Ciresului sont les premiéres artefacts de cette sorte connues jusqu’à maintenant sur le territoire de la Roumanie; elles contribuent d’une manière significative à l’augmentation du répertoire typologique des pointes de sagaie du Paléolithique supérieur dans les régions situées à l’est des Carpates (BELDIMAN, 2004b; 2005b).

- 23

Parmi les analogies de ces objets on peut mentionner les découvertes de Cosauti et Molodova V (CHIRICA et BORZIAC 1995; BORZIAC et al., 2006).

ConCLUSion

Les reconsidérations récentes des quelques découvertes anciennes ont mis en lumière l’existence exceptionnelle de cinq artefacts en ivoire au Paléolithique Supérieur de Roum-anie, attribués au Gravettien oriental/Epigravettien: un fragment de défense de mammouth débité par entaillage et probable par sciage transversal, découvert à Lespezi et daté autour de 18.000 uncal BP; un lissoir (ou poignard ?) découvert à Cotu Miculinti et daté autour de 19.000 uncal BP; trois pointes de sagaies découvertes dans le site de Piatra-neamţ - Poiana Ciresului et daté entre 16.000 et 19.000 uncal BP. Sur le plan paléotechnologique, le frag-ment de défense de Lespezi attire l’attention par l’attestation précoce probable du procédé de débitage par sciage transversal à la ficelle. Étant donné l’absence dans les deux sites men-tionnés ci-dessus d’autres témoignages signalés concernant le travail de l’ivoire, ainsi que la rareté des restes squelettiques de mammouth et la rareté générale des ivoires ouvrés au Paléo-lithique supérieur en Roumanie, on peut envisager une provenance extérieure, allogène, de cette catégorie d’artefacts, emportés par les communautés de chasseurs de rennes et grands herbivores de l’aire centrale ou orientale de l’Europe (les territoires actuels de la République Tchèque, d’Ukraine, de Russie et de la République de Moldavie).

La rareté générale des objets en ivoire dans les régions actuelles du pays génère un con-traste évident en rapport avec la situation connue à l’est du Prut (koZłowski, 1992; CHIRICA et BORZIAC, 1995; CHIRICA, 1996a). Une situation analogue est liée à des manifestations diverses de l’art mobilier (BORZIAC et CHIRICA, 1996); par exemple, jusqu’alors, après un demi-siècle de recherches systématiques, dans les sites paléolithiques roumains, aucune représentation animale ou humaine en ronde bosse n’a été encore retrouvé (OTTE et al., 1995; OTTE et BELDI-MAN, 1995; BELDIMAN, 2004a; 2005a; BELDIMAN et SZTANCS, 2006). Cette situation attend encore une explication plausible; elle n’est pas forcément attribuable au stade des recher-ches et reflète, probablement, des particularités culturelles de ces régions au Paléolithique Supérieur1.

______________________________________

1  Les photos et les dessins sans spécifications en légende de la figure appartiennent à l’auteur principal (CB).

24 -

R E F E R E n C E S

BELDIMAN, C. 1996 - Asupra utilizarii fildesului in paleoliticul superior din Romania (Sur l’utilisation de l’ivoire au Paléolithique supérieur de Roumanie). Studii şi cercetari de istorie veche şi arheologie, 4� (3): 325-333. Bucureşti (en roumain).

BELDIMAN, C. 1999a - Date privind industria paleolitica a materiilor dure animale in asezari din zona subcarpatica a Moldovei (Donées sur l’industrie des matières dures animales dans les sites de la région du piedmont de la Mol-davie). Analele Universitatii Crestine “dimitrie Cantemir”, Seria Istorie, 3: 41-71. Bucureşti (en roumain).

BELDIMAN, C. 1999b - Industria materiilor dure animale in paleoliticul superior, epipaleolitic, mezolitic si neoliticul timpuriu din Romania (l’industrie des matières dures animales au Paléolithique supérieur, à l’Épipaléolithique, au mésolithique et au Néolithique ancien de Roumanie). Thèse de doctorat, Institut d’Archéologie “Vasile Parvan”. Académie Roumaine, Bucarest (en roumain).

BELDIMAN, C. 2000 - Industria materiilor dure animale in asezarile paleolitice de pe versantul rasaritean al Carpatilor (L’industrie des matières dures animales dans les sites paléolithiques de la partie est des Carpates). Angustia, 5: �-29. Sfantu Gheorghe (en roumain).

BELDIMAN, C. 2003 - Date recente privind utilizarea fildesului in paleoliticul superior din Romania (Données ré-centes sur l’utilisation de l’ivoire au Paléolithique supérieur de Roumanie). In angelescu, M.-V., Bors, c. et Vasilescu, Fl. (dir.) Cronica cercetarilor arheologice din Romania. Campania 2002. A XXXVII-a Sesiune nationala de rapoarte arheologice, Covasna, 2-6 iunie 2003. CIMEC: 109-111, 404. Bucureşti (en roumain).

BELDIMAN, C. 2004a - Art mobilier au Paléolithique supérieur en Roumanie. In OTTE, M. (dir.), la Spiritualité. Actes du Colloque international de liège (10-12 décembre 2003). ERAUL, 106: 103-121. Liège.

BELDIMAN, C. 2004b - Bone and antler industry in the Upper Paleolithic of Romania: projectile points. In luik, H., CHOyE, A.M., BATEy, C.E. and lougas, L. (dir.) From hooves to horns, from mollusc to mammoth. manufac-ture and use of bone artefacts from Prehistoric times to the Present. Proceedings of the 4th Meeting of the ICAZ Worked Bone Research Group at Tallinn, 26th-31th August 2003: 15-32. Tallinn.

BELDIMAN, C. 2005a - L’ivoire au Paléolithique supérieur de Roumanie. In dujardin, V. (dir.), Table Ronde sur le Paléolithique supérieur récent. Industrie osseuse et parures du Solutréen au magdalénien en Europe, Angoulême (Charente, France), 28-30 mars 2003. Mémoire de la Société Préhistorique Française, 39: 2��-289. Paris.

BELDIMAN, C. 2005b - Upper Paleolithic of Romania from the perspective of bone and antler industry: projectile points. Istros, 12: 11-32. Braila.

BELDIMAN, C. 200� - Industria materiilor dure animale in preistoria Romaniei. Resurse naturale, comunitati umane şi tehnologie din paleoliticul superior pana in neoliticul timpuriu (l’industrie des matières dures animales dans la Préhistoire de la Roumanie: ressources naturelles, communautés humaines et technologie dès le Paléolithique supérieur au Néolithique ancien). Asociatia Romana de Arheologie. Studii de Preistorie Sup-plementum, 2. Éd. Pro Universitaria, Bucureşti (en roumain).

BELDIMAN, C. et SZTANCS, D-M., 2006 - Pendeloques paléolithiques et épipaléolithiques de Roumanie. Studii de preistorie, 3: 11-40. Bucureşti.

BILLAMBOZ, A. 19�� - L’industrie du bois de cerf en Franche-Comté au Néolithique et au début de l’Âge du bronze. gallia Préhistoire, 20 (1): 91-100. Paris.

Bitiri, M. et caPitanu, V. 19�2 - Asezarea paleolitica de la Lespezi, judetul Bacau (Le gisement paléolithique de Lespezi, dép. de Bacau). Carpica, 5: 39-68. Bacau (en roumain).

Bitiri-ciortescu, M., caPitanu, V. et carciuMaru, M. 1991 - Paleoliticul din sectorul subcarpatic al Bistritei in lu-mina cercetarilor de la Lespezi - Bacau (Le Paléolithique de la vallée moyenne de la rivière Bistrita en lumière des fouilles de Lespezi - Bacau). Carpica, 20: �-52. Bacau (en roumain).

BOLOMEy, AL. 1991 - Consideratii asupra resturilor de mamifere din statiunea gravettiana de la Lespezi - Lutarie (jud. Bacau) (Considérations sur les restes de mammifères du site gravettien de Lespezi - Lutarie, dép. de Bacau). Carpica, 20: 2�1-295. Bacau (en roumain).

- 25

BORZIAC, I.A. 1994 - Paleoliticul si mezoliticul in spatiul dintre Nistru si Prut (Le Paléolithique et le Mésolithique dans l’interfluve Dniestr-Prut). Thraco-dacica, 15 (1-2): 19-40 Bucureşti (en roumain).

BORZIAC, I.A., CHIRICA, V. et Valeanu, M.C. 2006 - Culture et sociétés pendant le Paléolithique supérieur à travers l’espace carpato-dniestréen. Bibliotheca Archaeologica Moldaviae, 6. PIM, Iaşi.

BORZIAC, I.A. et CHIRICA, C.V. 1996 - Pièces de marne du Paléolithique supérieur de la vallée du Dniestr. Préhistoire Européenne, 9: 393-401. Liège.

BOSINSKI, G. 1990 - Homo sapiens. Histoire des chasseurs du Paléolithique supérieur en Europe (40000 - 10000 av. J.C.). Errance, Paris.

Brudiu, M. 19�9 - Rezultatele cercetarilor arheologice din statiunea paleolitica de la Cotu Miculinti, com. Cotusca (jud. Botosani) (Résultats des fouilles archéologiques dans le site paléolithique de Cotu Miculinti, comm. Cotusca, dép. de Botosani). In STOICA, C. (dir.) Materiale şi cercetari arheologice. A XIII-a Sesiune anuala de rapoarte arheologice, oradea, 1979: �-16. Oradea (en roumain).

Brudiu, M. 1980a - Cercetari arheologice in statiunea paleolitica de la Cotu Miculinti, jud. Botosani (Fouilles archéologiques dans le site paléolithique de Cotu Miculinti, dép. de Botosani). In STOICA, C. (dir.) Materiale şi cercetari arheologice. A XIV-a Sesiune anuala de rapoarte arheologice, Tulcea, 1980: 5-12. Tulcea (en roumain).

Brudiu, M. 1980b - Prelucrarea oaselor si coarnelor de ren in asezarea paleolitica de la Cotu Miculinti (jud. Botosa-ni) (L’industrie de l’os et du bois de renne dans le site paléolithique de Cotu Miculinti, dép. de Botosani). Studii şi cercetari de istorie veche şi arheologie, 31 (1): 13-22 (en roumain).

Brudiu, M. 1983 - Rezultatele cercetarilor paleolitice din statiunea de la Cotu Miculinti (jud. Botosani) (Résultats des fouilles archéologiques dans le site paléolithique de Cotu Miculinti, dép. de Botosani). In STOICA, C. (dir.) Materiale şi cercetari arheologice. A XV-a Sesiune anuala de rapoarte arheologice, Brasov, 1983: 7-13. Bucureşti (en roumain).

Brudiu, M. 1986 - Sapaturile arheologice de la Cotu Miculinti (Les fouilles archéologiques dans le site paléolithique de Cotu Miculinti). In STOICA, C. (dir.) Materiale şi cercetari arheologice. A XVI-a Sesiune anuala de rapoarte arheologice, Vaslui, 1986: 5-8. Bucureşti (en roumain).

Brudiu, M. 198� - Le travail de l’os et du bois de renne dans le Paléolithique supérieur de la zone du Prut moyen. Répertoire typologique. In CHIRICA, V. (dir.) la genèse et l’évolution des cultures paléolithiques sur le ter-ritoire de la Roumanie. Bibliotheca Archaeologica Iassiensis, II: 73-86. Iaşi.

Brudiu, M. 1994 - Industria cornului si osului in paleoliticul superior din nord-estul Romaniei (L’industrie de l’os de du bois de renne au Paléolithique supérieur du Nord-est de la Roumanie). Studii şi cercetari de istorie veche şi arheologie, 45 (3): 275-284. Bucureşti (en roumain).

Brudiu, M. 2006 - Prelucrarea osului si cornului in gravetianul oriental din spatiul carpato-nistrean: resurse, reperto-riu, ocupatii (Le travail de l’os et du bois de renne dans le Gravettien oriental de l’espace carpato-dniestréen: ressources, répertoire, occupations). Acta musei Tutovensis, 1: 45-4�. Barlad (en roumain).

carciuMaru, M., Margarit, M., anghelinu, M., nita, l., carstina, o., cosac, M., Plesa, M. et duMitru, Fl. 2003 - Les dé-couvertes d’art paléolithique de la vallée de Bistrita dans le contexte de l’art mobilier paléolithique de Roumanie. An-nales de l’Université «Valahia» Targoviste, Section d’Archéologie et d’Histoire, 4-5: 16-2�. Targoviste (en roumain).

carciuMaru, M., anghelinu, M., Margarit, M., cosac, M. et CARSTINA, O. 2004a - Piatra neamţ-Poiana Ciresului. In angelescu, M.-V., oBerländer-tarnoVeanu, i. et Vasilescu, Fl. (dir.) Cronica cercetarilor arheologice din Romania. Campania 2003. A XXXVIII-a Sesiune nationala de rapoarte arheologice, Cluj-Napoca, 26-29 mai 2004. CIMEC: 231-232. Bucureşti (en roumain).

carciuMaru, M., Margarit, M., nita, l., anghelinu, M., cosac, M. et CARSTINA, O. 2004b - Les découvertes d’art mobilier paléolithique de Poiana Ciresului-Piatra neamţ (Roumanie). In OTTE, M. (dir.) la Spiritualité. Actes du Colloque international de liège (10-12 décembre 2003). ERAUL, 106: 123-126. Liège.

carciuMaru, M., anghelinu, M., Margarit, M., cosac, M. et CARSTINA, O. 2005 - Piatra neamţ-Poiana Ciresului. In angelescu, M.-V., oBerländer-tarnoVeanu, I. et Vasilescu, Fl. (dir.) Cronica cercetarilor arheologice din Romania. Campania 2004. A XXXIX-a Sesiune nationala de rapoarte arheologice, Jupiter-mangalia, 25-28 mai 2005. CIMEC: 265-267, 495-497. Bucureşti (en roumain).

26 -

carciuMaru, M., anghelinu, M., nita, l., Margarit, M., cosac, M., duMitru, Fl., duMitrascu, V., steguweit, l. et CARSTINA, O. 2006a - Piatra Neamt-Poiana Ciresului. In angelescu, M.-V., Vasilescu, Fl. (dir.) Cronica cercetarilor arheologice din Romania. Campania 2005. A Xl-a Sesiune nationala de rapoarte arheologice, Constanta, 31 mai - 3 iunie 2006. CIMEC: 262-264. Bucureşti (en roumain).

carciuMaru, M., anghelinu, M., steguweit, l., nita, l., Fontana, l., Brugère, al., haMBach, u., Margarit, M., duMitrascu, V., cosac, M., duMitru, Fl. et carstina, o. 2006b - The Upper Paleolithic site of Poiana Ciresului-Piatra Neamt (North-Eastern Romania). Recent results. Archäologisches Korrespondenzblatt, 36: 319-331. Berlin.

carciuMaru, M., anghelinu, M., lucas, g., nita, l., steguweit, l., Margarit, M., Fontana, l., Brugère, al., duMitrascu, V., haMBach, u., cosac, M., carstina, o. et duMitru, Fl. 2006c - Paleoliticul superior de la Poiana Ciresului (Piatra neamţ). noi rezultate, interpretari si perspective (Le Paléolithique supérieur de Poi-ana Ciresului - Piatra Neamt. Nouvelles résultats, interprétations et perspectives). Cercetari arheologice, 13: 11-37. Bucureşti (en roumain).

carciuMaru, M., anghelinu, M., nita, l., Margarit, M., cosac, M., duMitru, Fl., duMitrascu, V., steguweit, l., haMBach, u. et CARSTINA, O. 200�a - Piatra Neamt-Poiana Ciresului. In angelescu, M.-V. et Vasilescu, Fl. (dir.) Cronica cercetarilor arheologice din Romania. Campania 2006. A XlI-a Sesiune nationala de rapoarte arheologice, Constanta, 29 mai - 1 iunie 2007. CIMEC: 263-265. Bucureşti (en roumain).

carciuMaru, M., anghelinu, M., lucas, g., nita, l., steguweit, l., Margarit, M., Fontana, l., Brugère, al., duMitrascu, V., haMBach, u., cosac, M., carstina, o. et duMitru, Fl. 2007B - Santierul paleolitic de la Poi-ana Ciresului (Piatra Neamt): o sinteza a rezultatelor recente (1998-2005) (Le chantier archéologique paléo-lithique de Poiana Ciresului - Piatra neamţ: synthèse des résultats récentes, 1998-2005). Materiale şi cercetari arheologice, Serie noua, 2: 5-32. Bucureşti (en roumain).

CHIRICA, C.V. 1996a - Arta si religia paleoliticului superior in Europa Centrala si rasariteana (l’art et la religion du Paléolithique supérieur en Europe Centrale et orientale). Bibliotheca Archaeologica Iassiensis, VI. Ed. Helios, Iaşi (en roumain).

CHIRICA, C.V. 1996b - Manifestari artistice in paleoliticul superior in sud-estul Europei (Manifestations d’art au Paléolithique supérieur dans le Sud-est de l’Europe). In carciuMaru, M. et diaconescu, M. (dir.) lucrarile Simpozionului de arheologie, Targoviste, 23-25 noiembrie 1995: 43-45. Targoviste (en roumain).

CHIRICA, C.V. 2001 - gisements paléolithiques de mitoc. le Paléolithique supérieur de Roumanie à la lumière des découvertes de mitoc. Bibliotheca Archaeologica Iassiensis, XI. Ed. Helios, Iaşi.

CHIRICA, C.V. 2004 - Les significations artistiques et religieuses de certaines découvertes paléolithiques de l’espace carpato-dniestréen. In OTTE, M. (dir.) la Spiritualité. Actes du Colloque international de liège (10-12 décem-bre 2003). ERAUL, 106: 1��-186. Liège.

CHIRICA, V. 1989 - The gravettian in the East of the Romanian Carpathians. Bibliotheca Archaeologica Iassiensis, III. Institut d’Histoire et d’Archéologie ‘A.D. Xenopol’, Iaşi.

CHIRICA, V. et BORZIAC, I.A. 1995 - Les ivoires du Sud-Est de l’Europe: Bulgarie, Grèce, yougoslavie et Roumanie jusqu’au Dniestr. In Menu, M., taBorin, y., walter, Ph. et wideMan, F. (éds.) le travail et l’usage de l’ivoire au Paléolithique supérieur. Actes de la Table ronde de Ravello, 29-31 mai 1992: 199-210. Ravello.

CHIRICA, V. et TANASACHI, M. 1984-1985 - Repertoriul arheologic al judetului Botosani (Répertoire archéologique du département de Botosani), vol. I-II. Institut d’Histoire et d’Archéologie ‘A.D. Xenopol’, Iaşi (en roumain).

CHRISTENSEN M. 1999 - Technologie de l’ivoire au Paléolithique supérieur: caractérisation physico-chimique du matériel et analyse fonctionnelle des outils de transformation. BAR International Series, �51. Archaeopress, Oxford.

CLOTTES, J. (dir.) 1990 - l’art des objets au Paléolithique. Vol. I: l’art mobilier et son contexte. Vol. II: les voies de la recherche, Actes du Colloque international de Foix - mas d’Azil novembre 1987. Colloques internationaux du CNRS, Paris.

Feustel, r. 19�3 - Technik der Steinzeit. Archäolithikum - mesolithikum. Veröffentlichungen der Museums für Ur- und Frühgeschichte Thüringens 4, Weimar.

- 2�

HAHN, J. 1988 - Fiche Sagaie à base simple de tradition aurignacienne (1). In delPorte, h., hahn, j., Mons, l., Pinçon, g. et sonneVille-Bordes, d. de (éds.) Fiches typologiques de l’industrie osseuse préhistorique. Cahier I. Sagaies, U.I.S.P.P., Commission de nomenclature sur l’industrie de l’os préhistorique, responsable H. Camps-Fabrer. Publications de l’Université de Provence, Aix-en-Provence.

koZłowski, J.K. 1992 - l’art de la Préhistoire en Europe orientale. CNRS Éditions, Paris.

LOGAN, B. 1990 - The hunted of Grubgraben: an analysis of faunal remains. In Montet-white, A. (dir.) The Epi-gravettian site of grubgraben, lower Austria: The 1986 and 1987 excavations. ERAUL, 40: 65-85. Liège.

luMley, h. de (dir.) 1984 - Art et civilisation de chasseurs de la préhistoire (34000 - 8000 ans av. J.Ch.). Catalogue d’exposition, Paris.

Macgregor, a. 1985 - Bone, antler, ivory and horn. The technology of skeletal materials since the Roman period. Croom Helm, London and Totowa, N.J.

MONS, L. 1988 - Fiche Sagaie à base raccourci (5). In delPorte, h., hahn, j., Mons, l., Pinçon, g., sonneVille-Bordes, d. de (éds.) Fiches typologiques de l’industrie osseuse préhistorique. Cahier I. Sagaies. U.I.S.P.P., Commission de nomenclature sur l’industrie de l’os préhistorique, responsable H. Camps-Fabrer. Publications de l’Université de Provence, Aix-en-Provence.

OTTE, M. et BELDIMAN, C. 1995 - Sur les objets paléolithiques de parure et d’art en Roumanie: une pendeloque en os découverte à Mitoc, dép. de Botosani, Roumanie. memoria Antiquitatis, 20: 35-70. Piatra neamţ.

OTTE, M., CHIRICA, V. et BELDIMAN, C. 1995 - Sur les objets paléolithiques de parure et d’art en Roumanie: une pen-deloque en os découverte à Mitoc, dép. de Botosani, Roumanie. Préhistoire européenne, 5: 119-152. Liège.

Păunescu, al. 1989 - Structures d’habitat moustériennes mises au jour dans l’établissement de Ripiceni - Izvor (Roumanie) et quelques considérations concernant le type d’habitat au Paléolithique moyen à l’Est des Car-pates. In Patou, M. et FreeMan, l.g. (éds.) l’Homme de Néanderthal. Actes du Colloque international de liège (4-7 décembre 1986). Vol. VI: la subsistance. ERAUL, 33: 12�-143. Liège.

Păunescu, al. 1993 - Ripiceni - “Izvor”. Paleolitic şi mezolitic. Studiu monografic (Ripiceni - “Izvor”. Paléo-lithique et mésolithique. monographie archéologique). Ed. Academiei, Bucureşti (en roumain).

Păunescu, al. 1998 - Paleoliticul si epipaleoliticul de pe teritoriul moldovei cuprins intre Carpati si Siret. Stu-diu monografic (le Paléolithique et l’Épipaléolithique de la moldavie entre les Carpates et la rivière Siret. monographie archéologique). Ed. Satya Sai, Bucureşti (en roumain).

Păunescu, al. 1999 - Paleoliticul şi mezoliticul de pe teritoriul Moldovei cuprins intre Siret şi Prut. Studiu monografic (le Paléolithique et le mésolithique de la moldavie entre les rivières Siret et Prut. monographie archéologique). Ed. Satya Sai, Bucureşti (en roumain).

POPLIN, F. 19�4 - Deux cas particuliers de débitage par usure. In caMPs-FaBrer, H. (dir.) Premier Colloque interna-tional sur l’industrie de l’os dans la préhistoire, Abbaye de Sénanque, avril 1974. Aix-en-Provence.

SEMENOV, S.A. 1985 - Prehistoric Technology. An experimental study of the oldest tools and artefacts from traces of manufacture and use. Totowa-New Jersey.

Zaharia, n., Petrescu-diMBoVita, M., Zaharia, e. 19�0 - Asezari din moldova. de la paleolitic pana in secolul al XVIII-lea (Sites archéologiques de la moldavie. de Paléolithique jusqu’au XVIIIe siècle). Ed. Academiei, Bucureşti (en roumain).

Addresses des auteurs: CORNELIU BELDIMAN, Université Chrétienne «Dimitrie Cantemir», Faculté d’Histoire, Splaiul Unirii Nr. 1�6, Sector 4 – RO - 040042 BUCAREST 53.e-mail: cbeldiman58@yahoo.com

DIANA-MARIA SZTANCS, Université «Lucian Blaga», Faculté d’Histoire et pour l’Étude du Patrimoine «Nicolae Lupu», Master ‘Protection et Étude du Patrimoine Historique’, Bd. Victoriei Nr. 5-� – RO - 550024 SIBIU

- 29

Atti Soc. Preist. Protost. Friuli-V.G., Trieste, XVI, 2006-200� (2008): 29-51

VLADIMIR N. STANKO∗ and DMyTRO KIOSAK∗∗

tHE MESoLitHiC SEttLEMEnt oF tHE LowER DAnUbE-PRUt-DniEStER-SoUtH bUH intERFLUViALS:

tHE EARLY MESoLitHiC ASSEMbLAgES

SUMMARY – The mesolithic settlement of the lower danube-Prut-dniester-South Buh interfluvials: the Early mesolithic assemblages. This paper is a critical review of the present-day knowledge of the Early Mesolithic pe-riod of southwestern Ukraine. The sites of uncertain typological attribution were not taken into consideration. The general picture, which has been outlined on the basis of both settlement locations and structures and the chipped stone assemblages recovered from excavations and surveys, shows that the Early Mesolithic industries of this region undoubtedly differ from those already known from other territories of south and eastern Europe, mainly for the systematic occurrence of isosceles trapezes with retouched short side. The only site of this period so far radiocarbon-dated is Bilolisja, which gave a result of 8900±190 uncal BP (Ki-10886).

RiASSUnto – Il mesolitico del bacino del basso danubio-Prut-dniester-Bug: le industrie del mesolitico An-tico. Il presente lavoro consiste in una revisione critica delle attuali conoscenze sul Mesolitico antico dell’Ucraina sudoccidentale. I siti di incerta attribuzione non sono stati presi in considerazione. Il quadro generale che deriva dallo studio sia degli insediamenti e delle loro strutture, sia delle industrie litiche raccolte durante gli scavi e le ricognizioni di superficie, indica che i complessi del Mesolitico antico di questa regione, uno solo dei quali, Bilolisja, è stato datato radiometricamente a 8900±190 uncal BP (Ki-10886), sono indubbiamente diversi da quelli di altre regioni dell’Europa meridionale ed orientale, anche per la presenza di trapezi isosceli su lamella con il lato corto ritoccato.

Keywords: Ukraine, Prychornomor’a, Mesolithic, Bilolisja complex, Tsarynka tradition.

1. intRoDUCtion

The Mesolithic sites discussed in this paper were discovered in southwestern Ukraine since the late 1950s (fig. 1). Until these years, many researchers believed that the Meso-lithic period, characterised by microlithic assemblages, was not attested in northwestern Prychornomor’a (BIBIKOV, 1953: 103-105) and, consequently, that the last Palaeolithic hunters had interacted with the first Neolithic farmers of the Trypillja Culture. In 1954, P.I. ______________________________________

* Department of History, Mykolajiv Humanitarian State University P. Mohyla, UA** Faculty of Mathematics, Physics and Natural Sciences, Ferrara University, I

30 -

Boriskovskij discovered the site of Grebenyky, where an assemblage with geometric mi-croliths was found for the first time (BORISKOVSKIJ, 195�: 5-�; KOROBKOVA, 195�: 59-62; GRIGOR’EVA, 1961). Extensive field surveys, undertaken between 1955 and 1960, led to the discovery of more than 10 sites with geometric tools: Zavallja in the southern Buh, Myhajliv-ka (Bilolisja), Vasylivka, Dobrozhany, Baraboj I-IV and others. By 19�5 more than 50 sites were known in northwestern Prychornomor’a, most of which characterised by geometrics and other microlithic tools (KRASKOVSKIJ, 19�8: 40-61).

In the 19�0s V.N. DANyLENKO (1969: 46-14�) and V.I. MARKEVICH (19�4: 12�-144) sug-gested that the recently discovered Neolithic settlements of the Buh-Dniestern Culture origi-nated from the local Mesolithic background with a geometric chipped stone industry.

During the 1960s and 19�0s large-scale excavations were carried out at Ghyrzheve, Bilolisja (Mykhajlivka), Myrne and Frumushika (KRASKOVSKIJ and STANKO, 1966; STANKO, 1966; 196�; 19�1; 19�2; 19�6; 19��; 1980; 1982; CHERNySH, 19�5), and a few other new sites were investigated (CHETRARU, 19�3; KRASKOVSKIJ, 19�8).

The Mesolithic sites of the so-called Kukrek Culture (Synjukhin Brid, Abuzova Balka and others) were brought to light during surveys aimed at the discovery of sites with mi-croliths. Their complexes are characterised by a high number of microlithic tools among which are very few types of geometrics (DANILENKO, 1969: 5�-61; STANKO et al., 1981: 5-12).

The last quarter of the last century was marked by studies on the cultural history, palaeoecono-my and social development of the Mesolithic hunter-gatherers in the northwestern Prychornomor’a (COVALENKO and CHETRARU 199�; COVALENKO and COJ, 1999; STANKO et al., 1999).

The debates of the last decades resulted in a number of different approaches regard-ing the chronology, periodisation and origin of the Mesolithic in the Prut-Buh interfluvial. Nevertheless it is important to point out that up-to-now there is still no general agreement in the definition of the Early Mesolithic. V.N. Stanko attributes to this period the site he exca-vated at Bilolisja, and some other localities with characteristic surface material (Bilolisja IV, Kantemir, Vasylivka, Tsarynka, Gavryliv Jar). Both this author and D.Ja. Telegin suggested that the Kukrek type industries formed during an early stage at the beginning of the Meso-lithic, because they show many typological characteristics that relate them to the local Epi-gravettian tradition. A few other authors attribute all these sites to the end of the Palaeolithic (otherwise called Final Palaeolithic) (KOROBKOVA, 1989; 1989a; JANEVICH, 1990; GRIGOR’EVA, 1992; ZalіZnyak, 1995: 18). According these latter, around the beginning of the Holocene, the region was 1) uninhabited or 2) any trace of the Early Mesolithic settlement has not so far been discovered.

At present there is still no uniform view of the origin and chronology of the Late Me-solithic assemblages with geometric (Grebenyky Culture) and non-geometric microliths (Kukrek Culture or cultural entity). For instance V.N. STANKO (1986; 2006) believes that the Grebenyky Culture is rooted into the local tradition, L.L. ZalіZnyak (1998; 2005; 2005a; 2006) hypothesizes that the Grebenyky industry results from a direct migration of peoples from the Balkans, while other specialists have proposed other approaches.

Alongside with the above theoretical approaches and field investigations, radiocar-bon datings, geological and palaeoclimatic studies were undertaken in the 1980s and 1990s (STANKO and SVEZHENCEV, 1988; ZAJTSEVA et al., 199�; telegіn, 2002).

- 31

Fig. 1 - The research area with the location of the sites Bilolisja (1), Bilolisja IV (2), Kaghyl’nyk (3), Kantemir (4), Gavryliv Jar (5) and Tsarynka (6).

In this paper, the analysis of the archaeological assemblages follows the scheme pro-posed by V.N. STANKO (19�6), although the ideas of other authors are also reported from time to time, in order to provide the reader with an objective picture of the present state of the research and the interpretation of the many aspects of both each particular site and the entire Mesolithic period.

32 -

2. tHE EARLY MESoLitHiC SitES

In the Buh-Prut region, the end of the Pleistocene and the beginning of the Holocene are marked by the development of three different types of chipped stone assemblages. The Bilolisja type industry is distributed in the Danube-Dniester interfluvial. It is characterised by lunates, end scrapers on irregular blades and wide flakes, and circular scrapers, sub-prismatic cores, while burins are almost absent. Flakes predominate over blades in the debitage and the bladelets are the least represented group.

The Tsarynka type-sites were discovered in the north steppe zone of the Dniester-Buh interfluvial. Here a specific type of trapeze with retouched short side predominates, the luna-tes are absent, end scrapers and burins are obtained from thick blades and flakes. The blade-lets are few. Both the Bilolisja and Tsarynka industries do not seem to derive from the local Epigravettian complex, whose most probable descendant is the Kukrek industry (STANKO, 1982: 113-114; telegіn, 1982: 118-119).

2.1. tHE EARLY MESoLitHiC SitES oF tHE wEStERn PARt oF tHE REgion: tHE DAnUbE-DniEStER intERFLUViAL

Bilolisja is the most important Early Mesolithic site in the region. It was discovered by

A.M. Kremer in 195� and, after a further survey, it was published by V.I. KRASKOVSKIJ and A.M. KREMER (1959). In 1965-66 and 19�� V.N. STANKO (1985) excavated most of the site.

bilolisjaBilolisja is situated on the lower loess terrace facing the Sarata old riverbed. Its geographi-

cal coordinates are 45°56’04”N-29°41’08”E (fig. 2). The present-day course of the river origi-

Fig. 2 - Bilolisja: the Mesolithic site is located where the two people stand (photograph by P. Biagi).

- 33

Fig. 3 - Bilolisja: stratigraphy of the site according to V. DUBNJAK (pers. comm. 19��).

nated during the Riss-Würm interstadi-al, inside a more ancient valley covered by Late Pleistocene loess sediments. Around the beginning of the Holocene, the height of the Bilolisja terrace was not lower than that of the first terrace of the Danube (PETROUGNE, 19�1).

The archaeological occupation was found in situ in the lower part of layer Phi, at a depth of �0-80 cm, just above the lower-lying green pale-yellow loam horizon. Isolated finds were collected at the depth of 40-�0 cm, although this layer did not yield any trace of an ar-chaeological horizon in situ. The flint assemblage from this layer is represent-ed by trapezes with a retouched short side (fig. 6, n. 9), circular scrapers, two small double scrapers with retouched sides and a small number of flakes and blades. The modern topsoil did not yield any flint artefact.

The palynological analysis re-vealed four main zones (PASHKEVICH, 1981). The pollen spectrum of the low-

est zone corresponds to the tussock-cereal steppe vegetation cover. The pollens from the cul-tural layer Phi are typical for a motley grass-cereal steppe environment. The arboreal flora is represented by a small percentage of pine, birch and alder. The following upper zone (33-54 cm) is supposed to represent the Atlantic period (fig. 4).

Animal species niSP % Mni %

Equus caballus 65 �9.3 � 53.8

Bos Primigenius 10 12.2 3 23.1

Saiga tatarica 6 �.3 2 15.4

Vulpes vulpes 1 1.2 1 �.�

Large ungulates 93 - 89 -

Unidentifiable 89 - - -

total 264 100.0 13 100.0

% of unidentifiable ca. 34.0

Table 1 - Bilolisja: faunal assemblage (after BIBIKOVA, 19�8)

34 -

Fig. 4 - Bilolisja: pollen diagram by G.A. PASHKEVICH (1981).

The cultural layer yielded both animal bones (264 specimens, 89 of which unidentifi-able) and land snails (BIBIKOVA, 19�8; unpubl.). The faunal remains are characteristic of an open steppe landscape. The presence of Saiga tatarica bones would indicate a relatively arid environment (BIBIKOVA, 19�8: 28).

The visible part of the profile revealed the following sequence (fig. 3). Chernozem soil on the humus. Crumbly, loose sandy loam, 40-50 cm thick (layer Н). Pale-yellow, brownish, powder-sandy loam, 30-40 cm thick. V.A. DUBNJAK (pers. comm. 19��) subdivided it into two sub-horizons: dark-grey with pale-yellow inclusions (layer Нр: 45-65 cm) and pale-yel-low, brownish (layer Phi: 65-80 cm). The greenish, pale-yellow loam, 10-15 cm thick (layer Pk) is heavy, clayey, with siltstones, rich in consolidated calcareous sediments. The greenish, pale-yellow loam below is micaceous, microporous slowly turning into loam sand. Loam sand and sand (below 130 cm).

The sediments containing the cultural layer formed during the Early Holocene (PETROUGNE, 19�1: 115; BIBIKOVA, 19�8: 5�-61; PASHKEVICH, 1981: 83). This is confirmed also by a radiocarbon date obtained from a bone sample: 8900±190 uncal BP, 8550-�550 cal BC at 2σ (Ki-10886).

The cultural layer, 10 cm thick, is characterised by an increasing number of flint tools and animal bones. Its lowermost horizon showed the presence of two hearths, most probably excavated into the Mesolithic surface. Four main scatters of flints can be observed on the excavated surface, along the riverbank (fig. 5).

- 35

The first scatter was found in squares 2е-5гв. The preserved part of the spot has a di-ameter of 4 m. The finds include flakes, blades, shatters, scrapers, lunates, burins, retouched blades, animal bones and small charcoal pieces.

The second, oval-shaped scatter was excavated 3 m west of the first, in squares 9-13еб. Its diameter is slightly wider than 4 m. A circular hearth, 1.2 m in diameter, was uncovered in the eastern part of the accumulation, deepening in the upper layer of the greenish loam. Its fill consisted of very fragmented burnt bones, small charcoal pieces and a few burnt flints. The flint assemblage is composed of scrapers, burins, lunates, blades with oblique trunca-tion, perforators, retouched flakes and blades, shatters. Cores, tools and debitage wastes were found mainly west of the hearth.

The third scatter was discovered 1.5 m west of the second, in squares 15-18га. The inventory consisted of 1 core, a group of flakes and blades, two scrapers, lunates, retouched blades, shatters and other fragments.

Fig. 5 - Bilolisja: distribution map of the flint art e facts within a 1 sq m grid: unretouched pieces (a), cores and tools (b), trapeze (T), H cores (H) and retouched pieces (▲).

36 -

Further to the west, a relatively thick concentration of finds can be observed over a wide area (sq. 20-31еа3). This part of the site might represent the remains of a few other scatters, disturbed in antiquity. In this area, the green, pale-yellow loam horizon, below the cultural layer, is heavily damaged. It was not found also along the western edge of the excavation trench. We can tentatively delineate only one scatter (5 m Ø) with a hearth in its centre and a few smaller scatters.

The fourth scatter was excavated in squares 22-26в-а3 in the same alignment of the al-ready-described scatters along the riverbank. The hearth, which consisted of an oval-shaped thin layer of ash, 50-60 cm in diameter, is badly preserved. The flint assemblage is composed of cores, scrapers, burins, lunates, trapezes, backed bladelets, retouched blades and flakes.

An alignment of micro-scatters is situated along the wall of the south trench. It is sepa-rated by an empty space some 3 m wide from the high-density areas near the western hearth. The best-preserved structure is a circular scatter uncovered in squares 28-31ев, which con-tained a relatively high number of cores, rejuvenation flakes, microflakelets and shatters.

The petrographic analysis of the chipped stone artefacts indicates that the inhabitants of the site exploited Dniester or Prut flint pebbles, which were the most important raw ma-terials employed also for the manufacture of the Late Mesolithic industries of the region (PETROUGNE, 19�1: 11�).

The cores include single and double platform, prismatic and sub-prismatic specimens (fig. 5: 18-20)

Category number %

Flaked pebble 1 0.1

Cores 50 4.2

Core-like fragments, pieces of cores and rejuvenation flakes 50 4.2

Flakes �19 60.2

Blades 3�5 31.3

total 1195 100.0

Tools 1�2 14.8

On flakes 69 6.0

On blades 106 8.8

Table 2 - Bilolisja: frequency and percentage of the main technological groups and debris types.

The flat forms represent more than 50% of the total assemblage. They often have addi-tional use wears. All the cores are of a small dimension. Their length does not usually exceed 4 cm.

The debitage consists mainly of flakes, microflakelets, shatters and splintered pieces. There are a few prismatic blades with parallel ridges and sides. Most blades are irregular and with a wavy profile.

- 3�

Fig. 6 - Bilolisja chipped stone assemblage: lunates (1-�, 10), trapezes (8, 9 and 13), curved backed points (11, 15), ‘quadrangle’ (12), point (14), truncations (16, 1�, 19), burin (18), end scrapers (20-29) and cores (30-32).

38 -

tool types number %

Scrapers �9 45.4

end scrapers on blade 19

end scrapers on shortened blade 12

end scrapers on flake 6

circular �

on flake 24

on core-like fragment 2

double 3

parts of working edges of scrapers 6

Burins 15 8.6

on fracture 6

on truncation 3

dihedral 3

multiple 3

Geometric microliths 23 13.2

segments 19

trapezes 2

triangles 1

backed blade with two opposed truncations 1

Points 15 8.6

truncations 4

borers (with alternate retouch) 5

perforators 5

curved backed point 1

Composite tools 5 2.9

backed blades and bladelets 5 2.9

retouched blades and bladelets 13 �.4

retouched flakes 15 8.6

notched pieces 4 2.4

total 174 100.0

Table 3 - Bilolisja: list of the tool types.

- 39

The scrapers were obtained from both flakes and blades. The scrapers on thick, irregular blades and blade-like flakes predominate over the end scrapers.

There is a high percentage of short and shortened blanks (fig. 6, nn. 22 and 23). Fan-shaped end scrapers with a thick working edge constitute a small, distinctive group (fig. 6, n. 24). There are a few end scrapers with a wedge-like working edge. The circular and semi-circular end scrapers on a flake represent some 40% of the end scrapers (fig. 6, nn. 26-29).

The burins are few and atypical, made on both flakes and blades (fig. 6, n. 18).The class of the geometric microliths is very significant, with various types of lunates.

They are represented both by elongated specimens, obtained from rough, irregular blades and small, finely retouched types on thin blanks (fig. 6, nn. 1-� and 10).

The trapezes are of two different types: the first has a retouched short side (fig. 6, nn. 8 and 9), while the second is obtained from a wide blade with two opposite oblique, abrupt straight truncations one of which is direct, the other inverse (fig. 6, n. 13). One isosceles triangle and one ‘quadrangle’1 were also recovered (fig. 6, n. 12).

Oblique truncations, perforators and borers are all from blade-like flakes, or irregular blades (fig. 6, nn. 15-1� and 19). An arched, backed point was made on a thick blade. It re-calls the Azilian points or lancets (fig. 6, n. 11).

To sum up, the assemblage from Bilolisja can be attributed, as well as other complexes, to the ‘Danubian Mesolithic’, defined by J.K. koZłowski (19�3).

G.F. KOROBKOVA (1989) studied the functional analysis of the assemblage (table 4).

tool types number %

Projectile inserts 34 9.3

Knives 123 33.�

Scrapers �8 21.4

Side-Scrapers 18 4.9

Strickles 35 9.6

Chisels 8 2.2

Burins 15 4.1

Piercers 12 3.3

Borers 8 2.2

Joiner blades 12 3.3

Saws 5 1.4

Multifunctional tools 1� 4.6

total 365 100.0

Table 4 - Bilolisja: list of the tools according to their function (after KOROBKOVA, 1989).

______________________________________

1  Backed retouched bladelet with two opposite, straight, horizontal truncations.

40 -

Fig. � - Chipped stone industries from Bilolisja IV (1 and 2), Kaghyl’nyk (3-10) and Kantemir (11-21). Trapezes with retouched short side (1 and 3), end scrapers (2, 4, 6, 8-9, 14-16, 18 and 19), notched blade (5), burin (�), re-touched blades (10-13) and cores (1�, 20 and 21).

- 41

Three more sites (Kaghyl’nyk, Kantemir and Bilolisja IV) are known only from surface collections. They were formerly attributed to the Early Mesolithic of the Danube-Dniester interfluvial.

Kaghyl’nykKaghyl’nyk is located on the low bank of the eponymous river, not far from the

point where it flows into the Lyman Sasyk (Kunduk), at 45°51’16”N-29°39’32”E. S.A. dVorjaninoV (19�6: 154-15�), who discovered of the site, collected �4 flint artefacts from its surface.

The cores (4 specimens) have a sub-prismatic shape, with one, two and three platforms. One specimen is flat. The number of blades is higher than that of the flakes. The blades are irregular and the bladelets are represented by only 4 pieces.

The end scrapers on a shortened blade (fig. �, nn. 4, 6 and 8) always predominate. The other implements include 1 atypical burin on a blade (fig. �, n. �), 1 trapeze with retouched shorter side (fig. �, n. 3), 1 notched blade (fig. �, n. 5) and 1 retouched blade (fig. �, n. 10).

KantemirKantemir (Zelene). The site of Kantemir is located on the right bank of the eponymous

river at its confluence with the Sarata, 30-40 m above the course of the river itself (KREMER, 1960).

The assemblage (105 specimens) consists of surface finds. It includes 2 flat cores (fig. �, nn. 20 and 21), scrapers of circular and semi-circular shapes on blades and flakes (fig. �, n. 14-16, 18 and 19) (nail-scrapers, scrapers with a wedge-like working edge), bladelets with a simple, marginal retouch on the side (fig. �, nn. 11-13). The blades are more numerous than the flakes. The attribution of this site to the Early Mesolithic is highly questionable.

bilolosja iVBilolosja IV (Bilolisja-Slidy) is situated on the right, low bank of the Sarata River

(DVORJANINOV, 19�6: 155), 2 km downstream from Bilolisja. The surface collection is com-posed of 8 artefacts: 1 end scraper on a blade (fig. �, n. 2), 1 semicircular scraper on a flake, 1 trapeze with retouched short side (fig. �, n. 1), 1 core-like fragment, 1 blade and 3 flakes.

2.1.1. DiscussionOnly one informative site of this period (Bilolisja) has so far been investigated in the

Danube-Dniester interfluvial. The materials from the other three sites are few and not very representative. Their attribution to the Early Mesolithic is still to be demonstrated.

The detailed study of Bilolisja gives us an opportunity to put forward a few suggestions about the social and economic life of its inhabitants. The scatters of finds along the bank of the Sarata River might represent the remains of 4(-5?) domestic activity areas, which might indicate the presence of light, temporal dwellings. The structure of the tool assemblage is typical for a seasonal hunting camp settled during a warm period of the year. The composi-tion of the faunal remains is also characteristic for the spring-summer period. Hunting horse

42 -

was optimal during the spring on heat, or by the end of the summer/beginning of the autumn, the period of herds’ formation (GORELIK, 2001: 280). Saiga tatarica grazed in the steppes of northern Prychornomor’a in the spring-summer period and moved to Crimea before the beginning of the cold season (BIBIKOVA and STARKIN, 1985).

The limited inhabited surface, and number of domestic areas, would indicate that the total number of the Bilolisja inhabitants probably did not exceed 20 individuals. The simple pattern of the space organisation is not represented by any feature due to logistic activities. Most probably the inhabitants were foragers and not collectors, following the terminology proposed by L. BINFORD (1989).

Bilolisja is the site from which the first appearance of aurochs in the steppes of the northern Prychornomor’a is represented. This unique hunting activity of the Bilolisja inhabit-ants might be explained as the infiltration of a small group of hunters from the Balkan-Danu-bian region to the most western part of steppes. The analysis of the chipped stone assemblage seems to support this hypothesis.

The flaking technique and the composition of the retouched pieces do not find any parallel in the lithic industries of the preceding Late Palaeolithic sites of the northern Prychornomor’a. The closest analogies can be extended to the assemblages from the so-called Tardi/Epigravettian of the Balkan-Danubian region, which are characterised by a high percentage of various types of abrupt retouched microliths.

If the similarities between the Bilolisja assemblage and those of the Balkan Tardi/Epi-gravettian are clear, the definition of their direct analogies is not simple. D.Ja. TELEGIN (1982: �0) supposed that the complex of Bilolisja is related to the ‘Romanello-Azilian’ group of Romania. In effect, the geographic proximity and the tool inventory characteristics are com-mon to Bilolisja, Cuina Turcului (Păunescu, 1979) and Baile Herculane (Păunescu, 2001: 135-148). The new publication of the assemblages from the two above-mentioned Romanian sites can provide further suggestions regarding the relationships between Bilolisja and the two Iron Gates sites.

Both the Cuina Turcului layers are older than Bilolisja (12,600±120 uncal ВР [Bln-803]: 13700-12200 cal BC at 2σ; 12,050±120 uncal BP [Bln-804]: 13400-11600 cal BC at 2σ; and 10,125±200 uncal BP [Bln-802]: 10700-9200 cal BC at 2σ [Păunescu, 1979: 15; Borić, 2001: 103). They represent the last Epigravettian episodes in this territory. They yielded high number of Epigravettian tools sensu stricto: bladelets with a straight backed side, points on bladelets with an obliquely truncated base, arched backed points and even small Gravettian-like points (Păunescu, 19�0: 130-141; DINAN, 1996).

These characteristics are badly represented in the Bilolisja collection. The occurrence of lunates, isolated trapezes with retouched short side, triangles, ‘quadrangles’ and the overall similarity of the other tool types link Bilolisja and both the Cuina Turcului levels. Thus the Bilolisja industry might represent a developed aspect of the Cuina Turcului complex.

It is important to point out that none of the steppe Prichernomor’a Mesolithic investiga-tors has ever noticed the presence of trapezes with retouched short side in the Danubian Tar-digravettian assemblages. This is the reason why the occurrence of these tools at Kaghyl’nyk and Bilolisja IV was explained as caused by the migration of the Tsarynka tradition bearers from the northeast. The systematic recurrence of this tool in the Danubian Tardigravettian inventory gives an opportunity to reject these hypotheses.

- 43

2.2. tHE EARLY MESoLitHiC SitES oF tHE EAStERn PARt oF tHE REgion: tHE SoUtHERn bUH-DniEStER intERFLUViAL

Now we will turn our attention further to the east, to the steppe stripe of the southern Buh-Dniester interfluvial. No Early Mesolithic site has ever been excavated in this region. This period is represented only by surface collections, although the finds from Tsarynka and Gavryliv Jar can undoubtedly be assigned to this period. Furthermore there are a few more, highly questionable sites, which have been previously attributed to the Early Mesolithic.

tsarynkaTsarynka is located on a high (100 m above the river level) isolated terrace of the right

bank of the Ivashkova Balka, the right tributary of the Savranka River, which flows into the Southern Bug.

The archaeological collection includes only surface flint finds. The test-trenches opened in 19�5 failed to recover the cultural horizon of this site. The number of flint specimens con-sists of 3,854 pieces (table 5).

Two rounded pebbles (4 and 6 cm Ø respectively) were employed as hammer-stones. They show characteristic star-shaped hit scars along their ridges.

Flat cores with one or two platforms are well represented. They often have a signs of curation (fig. 8, nn. 31-33). The prismatic flat cores, with traces of cortex on their back surface, predominate. The sub-prismatic cores are few. The striking platforms are mainly oblique.

Artefact types number %

Chipped pebbles, core-like fragments and undetermined debris 161 4.2

Cores �3 1.9

Hammers 2 -

Flakes of rejuvenation 111 2.9

Flakes 1660 43.1

Blades 1269 32.9

Bladelets 1�3 4.5

Small chips and burin spalls 406 10.5

total 3854 100.0

Table 5 - Tsarynka: technical parameters of primary flaking (after S.P. Smol’JANINoVA, 1990).

The flake scars are usually 3-6 cm long. The flakes are more numerous than the blades. The retouched tools are both obtained on flakes (5.0%) and blades (�.1%). The Tsarynka blades are often fragmented. The retouched implements constitute 9.3 % of the total assem-blage.

44 -

Fig. 8 - Tsarynka flint types. Trapezes with retouched short side (1-11), retouched blades (12, 15, 1�), truncations (13, 14), notched blade (16), scrapers (18-26), burins (2�, 29, 30), point (28) and cores (31-33).

- 45

The scrapers represent the most numerous class. More than a quarter is on shortened blades. The scrapers on a flake are also well represented. The majority of the burins are made on the blade angle without the preparation of the burin blow platform. The burins with a retouched platform are few and atypical. There are numerous retouched blades. A small, distinctive group is represented by blades and flakes with trimming scars (adze-like tools).

tool types number %Scrapers 152 36.2

End scrapers on blades 21 -End scrapers on shortened blades 36 -End scrapers on flakes 3 -circular 3 -On flakes 68 -Nosed-scrapers 2 -On chips 3 -double 8 -Fragments of working edges 8 -

Burins 39 9.2Angle 11 -On broken blade 10 -On truncation 10 -Dihedral 5 -On flakes 2 -Double 1 -

Geometric microliths 18 4.3Trapezes 18 -

Points 18 4.3Blades with oblique truncation 12 -Borers 3 -Perforators 3 -

Combined tools 5 1.2Scrapers-burins 2 -Notched piece-point 1 -Scraper-point 1 -Notched piece-scraper 1 -

Backed blades 11 2.6Blades with marginal retouch 39 9.2Truncated blades � 1.�Notched blades 36 8.6Flakes with retouch and macrotraces of utilisation �3 1�.4Notched flakes 13 3.1Uncommon pieces 9 2.2total 420 100.0

Table 6 - Tsarynka: tool types with secondary treatment.

The main characteristic of Tsarynka is the presence of very typical trapezes with retouched short side. They are on thick, rough, irregular blades with an abrupt wide-

46 -

facetted retouch at both edges and on the short side. The latter can be straight, although usually it is wavy or notched. The trapezes are elongated and sometimes asymmetrical (fig. 8, nn. 1-11). The trapezes are often broken or show impact traces (fig. 8, nn. 2, 4, 5, 10 and 11).

The assemblage also includes oblique truncations (fig. 8, n. 13 and 14), perforators (fig. 8, nn. 1� and 28) and notched blades (fig. 8, n. 16).

The general style of tool production is quite archaic; a few authors attributed the Tsaryn-ka assemblage to the Late Palaeolithic (GRIGOR’EVA, 1982: 4�-48; KOROBKOVA, 1989: 68).

gavryliv JarThe site of Gavryliv Jar can be most probably attributed to the Early Mesolithic

(KREMER, 1960; DVORJANINOV, 19�6; STANKO, 1980: 103; SMOL’JANINOVA, 1990: 62). It was discovered on the high left bank of the Kodyma River, a right tributary of the

southern Buh, 60 km southeast of Tsarynka. 211 flint artefacts were collected from a surface of 30x50 m. The assemblage consists of 15 cores (fig. 9, nn. 14 and 15), 3 core-like frag-ments, 5 crested blades, 4 platform rejuvenation flakes, 45 flakes, 68 blades, 6 bladelets, 5 burin spalls and 36 chips and unidentifiable pieces. The tools are represented by 10 end scrapers (fig. 9, nn. 2-4 and �-11), 2 burins, 8 retouched blades (fig. 9, nn. 5 and 6), 2 notched blades, 1 trapeze with retouched short side (fig. 9, nn. 1).

2.2.1. DiscussionThe industries from Tsarynka and Gavryliv Jar have been interpreted in different ways

in the literature. Some authors, on the basis of the flaking style attributed them to the Late Palaeolithic, although important features distinguish these sites from the latest Palaeolithic assemblages of the northwestern Pontic Area.

First of all the appearance of scrapers, burins and perforators is different. More than 2/3 of the scrapers are on flakes, circular and semicircular forms included. The end scrapers are represented by short forms. The burins are not numerous and simple. The flat burin spalls are common. The microlithic tools, which are characteristic for the Epigravettian of the northern Prychornomor’a, are rare, and the micro-points absent. A particular type of microlith, the trapeze with retouched short side is typologically very distinctive and it represents the greater majority of the microliths. These trapezes are easy to distinguish from both the Late Paleo-lithic double truncations and the backed points.

The development of the Tsarynka tradition probably took place after the end of the Epigravettian complex. The Tsarynka industry might be traced back to the period of the Borschevo II sites type industry (STANKO, 1986: 22). This lithic industry is characterised by short scrapers on blades and flakes, oblique truncations, burins with flat burin spall facets and the absence of points on bladelets (BORISKOVSKIJ, 1953; EFIMENKO and BORISKOVSKIJ, 1953).

The origin of the trapezes with short, retouched side might be rooted in the abrupt re-touch, arched points, which are typical for the Borschevo II assemblage. The trapezes with short retouched side appeared simultaneously on vast areas of both Central and Eastern Europe, although their morphological traits are quite distinctive according to the differ-ent regions. The isosceles trapezes, which made their appearance at the end of the Late

- 4�

Fig. 9 - Gavryliv Jar chipped stone industry: trapeze with a retouched short side (1), end scrapers (2-4, �-11), side scrapers with marginal retouch (5, 6, 12 and 13) and cores (14 and 15).

Palaeolithic in Crimea and the Lower Dnieper region (Osokorivka), are different. They are obtained from wide, regular, thin blades, with or without retouched short side. They can be included into one single typological class together with the lunates. They are obtained with a more regular, abrupt retouch, which contrasts with that of the Tsarynka trapezes (KOLOSOV, 1964: 45).

To sum up, the cultural processes in the northwestern Prychornomor’a around the beginning of the Holocene are unique although still insufficiently studied. Nevertheless their main cultural traits can be defined, and some hypotheses were proposed in order to define the directions of future investigations.

48 -

R E F E R E n C E S

BIBIKOVA, V.I. 1978 - Fauna iz mezoliticheskih poselenij Belolesje i Ghirzhevo (nizhneee Podnestrovje) [Fauna from the Mesolithic settlements Bilolisja and Ghyrzheve (Lower Dniester region)]. In KARySHKOVSKIJ, P.O. (ed.) Arheologicheskie issledovanija Severo-zapadnogo Prichernomorja [Archaeological investigations in northwestern Prychornomor’a]: 17-29. naukova dumka, Kyiv (in Russian).

BIBIKOVA, V.I. unpubl. - Teriofauna rannemezoliticheskogo poselenija Belolesja (Severo-zapadnoe Prichernomorje) [mammalian fauna from the Early mesolithic settlement Bilolisja (northwestern Prychornomor’a)]. Unpub-lished manuscript in possession of V.N. Stanko (in Russian).

BIBIKOVA, V.I. and STARKIN, A.V. 1985 - Ostatki sajgaka pozdneplejstotsenovogo vozrasta iz raskopok stojanki Anetovka 2 (Ukraina) [The rests of the Late Pleistocene Saiga from the Anetivka 2 site excavations]. Vestnik zoologii, 3: 4�-51 (in Russian).

BINFORD, L. 1989 - debating archaeology. Academic Press, London.

Borić, D. 2001 - Mesolithic and Early Neolithic Hunters and Fishers in the Danube Gorges: An Analysis of Archaeo-zoological Data. In KERTÉSZ, R. and MAKKAy, J. (eds.) From the mesolithic to the Neolithic. Archaeolingua, 11: 101-124. Budapest.

BORISKOVSKIJ, P.I. 1953 - Paleolit Ukrainy [The Palaeolithic of Ukraine]. materialy i issledovnija po arheologii SSSR, 40 (in Russian).

BORISKOVSKIJ, P.I. 1957 - Razvedki pamjatnikov kamennogo veka mezhdu Tiraspolem i Razdel’noj [The Stone age sites surveys from Tiraspol’ to Rozdil’na]. materialy po arheologii Severnogo Prichernomorja, 1: 5-� (in Russian).

CHERNySH, O.P. 19�5 - Starodavne naselennja Podnіstrov’ja v dobu mezolіtu [The ancient population of Podni-strov’a in the mesolithic]. naukova dumka, Kyiv (in Ukrainian).

CHETRARU, N.A. 1973 - Pamjatniki jepoh paleolita i mezolita [The Palaeolithic and Mesolithic sites]. In Ar-heologicheskaja karta moldavskoj SSR. [Archaeological map of moldova SSR], 3. Stiinta, Chisinau (in Russian).

COHEN, V. 1999 - The Population of South-Russian Plain after the maximum of the second pleniglacial. Folia Quar-tenaria, �0: 363-384.

COVALENKO, S.I. and CHETRARU, N.A. 199� - Novye dannye o mezoliticheskoj stojanke Sarateny v Nizhnem Poprut’e [new data on the Mesolithic site Sarateni in the Lower Prut Region]. In CHETRARU, N. and tentiuc, i. (eds.) Vestigii arheologice din moldova: 36-59. Chisinau (in Russian).

COVALENKO, S.I. and COJ, V.B. 1999 - K voprosu o razvitii pozdnemezoliticheskih industrij v Karpato-Dnestrovskom regione [On the question of the Late Mesolithic industry development in the Carpathian-Dniester region]. Stratum plus, 1: 25�-262 (in Russian).

DANILENKO, V.N. 1969 - Neolit Ukrainy [The Neolithic of Ukraine]. Naukova dumka, Kyiv (in Russian).

DINAN, E. 1996 - Preliminary lithic analysis of the Epigravettian levels from the Iron Gates site of Cuina Turcului. mesolithic miscellany, 1� (2): 25-40.

DVORJANINOV, S.A. 19�6 - Predvaritel’nye itogi razvedok pamjatnikov kamennogo veka v Tatarbunarskom rajone Odesskoj oblasti [Preliminary results of the Stone Age sites surveys in the Tatarbunary district of Odessa region]. materialy po arheologii Severnogo Prichernomorja, 8: 154-15� (in Russian).

EFIMENKO, P.P. and BORISKOVSKIJ, P.I. 1953 - Paleoliticheskaja stojanka Borshevo II [The Palaeolithic site Borschevo II]. materialy i issledovnija po arheologii SSSR, 39: �9-92 (in Russian).

GORELIK, A.F. 2001 - Pamjatniki Rogaliksko-Peredel’skogo rajnona. Problemy final’nogo paleolita Jugo-Vostochnoj Evropy [The sites of Rogalyk-Peredil’s’ke region. The issues of the Final Palaeolithic in southeastern Europe]. RIO «LIVD», Kyiv-Lugansk (in Russian).

- 49

GRIGOR’EVA, G.V. 1961 - novye dannye o tardenuazskoj stojanke Grebenniki v nizhnem Podnestrov’e [new data on the Tardenoisian site Grebenyky in the Lower Dniester Region]. zapiski odesskogo arheologicheskogo obschestva, 1: 226-230 (in Russian).

GRIGOR’EVA, G.V. 1982 - Pozdnepaleoliticheskie pamjatniki s geometricheskimi mikrolitami na Juge Russkoj ravniny [The Late Palaeolithic sites with geometric microliths in the south Russian Plain]. Tezisy 11 kon-gressa INQUA, 3: 4�-48 (in Russian).

GRIGOR’EVA, G.V. 1992 - Pozdnepaleoliticheskie pamjatniki s geometricheskimi mikrolitami v Severnom Prichernomor’e [The Late Palaeolithic sites with geometric microliths in northern Prychornomor’a]. In Pіzn’opaleolіtychnі pam’jatki Pіvnіchnogo Prychornomor’a [The late Palaeolithic sites in nor-thern Prychornomor’a]: 11-20. Kherson (in Russian).

JANEVICH, A.A. 1990 - K probleme zapadnyh geneticheskih svjazej rannego mezolita gornogo Kryma [On the issue of the western genetic connections of the Early Mesolithic of mountainous Crimea]. In Problemy pervobyt-noj arheologii Severnogo Prichernomor’ja (K stoletiju osnovanija Khersonskogo muzeja drevnostej). Tezisy dokladov jubilejnoj konferencii: 2�-28. Kherson (in Russian).

KOLOSOV JU.G. 1964 - Nekotorye pozdnepaleoliticheskie stojanki porozhistoj chasti dnepra [Selected late Palaeoli-thic sites from the dnieper rapids region]. Svod arheologicheskih istochnikov, A1-5 (in Russian).

KOROBKOVA, G.F. 1957 - Tardenuazskaja stojanka Grebenniki v nizhnem Podnestrov’e [Tardenoisien site Grebe-nyky in the Lower Dnieper region]. Kratkie soobshenija Instituta Istorii material’noj Kul’tury, 6�: 59-62 (in Russian).

KOROBKOVA, G.F. 1989 - Predposylki slozhenija proizvodjawego hozjajstva v Severo-Zapadnom Prichernomor’e [The prerequisites of the food production appearance in the northwestern Prychornomor’a]. In BEREZANS-KAJA, S.S. (ed.) Pervobytnaja arheologija [Archaeology of the Prehistory]: 63-86. naukova dumka, Kyiv (in Russian).

KOROBKOVA, G.F. 1989a - Tehnologija i funkcii orudij v uslovijah regional’noj adaptacii (na primere verhnego paleolita-mezolita Severo-Zapadnogo Prichernomor’ja) [Technology and functions of tools under local ad-aptative conditions]. In Problemy kul’turnoj adaptacii v jepohu verhnego paleolita [Problems of the cultural adaptation in the Upper Palaeolithic period]: 48-52. nauka, Leningrad (in Russian).

koZłowski, j. 1973 - The problem of so-called Danubian Mesolithic. In koZłowski s.k. (ed.) The mesolithic in Europe. International Archaeological Symposium on the mesolithic in Europe: 314-330. Warsaw.

KRASKOVSKIJ, V.I. 19�8 - Pamjatniki paleolita i mezolita Severo-zapadnogo Prichernomor’ja [The Palaeolithic and mesolithic sites in northwestern Prychornomor’a]. Naukova dumka, Kyiv (in Russian).

KRASKOVSKIJ, V. and KREMER, A. 1959 - Mihajlіvs’ka epіpaleolіtichna stojanka [The Epipalaeolithic site Myhajliv-ka]. materialy z arheologiji Pivnichnogo Prychornomor’a, 6: 126-130 (in Ukrainian).

KRASKOVSKIJ, V.I. and STANKO, V.N. 1966 - K arheologicheskoj karte-sheme pamjatnikov kamennogo veka Nizhnego Podnestrov’ja [On the archaeological map of the Stone Age sites of the Lower Dniester region]. materialy po arheologii Severnogo Prichernomorja, 5: 235-244 (in Russian).

KREMER, A.M. 1960 - Materialy razvedok pamjatnikov kamennogo veka na r. Sarate [The materials recovered in the course or the Stone Age survey in the Sarata river valley]. zapiski odesskogo arheologicheskogo obschestva, 1: 224-225 (in Russian).

MARKEVICH, V.I. 19�4 - Bugo-dnestrovskaja kul’tura na territorii moldavii [Bug-dniester Culture in the moldova territory]. Stiinta, Chisinau (in Russian).

PASHKEVICH, G.A. 1981 - Dinamika rastitel’nogo pokrova Severo-Zapadnogo Prichernomor’ja v golocene, ego izmenenija pod vlijaniem cheloveka [The plant cover dynamics of northwestern Prychornomor’a in the Holocene and its changes under the anthropogenic influence]. In Antropogennye faktory v istorii razvitija sovremennyh jekosistem [Anthropogenic factors in the history of ecosystems development]: 74-86. nauka, Moscow (in Russian).

50 -

Păunescu, AL. 19�0 - Evoluţia uneltelor şi armelor de piatră cioplită descoperite pe teritoriul României. Biblioteca de Arheologie, XV. Editura Academiei Republiciii Socialiste România, Bucharest (in Romanian).

Păunescu, al. 1979 - Cercetările de la Cuina Turcului-Dubova (Jud. Mehedinţi). Tibiscus, 5: 11-56. Timişoara (in Romanian).

Păunescu, al. 2001 - Paleoliticul şi Mezoliticul din Spaţiul Transilvan. AGIR, Bucharest (in Romanian).

PETROUGNE, V.F. 1971 - O geologicheskoj pozicii i obrabotannom kremne mezoliticheskoj stojanki Beloles’e [On the geological position and the chipped flint of the Mesolithic site Bilolisja]. materialy po arheologii Severnogo Prichernomorja, �: 110-11� (in Russian).

SMOL’JANINOVA S.P. 1990 - Paleolit i mezolit Stepnogo Pobuzh’ja. [The Palaeolithic and mesolithic of the steppe Pobuzh’a]. Naukova dumka, Kyiv (in Russian).

STANKO, V.N. 1966 - Mezoliticheskaja stojanka Girzhevo v Odesskoj oblasti [The Mesolithic site Gyrzhyve in Odes-sa region]. Sovetskaja arheologia, 2: 96-103 (In Russian).

STANKO, V.N. 1967 - nekotorye voprosy pozdnego mezolita Severnogo Prichernomor’ja [Some issues of the Late Mesolithic of northern Prychornomor’a]. zapiski odesskogo arheologicheskogo obschestva, 2: 165-168 (in Russian).

STANKO, V.N. 1971 - Mezolit Dnestro-Dunajskogo mezhdurech’ja [The Mesolithic of Dniester-Danube interfluvial]. materialy po arheologii Severnogo Prichernomorja, 3: 93-110 (in Russian).

STANKO, V.N. 1972 - Tipy pamjatnikov i lokal’nye kul’tury v mezolite Severnogo Prichernomor’ja [The site types and local cultures of the Mesolithic of northern Prychornomor’a]. materialy i issledovnija po arheologii SSSR, 185: 252-261 (in Russian).

STANKO, V.N. 1976 - Periodizatsija pamjatnikov mezolita Severnogo Prichernomor’ja [Periodisation of the Meso-lithic sites of northern Prychornomor’a]. materialy po arheologii Severnogo Prichernomorja, 8: 15-21 (in Russian).

STANKO, V.N. 19�� - Osnovnye osobennosti i hronologija pamjatnikov mezolita stepej Severnogo Prichernomor’ja [The main peculiarities and chronology or the Mesolithic sites in the steppes of northern Prychornomor’a]. Kratkie soobshenija Instituta Arheologii AN SSSR, 149: 46-53 (in Russian).

STANKO, V.N. 1980 - Rannij mezolit stepej Severnogo Prichernomor’ja [The Early Mesolithic of the northern Prychoronomor’a]. In Pervobytnaja arheologija – poiski i nahodki [The archaeology of prehistory – research-es and finds]: 90-109. naukova dumka, Kyiv (in Russian).

STANKO, V.N. 1982 - mirnoe. Problema mezolita stepej Severnogo Prichernomor‘ja [myrne. Problem of the meso-lithic of northern Prychornomor’a steppes]. naukova dumka, Kyiv (in Russian).

STANKO, V.N. 1985 - K probleme zapadnyh svjazej mezolita Severnogo Prichernomor’ja (po materialam poselenija Beloles’e) [On the problem of the western connections of the Mesolithic of the northern Prychornomor’a (accord-ing to the Bilolisja site materials)]. In Novye materialy po arheologii Severo-zapadnogo Prichernomor’ja [New materials on the archaeology of northwestern Prychornomor’a]: 31-45. Naukova dumka, Kyiv (in Russian).

STANKO, V.N. 1986 - K probleme slozhenija grebenikovskoj kul’tury [On the problem of the Grebenyky culture gen-esis]. In Issledovanija po arheologii Severnogo Prichernomor’ja [The studies in northern Prychornomor’a archaeology]: 13-26. naukova dumka, Kyiv (in Russian).

STANKO, V.N. 1991 - Kul’turno-istoricheskij process v mezolite Severo-Zapadnogo Prichernomor’ja [The cultural historical process in the Mesolithic of northwestern Prychornomor’a]. In Severo-zapadnoe Prichernomor’e – kontaktnaja zona drevnih kul’tur [northwestern Prychornomor’a – contact zone of ancient cultures]: 5-17. Naukova dumka, Kyiv (in Russian).

STANKO, V.N. 1996 - Hozjajstvo naselenija stepej Severnogo Prichernomor’ja v mezolite [The economy of the popu-lation of the steppes of the northern Prychornomor’a in the Mesolithic]. Zapysky іstorychnogo fakul’tetu, 3: 3-14 (in Russian).

- 51

STANKO, V.N. 2006 - The Dynamics of population forming processes in the North-Western Pontic area in the Late Palaeolithic and Mesolithic. In yANKO-HOMBACH, V. (ed.) Extended Abstracts of 2nd Plenary meeting and Field Trip of Project IgCP-521 “Black Sea-mediterranean corridor during the last 30 ky: sea level change and human adaptation (2005-2009): 163-164. Astroprint, Odessa.

STANKO, V.N., DOLUHANOV, P.M., SEFERIADES, M., SMyNTyNA, E.V., PILIPENKO, G.P. and GOLOBORODOVA, E. 1999 - Mezolit juzhnoj Bessarabii [The Mesolithic of southern Bessarabia]. Zapysky іstorychnogo fakul’tetu, 8: 8-6� (in Russian).

STANKO, V.N., PETROUGNE, V.F. and MAKSIMJUK, T.I. 1981 - Pozdnemezoliticheskoe mestonahozhdenie kukrekskogo tipa na Juzhnom Buge [The late Mesolithic site of Kukrek type of the southern Bug]. In Pamjatniki drevnih kul’tur Severo-zapadnogo Prichernomor’ja [The ancient cultures’ sites of northwestern Prychornomor’a]: 5-12. Naukova dumka, Kyiv (in Russian).

STANKO, V.N. and SVEZHENCEV, JU.S. 1988 - Hronologija i periodizacija pozdnego paleolita i mezolita Sever-nogo Prichernomor’ja [Chronology and periodisation of the Late Palaeolithic and Mesolithic of northern Prychernomor’a]. Bulleten Komissii po izucheniju chenvertichnogo perioda, 5�: 116-120 (in Russian).

telegіn, d.ja. 1982 - Mezolіtichnі pamjatki Ukraini [The mesolithic sites of Ukraine]. naukova dumka, Kyiv (in Ukrainian).

telegіn, d.ja. 2002 - Іgrens’ke poselennja na Podnіprov’ji ta problema zhytlobuduvannja v mezolіtі Shіdnoji Evropy [The settlement of Igren’ in the Dnieper region and the issue of house building in the Mesolithic of Eastern Europe]. Shljah, Lugans’k (in Ukrainian).

ZAJTSEVA, G.I., TIMOFEEV, V.I., ZAGORSKAJA, I. and KOVALJUKH, N.N. 199� - Radiouglerodnye daty pamjatnikov mezolita Vostochnoj Evropy [The radiocarbon dates of the Mesolithic sites in eastern Europe]. Radiouglerod i arheologija, 2: 11�-12� (in Russian).

ZalіZnyak, l.l. 1995 - Fіnal’nyj paleolіt Ukrajiny [The Final Palaeolithic of Ukraine]. Arheologіja, 1 (in Ukrai-nian).

ZalіZnyak, l.l. 1998 - Peredіstorіja Ukrajiny X-V tis. do n.e. [Prehistory of Ukraine X-V millennia BC]. Bіblіoteka ukrajintsja, Kyiv (in Ukrainian).

ZalіZnyak, l.l. 2005 - Chornomors’kyj potop ta jogo arheologіchnі naslіdky [The Grebenyky culture as a possible consequence of the Black Sea Flood]. Arheologіja, 3: 3-12 (in Ukrainian).

ZalіZnyak, l.l. 2005a - Fіnal’nyj paleolіt і mezolіt kontynental’noji Ukrajiny. Kul’turnyj podіl ta perіodizatsіja [Final Palaeolithic and mesolithic of continental Ukraine. Cultural subdivisions and periodisation]. Shljah, Kyiv (in Ukrainian).

ZALIZNyAK, L. 2006 - Grebeniky culture of Odessa region as a possible consequence of the Black Sea Flood. In yANKO-HOMBACH, V. (ed.) Extended Abstracts of 2nd Plenary meeting and Field Trip of Project IgCP-521 “Black Sea-mediterranean corridor during the last 30 ky: sea level change and human adaptation (2005-2009): 180-181. Astroprint, Odessa.

Authors’ addresses: VLADIMIR N. STANKO, Petro Mohyla State University for the Humanities in Mykolajiv 10, 68 Desantnykiv St. – UA - 54003 MyKOLAJIVe-mail: vladimirstanko@ukr.net

DMyTRO KIOSAK, Progetto Erasmus Mundus, Facoltà di Scienze Matematiche, Fisiche e Naturali, Università degli Studi di Ferrara, Corso Porta di Mare 2 – I - 44100 FERRARAe-mail: dkiosak@ukr.net

- 53

Atti Soc. Preist. Protost. Friuli-V.G., Trieste, XVI, 2006-200� (2008): 53-�3

FEDERICO BERNARDINI∗, ANTONIO ALBERTI∗∗, GABRIELLA DEMARCHI∗∗, EMANUELA MONTAGNARI KOKELJ∗, FRANCESCO PRINCIVALLE∗∗

and AnTOn VELUŠČEK∗∗∗

An ARCHAEoMEtRiC StUDY oF tHE PREHiStoRiC PoLiSHED StonE tooLS FRoM tHE LJUbLJAniCA RiVER

(SLoVEniA)

SUMMARY – An archaeometric study of the prehistoric polished stone tools from the ljubljanica River (Slovenia). The archaeometric analysis of 31 polished stone tools from the Ljubljanica River near the site of Hočevarica is presented in this article. Amongst 20 axe blades, massive sub-rectangular specimens obtained mainly from local pyroclastic rocks are the most abundant. One small triangular piece is made from jade from northwestern Italy. Four of � shaft-holed axes are obtained from serpentinite whose probable origin is to be sought in the Hohe Tauern area (central Austria).

RiASSUnto – Studio archeometrico degli strumenti in pietra levigata provenienti dal fiume ljubljanica (Slove-nia). Questo lavoro riguarda l’analisi archeometrica di 31 strumenti in pietra levigata provenienti dal fiume Ljublja-nica presso il sito di Hočevarica. Tra le 20 lame d’ascia raccolte, gli esemplari massicci di forma subrettangolare, ottenuti da rocce locali piroclastiche, sono i più comuni. Un solo piccolo manufatto, di forma triangolare, è stato confezionato con giada proveniente dall’Italia nord occidentale. Quattro delle � asce forate sono di serpentinite, la cui origine è probabilmente da ricercare nell’area degli Alti Tauri in Austria centrale.

Keywords: Slovenia, Ljubljanica River, Neolithic/Chalcolithic, polished stones, raw material.

intRoDUCtion

A few studies have been so far conducted on the stone axes from Slovenia (BUSER, 1980, 198�; LUBšINA-TUšEK, 1993; KAVUR, 2005), while archaeometrical analyses have been carried out only in a few cases (PELOI, 1996-199�; D’AMICO et al., 2001). An inter-national research project on the greenstone axes from Friuli-Venezia Giulia, Slovenia and Croatia began in 2004 (MONTAGNARI KOKELJ et al., 2006), and a preliminary report has been ______________________________________

*∗∗  Dipartimento di Scienze dell’Antichità, Università degli Studi di Trieste, I*** Dipartimento di Scienze della Terra, Università degli Studi di Trieste, I*** Inštitut za arheologijo, Znanstvenoraziskovalnega centra SAZU, Ljubljana, SI

54 -

recently published (ALBERTI et al., 200�). The polished stone tools from the Ljubljanica River (Ljubljansko barje) have been briefly taken into consideration. The aim of this paper is to describe the typology of the tools, the results of the archaeometrical analyses, and to discuss their archaeological implications.

tHE PoLiSHED StonES PRiVAtE CoLLECtion FRoM tHE LJUbLJAniCA RiVER

The assemblage is composed of 31 polished stone tools found underwater in the bed of Ljubljanica River, close to the site Hočevarica in Ljubljansko barje (Velušček, 2004a: 53-55) (fig. 1). They are � shaft-holed axes, 20 axe blades, 2 chisels, 1 whetstone and a triangular object whose function is still undefined.

tYPoLogY

Axe bladesThe collection comprises 13 complete tools, 5 proximal and 2 distal fragments. Most of

them show common typological features: a considerable size, a squared heel, a sub-rectan-gular shape and an oval or, more often, rectangular transverse section (fig. 2, nn. 2-4 and 46; 3, nn. 51, 56, 5� and 59). The hammer n. 55 is probably obtained from a broken axe blade (fig. 3). The shape of the axes of this group does not find analogies with that of the polished stone tools from northern Italy. In this latter region rectangular axes make their appearance since the Chalcolithic, probably in connection with new handling systems (DE MARINIS, 1996: 1�6). They are always smaller than those from Ljubljanica, which, on the other hand, can be compared with some Slovenian finds (LUBšINA-TUšEK, 1993: 136: pl. 5, fig. 1�; 158: pl. 2�, fig. 4).

two small axes Two blades have peculiar shapes and very small dimensions, which contrast with the

characteristics of other tools. One has a flat triangular body with a curved, asymmetrical cutting-edge, lateral distinct sides and a sub-rectangular transverse section (fig. 2, n. �). The surfaces are finely polished except for the lateral sides, which are covered with dense ham-mering traces. The cutting-edge is sharp and does not show traces of use-wear; only a small fracture, 11 mm long, is visible. The other axe blade has a flat, trapezoidal body with a curved cutting-edge damaged by use fractures, distinct lateral sides, and a rectangular trans-verse section (fig. 3, unnumbered). The surfaces are well-polished. The first specimen finds parallels in the polished stone industries of northern Italy, which are often obtained from HP metaophiolites from the western Alps (D’AMICO et al., 2004; D’AMICO, 2005; D’AMICO and STARNINI, 2006; for Friuli-Venezia Giulia see D’AMICO et al., 199�; PESSINA and D’AMICO, 1999). The second shows an uncommon shape, which probably is due to the re-utilisation of a broken tool.

Two more tools are difficult to define. Number 15 is a carbonatic rock flake with very weathered surfaces (fig. 2). For this reason it is impossible to recognise any technological

- 55

Fig. 1 - Discovery location of the tools and archaeological sites mentioned in the text (drawing by m. Belak).

56 -

trace. It is supposed not to be a tool. Number 44 (fig. 2) has a flat sub-trapezoidal body with distinct edges; the cutting-edge is not sharp and its profile rounded.

ChiselsAre represented by only 2 tools (fig. 2, n. 45; 3, n. 54). Number 45 has a sub-rectangular

massive body with a tapering distal part, an oval transverse section and a squared heel. The cutting-edge is rectilinear and short, with use notches. The distal edge is well polished, while the other surfaces are covered with hammering traces, except for the heel, which shows a glossy surface perhaps due to its utilisation or handling. The chisel n. 54 (fig. 3) is charac-terised by a sub-trapezoidal massive body tapering from the heel to the cutting edge, 25 mm long and curved; the heel is sub-rectilinear and the transverse section flat-convex. The sur-faces are well polished. Although the chisels morphology is different from that of the other tools, their dimension and raw material recall the main group of axe blades.

Shaft-holed axesThis group comprises � tools (fig. 4). Two show a narrow, biconvex shape in frontal

view and a sub-rectangular shape in lateral one. The cutting-edge is curved, when preserved; the hole is located at the proximal edge of the axe. All the surfaces are very finely polished; polishing stripes in various directions are visible (fig. 4, nn. 9 and 12). The working process is very accurate. The fragmented axe n. 11 (fig. 4) might be a part of a similar tool, although it is somewhat larger than nn. 9 and 12. This morphology does not recall that of the shaft-holed axes from Friuli-Venezia Giulia collections (PELOI, 1996-199�: 124 and 125). A similar tool, probably made from serpentinite, was found near Töplitsch in Carinthia (PICCOTTINI, 19��: 291; VAHLKAMPF, 19�9: �-8, fig. 1; VON USLAR, 1991: 251).

The other shaft-holed axes show a considerable morphological variety. Number 1 is complete: in frontal view it shows a bi-convex shape, a flat tapering heel, a symmetric bicon-vex cutting edge and a medial perforation; in lateral view its shape is sub-rectangular with a curved cutting edge and a squared heel with blunt angles. Its transverse section is sub-rectan-gular (fig. 4). The surfaces are polished with hammering traces not completely erased by the polishing process. The tool n. 5 is a distal fragment (fig. 4) whose original shape cannot be identified. In frontal view it shows curved lateral surfaces and a symmetric biconvex cutting edge, curved in lateral view. The transverse section is sub-rectangular. All the surfaces are very finely polished, and polishing stripes are not visible at a macroscopic level. In frontal view the body of axe n. 8 (fig. 4) is sub-triangular with an irregular heel, curved lateral sur-faces and a flat symmetric cutting edge; the transverse section was probably sub-rectangular. The shapes of these shaft-holed axes do not have precise parallels in the stone artefacts from other localities. The triangular body of axe n. 8 is very common both in Friuli-Venezia Giulia (D’AMICO et al., 1996: figs. 153-155) and Slovenia (LUBšINA-TUšEK, 1993). The narrow bi-convex body with a central hole of axe n. 1 is not frequent in the axes from north-eastern Italy (D’AMICO et al., 1996: figs. 153-155).

Number 14 (fig. 4) might be an unfinished shaft-holed axe, but this interpretation is uncertain. It is a triangular implement with a small shaft hole near its narrow edge; the trans-verse section is sub-rectangular. All the surfaces are covered with fine and dense hammering traces and are unpolished. The largest (cutting) edge shows a rounded profile.

- 5�

3 cm

Fig. 2 - Axe blades and chisels from the Ljubljanica River (drawings by T. Korošec).

58 -

3 cm

Fig. 3 - Axe blades and chisels from the Ljubljanica River (drawings by T. Korošec).

other toolsThe collection comprises a pierced whetstone with a small bi-pyramidal hole (fig. 5, n.

13). Another specimen has a triangular shape, with small notches in the central part of the edges (fig. 5, n. 60). It might be a weight.

- 59

3 cm

Fig. 4 - Shaft-holed axes from the Ljubljanica River (drawings by T. Korošec).

60 -

Fig. 5 - Triangular object and whetstone from the Ljubljanica River (drawings by T. Korošec).

RAw MAtERiALS AnD PoSSibLE PRoVEnAnCES

Analytical methodsAll the 31 tools have been studied by non-destructive stereo-microscopic method. Fur-

thermore, a test with 10% solution of hydrochloric acid has been performed on 3 tools. 21 specimens have been analysed also through optical polarized-light microscopy (OM) and X-ray diffraction (XRD) to define their mineralogical and petrographic features. The surface of axes number � and 9 has been studied by XRD without taking any powder: this method is adequate if the rocks are homogeneous and fine-grained. The identification of the tools analysed only at the stereomicroscope has been possible by comparison with the analysed tools.

ResultsThe lithological composition of the Ljubljanica collection and the analytical methods

are summarised in table 1 below.The small axe blade n. � is from jade, a raw material available in northwestern Italy,

as several studies have demonstrated (D’AMICO et al., 2004; D’AMICO, 2005; D’AMICO and STARNINI, 2006; PÉTREQUIN et al., 2005; 200�).

Most of the polished stone tools (19 out of 31) were obtained from tuffitic and pyroclas-tic rocks. The petrographic features (see table 2) show some relationships. Close similarities have been observed between axes 48 and 49, 52 and 60, 54 and 55. This lithology sensu latu, which is present in some 2% of the Slovenian territory, outcrops in the northern part of the country between Kranj and Celje (KOMAC, 2005: fig. 2). The pyroclastic rocks occur in the middle Triassic formations. According to some authors they are associated with spilite and keratophire lithologies (Mioč and Žnidarčič, 1983; PREMRU, 1983; DOZET, 2003: 6�5-6��), or in Tertiary deposits known as Smerkovac Tertiary (HANFLAND et al., 2004). Axe n. 5� is ob-tained from a spilite with a hyaloophitic texture. Sericitized 1 to 2 mm-sized plagioclase laths are set in an altered groundmass of palagonite full of small opaque specks, with a few highly oxidized mafic minerals (probably pyroxenes). Axe n. 59 is from an altered, pervasively oxi-dized keratophyre with a porphyritic texture (porphyricity index around 10) with 1 to 2 mm-sized feldspar phenocrysts masked by sericite aggregates, set in a fine-grained heterogeneous,

- 61

Sample typology Rock name Provenance Analytical methods

1 shaft hole axe carbonatic rock neighbouring areas stereomicroscopy; acid test2 axe blade pyroclastic rock neighbouring areas (northern Slovenia) stereomicroscopy3 axe blade fine tuff neighbouring areas (northern Slovenia) OM; XRD4 axe blade pyroclastic rock neighbouring areas (northern Slovenia) stereomicroscopy5 shaft hole axe fine tuff neighbouring areas (northern Slovenia) OM; XRD6 axe blade ash tuff neighbouring areas (northern Slovenia) OM; XRD� axe blade jade north-western Italy XRD on surface8 shaft hole axe serpentinite Hohe Tauern (Austria)? OM; XRD9 shaft hole axe serpentinite Hohe Tauern (Austria)? XRD on surface11 shaft hole axe serpentinite Hohe Tauern (Austria)? OM; XRD12 shaft hole axe serpentinite Hohe Tauern (Austria)? OM; XRD13 whetstone tuffaceous mudstone neighbouring areas (northern Slovenia) OM; XRD14 shaft hole axe ? carbonatic rock neighbouring areas stereomicroscopy; acid test15 axe blade ? carbonatic rock neighbouring areas stereomicroscopy; acid test44 axe blade ? ash tuff neighbouring areas (northern Slovenia) OM; XRD45 chisel axe pyroclastic rock neighbouring areas (northern Slovenia) stereomicroscopy46 axe blade pyroclastic rock neighbouring areas (northern Slovenia) stereomicroscopy4� axe blade serpentinite Hohe Tauern (Austria)? OM; XRD48 axe blade tuffaceous sandstone neighbouring areas (northern Slovenia) OM; XRD49 axe blade coarse crystal–lithic tuff neighbouring areas (northern Slovenia) OM; XRD50 axe blade tuffaceous sandstone neighbouring areas (northern Slovenia) OM; XRD51 axe blade pyroclastic rock neighbouring areas (northern Slovenia) stereomicroscopy52 axe blade coarse crystal tuff neighbouring areas (northern Slovenia) OM; XRD53 axe blade tuffaceous mudstone neighbouring areas (northern Slovenia) OM; XRD54 chisel axe coarse tuff neighbouring areas (northern Slovenia) OM; XRD55 hammer coarse tuff neighbouring areas (northern Slovenia) OM; XRD56 axe blade tuffaceous sandstone neighbouring areas (northern Slovenia) OM; XRD5� axe blade altered spilite neighbouring areas (northern Slovenia) OM; XRD59 axe blade keratophyre neighbouring areas (northern Slovenia) OM; XRD60 weight ? tuffaceous sandstone neighbouring areas (northern Slovenia) OM; XRD(unnum-bered) axe blade serpentinite Hohe Tauern (Austria)? OM; XRD

Table 1 - Lithological composition and possible sources of the Ljubljanica River tools.

devitrified and altered, blotched groundmass full of Fe-hydroxides in irregular streaks and spots, with few opaque specks arranged along a barely distinguishable flowage banding. It is noteworthy that all these rocks are also found in the alluvial deposits of the Sava River close to Ljubljansko barje. Some 20 pebbles were collected from the Sava River near Ljubljana and analysed through optical microscopy, showing similarities with the pyroclastic rocks used for the axes. Therefore, also the Sava alluvial deposits might represent raw material sources for the Ljubljanica axes made from pyroclastic rocks, spilites, and keratophyres.

Regarding the serpentinites, a comparative procedure has been adopted to identify their provenance area, following the approach used by other scholars (MAJEROWICZ et al., 2000; SKOCZyLAS et al., 2000).

First, the petrographic features of the Ljubljanica serpentinite tools have been compared with the serpentinite implements from the archaeological sites in northeastern Italy (D’AMICO et al., 199�: 419; 2001) and Istria (Croatia). They all are composed of fine to medium-grained antigorite, various amounts of magnetite in irregular aggregates or local nets, and frequent clinopyroxenes relics variably replaced by tremolite and antigorite.

Second the serpentinite geological samples collected from the outcrops in northeastern Slovenia (Pohorje area; VRABEC et al., 200�), Burgenland (eastern Austria, close to the border

62 - TEXTURAL AND MINERALOGICAL FEATURES

ClastsSample Rock name Composition

%Grain size

(in mm)Composition

Matrix

3 fine tuff rhyolitic 10-15 0.1-0.5 alk feld (partially caolinized) > qz > saussuritized pl siliceous, vitric to partially devitrified fine and unsorted ash

5 fine tuff rhyolitic 5-10 0.1-0.2 glass shards >> qz > alk feld, minor oxidized biot siliceous devitrified hydrated (smectitic) and argillified fine ash

6 ash tuff trachytic to rhyolitic / / /siliceous, vitric strongly devitrified and smectitic fine ash

13 tuffaceous mudstone andesitic, Fe-rich 5-10< 0.3 coarse-ash and very fine sand- grains

argillified feld > biot > qz, minor muscovite/sericite flakes mixed siliceous, argillaceous to ferrogineous with abundant opaque specks fine ash

44 ash tuff rhyo-dacitic 10-15 0.1-0.3 feld > qzsiliceous, devitrified and altered (argillified and smectitic) fine ash

48tuffaceous sandstone

dacitic �0-80average: 0.2-0.3; max size: 0.5

qz > albitic pl > porphyritic rocks with felsitic groundmass > chalcedony > alk feld > oxidized opaques

mixed siliceous and argillaceous, devitrified and smectitic grain-supported fine ash

49coarse crystal- to lithic tuff

rhyolitic 30-40average: 1-2; max size: 3

qz > alk feld > saussuritized pl >> subordinate felsitic rocks (also with micrographic texture), oxidized biot, altered (argillic to chloritic and smectitic) lithic fragments

siliceous fine ash, partially devitrified with eutaxitic flowage marks; few opaque specks

50tuffaceous sandstone

trachytic 50-60 1-3 k-feld > fine grained felsitic rocks > albitic pl, minor qz and opaque specks

mixed siliceous to argillic and chloritic (smectitic), grain supported fine ash

52 coarse crystal tuff rhyolitic 20-30average: 0.5-1; max size: 3

qz > altered pl > alk feld >> subordinate felsitic rocks > oxidized biot, devitrified glass shards

siliceous fine ash, devitrified with faint eutaxitic flowage marks

53 tuffaceous mudstone trachytic 10 01-0.2 pl (sericitized and saussuritized), minor qz

mixed siliceous (faintly devitrified) and argillaceous very fine, unsorted ash with abundant carbonaceous matter in long threads and irregular specks

54 coarse tuff trachy-andesitic 30-40average: 0.5; max size: 1.5

pl (sericitized and saussuritized) >> qz, alk feld, opacized or strongly oxidized mafic minerals, flattened pumices (almost undistinguishable from matrix), opaque specks

siliceous devitrified fine ash, with flow banding showing faint eutaxitic character

55 coarse tuff trachytic to andesitic

40-50 poor sorting; average: 0.1-0.3; max size 1.5

pl (sericitized and saussuritized) >> phyllites, mafic minerals (chloritized or strongly altered), qz, fine-grained rocks, opaque specks

siliceous fine ash with carbonaceous matter

56tuffaceous sandstone

dacitic to rhyo-dacitic

�0-80 poor sorting; max size: 1

altered and turbid pl > qz > K-feld >> albitic pl, volcanic rocks, fragments of shales, oxidized pyrite

siliceous devitrified, but heterogeneous, altered (sericitized and argillified) unsorted fine ash, with clay and carbonaceous matter

60tuffaceous sandstone

dacitic �0-80average: 0.3-0.4; max size: 0.�

qz > albitic pl > porphyritic rocks with felsitic groundmass >> phyllitic rocks, chalcedony, alk feld, chloritized biot, muscovite, oxidized opaques

mixed siliceous and argillaceous, devitrified and smectitic grain-supported fine ash

Table 2 - Petrographic features of tuffitic and pyroclastic artefacts. A

bbreviations: qz, quartz; pl, plagioclase; alk feld, alkali feldspar; feld, feldspar; biot, biotite.

- 63

with Hungary; ERVEN, 19�2; KOLLER, 1985), Styria and the Hohe Tauern (central Austria; MELCHER et al., 2002; MELCHER and MEISEL, 2004) have been considered. Both the minera-logical and petrographic features of the Ljubljanica serpentinite axes are markedly differ-ent from those sampled from the outcrops of the Slovenska Bistrica Ultramafic Complex (SBUC), in the southeastern region of the Pohorje Mountains. The Metaultrabasites of this district constitute a small body some 5x1 km in size composed of serpentinized harzburgites, with local lenses or pods of garnet lherzolites (JANAK et al., 2004: 2; 2006: 20-21; SASSI et al., 2004: 23�; VRABEC et al., 200�). Thus we can exclude the Pohorje rocks as a source area. The preliminary study of the serpentinites collected from other localities in the Eastern Alps, and the bibliographic data, show that the Ljubljanica tools are similar to the rocks that outcrop in the Hohe Tauern, in particular to some serpentinites from the Blauspitze Mountains near Kals in the Matrei Zone (eastern Tyrol), and from the upper Möll Valley, south of Heiligenblut (Carinthia). The metaultrabasites from the Matrei Zone are transformed into antigorite ser-pentinites characterised by interpenetrating and mesh textures. The primary cpx is replaced by aggregates of tremolite, chlorite and carbonate; metamorphic diopside has also been ob-served (MELCHER et al., 2002: 98). Therefore the latter localities can be considered possible candidates as source areas for the Ljubljanica artefacts.

DiSCUSSion AnD ConCLUSionS

The polished stone implements from the Ljubljanica River cannot be undoubtedly referred to one archaeological context. However, the morphological and lithological homogeneity of some shaft-holed axes, and of the main group of sub-rectangular axe blades suggests that the tools have not been transported by water from a distant place. It seems likely to relate some of the specimens to Hočevarica pile dwelling, located very close to their finding place.

In spite of the contextual problems, the collection indicates a wide exploitation of local raw materials and only a sporadic presence of long-distance, imported tools. Only one jade axe of Italian provenance is present, while the most common sub-rectangular axe blades are mainly made from pyroclastic rocks from central-northern Slovenia.

Table 3 below summarizes all the discoveries of HP metaophiolite tools - mainly axe blades - from Friuli-Venezia Giulia, Slovenia and Croatia.

The HP metaophiolite axe distribution is incomplete; nevertheless it allows some pre-liminary considerations. In the Neolithic sites of northern Italy, the HP metaophiolite tools represent some 90% of the total assemblages, while in the Alpine valleys and the northeastern regions their percentage decreases to some 60-�0% (D’AMICO et al., 2004; D’AMICO, 2005: 236; D’AMICO and STARNINI, 2006: 264). A first analysis of the blades from the Trieste Karst shows the presence of HP metaophiolites axes, although the utilization of other raw materials and the presence of other tool types, as for instance chisel axes, are also documented. These are only general and preliminary indications, which will be tested in future studies. The present-day literature lists some 20 axes mainly from caves or rock-shelters (MONTAGNARI KOKELJ, 2001).

The data collected from the northeastern Adriatic coastal regions show a widespread, although not quantitatively significant presence of HP metaophiolite tools. They seem to con-

64 -

firm the existence of relationships between northern Italy and the eastern Adriatic coast dur-ing the Neolithic. The archaeological data so far available suggest that the HP metaophiolite ground tools reached the northeastern Adriatic coast for the first time during the development of the Danilo/Vlaška Culture. It is probable that polished tools of Italian origin reached the Croatian coastline in the following period, as suggested by a small jade chisel from Loznati (Cres Island), which is a typical Square Mouthed Pottery Culture implement (ALBERTI et al., 200�).

The relationships between northern Italy and central Slovenia, only 100 km from the Friuli plain, look quite different. At present, only 4 small axe blades made from HP metaophi-

Site name Site typen. of tools

Valer (PN, Friuli-Venezia Giulia) open-air settlement 1

Cormons (GO, Friuli-Venezia Giulia) surface find 1

Marano (UD, Friuli-Venezia Giulia) surface find 1

Muzzana del Turgnano (UD, Friuli-Venezia Giulia) surface find 22

Pavia di Udine (UD, Friuli-Venezia Giulia) open-air settlement 1

Piancada di Palazzolo dello Stella (UD, Friuli-Venezia Giulia) surface find 5

Precenicco (UD, Friuli-Venezia Giulia) surface find 5

Sammardenchia (UD, Friuli-Venezia Giulia) open-air settlement 1��

Grotta Aurisina (TS, Friuli-Venezia Giulia) cave site 1

Grotta delle Gallerie (TS, Friuli-Venezia Giulia) cave site 1

Grotta delle Porte di Ferro (TS, Friuli-Venezia Giulia) cave site 1

Novigrad (Istrian peninsula, Croatia) surface find 2

Losnati (island of Cres, Croatia) surface find 2

Velj Losjnj (island of Losjnj, Croatia) surface find 1

Vrbnik (island of Krk, Croatia) surface find 2

Island of Krk (Croatia) surface find ? 5

Ražanac (Zadar, Croatia) open-air settlement 2

Split (Croatia) surface find ? 1

Markova cave (island of Hvar, Croatia) cave site 1

Nakovana (Croatia) surface find ? 1

Island of Korčula (Croatia) surface find ? 3

Staro Čiče (Croatia) open-air settlement 1

Ljublianica River (Ljubljansko barje, Slovenia) surface find 1

Stare gmajne ( Ljubljansko barje, Slovenia) pile dwelling settlement 1

Deschmann pile-dwellings (Ljubljansko barje, Slovenia) pile dwelling settlement 2

Table 3 - Polished stone tools made from HP metaophiolites from Friuli-Venezia Giulia, Slovenia and Croatia (only scientific identifications are listed except for the data from Petrić, 1995 and PESSINA et al., 2006, based on macroscopic observation). Data sources: Valer (D’AMICO and STARNINI, 2006); Cormons (PESSINA et al., 2006; PÉTREQUIN, pers. comm.); Marano, Muzzana del Turgnano, Pavia di Udine, Piancada di Palazzolo dello Stella, Pre-cenicco (PESSINA et al., 2006); Sammardenchia (D’AMICO et al., 199�; PESSINA and D’AMICO, 1999); Grotta Aurisina, Grotta delle Gallerie, Grotta delle Porte di Ferro (unpublished); Novigrad, Loznati, Veli Lošinj, Vrbnik (ALBERTI et al., 2007); Island of Krk, Ražanac, Split, Markova spilja, nakovana, Island of Korčula (Petrić, 1995); Staro Čiče (Burić, 2000); Ljubljanica River (ALBERTI et al., 200�); Stare gmajne (unpublished); Deschmann pile-dwellings (PELOI, 1996-199�; D’AMICO, 2005).

- 65

olites are known in Slovenia, 2 of which from the Deschmann pile-dwellings (KOROšEC and KOROšEC, 1969: pl. �6, figs. 11, 15; PELOI, 1996-199�: 304; D’AMICO, 2005: 24�), 1 from Stare gmajne (Velušček, 2004b: ��) and 1 from the Ljubljanica River (fig. 1, n. �). The number of HP metaophiolite axe blades in Slovenia might increase if the polished stone tools from Čatež-Sredno polje will be analysed; among the stone axes there are a few small blades made from imported raw material (KAVUR, 2005: 131-144). The axe from Stare gmajne, dated to the second half of the 4th millennium cal BC (Vera-2561: 4530±35 uncal BP, 3360-3090 cal BC at 2σ: see also Velušček, 2004b: ��), provides us with the only reliable chronological information.

Regarding the shaft-holed axes, those in the Ljubljanica collection are mostly made from serpentinites. From a lithological point of view these specimens show some affinities with the shaft-holed axes from Friuli-Venezia Giulia and Istria. Shaft-holed axes made from serpentinite and metaultramafite are very numerous in Friuli-Venezia Giulia and Ljubljan-sko barje (fig. 6), while meta-amphibolic gabbro is the commonest lithology in the Istrian Peninsula (D’AMICO et al., 2001; ALBERTI et al., 200�). The archaeometric data currently available show that the serpentinite axes in northern and eastern Croatia are not as frequent as the tools obtained from other metamorphic or magmatic rocks (BALEN et al., 2002; šIMEK et al., 2002).

Serpentinite shaft-holed axes come from sites attributed to a long period spanning from the Neolithic (for instance, the shaft-holed axe SAM306 from layer 4, structure 153 at Sam-mardenchia: PESSINA and D’AMICO, 1999: 56, fig. 29) to late prehistory (for instance, some shaft-holed axes from the Deschmann pile-dwellings: KOROšEC and KOROšEC, 1969; PELOI, 1996-199�).

The preliminary petrographic studies of the serpentinite outcrops in the Eastern Alps car-ried out by the present authors suggest that the most probable source areas of the Ljubljanica serpentinite artefacts lie in the Hohe Tauern zone of central Austria and their related second-ary deposits. Besides the petrographic evidence, this hypothesis is also supported by the fol-lowing archaeological data: a) the distribution of the shaft-holed axes in the Caput Adriae: the percentage of serpentinite shaft-holed axes is very high (almost 60%) in Friuli-Venezia Giulia and Ljubljansko barje, while it decreases to some 20% towards the south in the Istria peninsula (D’AMICO et al., 2001; ALBERTI et al., 200�; fig. 6); b) the relationships between Hočevarica and southern Austria (Carinthia and southern Styria) are demonstrated by the common presence of ceramics attributed to the HKBV horizon (Velušček, 2004a: 258-259); c) 2 shaft hole axes and 18 cores resulting from piercing, come from Kanzianiberg, very close to the Hohe Tauern serpentinite primary occurrences; suggesting a production of axes for external exchanges (PEDROTTI 1990: 221); d) the axes n. 9 and n. 12 are very similar to a serpentinite specimen found near Töplitsch in Carinthia (PICCOTTINI, 19��: 291; VAHLKAMPF, 19�9: �-8, fig. 1; VON USLAR, 1991: 251).

To conclude: it can be stated that the lithological assemblage of the Ljubljanica col-lection is mainly from local or north Slovenian sources (pyroclastic and tuffitic rocks some 60%, carbonatic rocks around 10% and a small percentage of spilitic/keratophyric rocks), with a rather high percentage (some 20%) of serpentinite implements, whose source is prob-ably to be sought in the Hohe Tauern area of central Austria. The collection includes only one jade axe. Therefore this assemblage seems very different from the northern Italian polished

66 -

Fig. 6 - Serpentinite shaft-holed axes in the Caput Adriae (only scientific identifications are considered). 1: S. Tomè (PN, Friuli-Venezia Giulia), n. AQ9514; 2: Meduno, (PN, Friuli-Venezia Giulia), n. AQ25060�; 3: Gradisca di Provesano (PN, Friuli-Venezia Giulia), n. AQ221855; 4: S. Eliseo di Caporiacco (UD, Friuli-Venezia Giulia), n. AQ2230�8; 5-22: Sammardenchia (UD, Friuli-Venezia Giulia), nn. AQ225152, AQ223082; AQ2251�0, SAM6, SAM82/1, SAM90, SAM143, SAM148, SAM211-SAM213, SAM216, SAM301, SAM306, SAM315; Pozzuolo del Friuli (UD, Friuli-Venezia Giulia) nn. AQ220591, AQ116005, AQ223080; 23: Pavia di Udine (UD, Friuli-Venezia Giulia), n. AQ225151; 24: Castions di Strada (UD, Friuli-Venezia Giulia), n. 232515; 25: S. Stefano Aquileiese (UD, Friuli-Venezia Giulia), n. AQ455�15; 26: Novacco (UD, Friuli-Venezia Giulia), n. AQ331243; 2�-28: Mossa (GO, Friuli-Venezia Giulia), n. 1325, no number; 29: Nova Gorica (western Slovenia), no number; 30: Opicina (TS, Friuli-Venezia Giulia), no number; 31: Montedoro (TS, Friuli-Venezia Giulia), n. TS2280; 32: Grotta Sottomonte (TS, Friuli-Venezia Giulia), n. TS25�85; 33: S. Ivan pod Sterne (Istria, Croatia), n. P-194; 34: Sandaya (Istrian peninsula, Croatia), P-15059; 35-36: Istria (Croatia), nn. P-12, P-14; 3�-40: Ljubljanica (Ljubljansko barje, Slovenia), nn. 8-9, 11-12; 41-50: Deschmann pile-dwellings (Ljubljansko barje, Slovenia), nn. B45, B46, B48, B50, B52, B54, B56, B59, B60, B62. Data sources: Castions di Strada, Grotta Sottomonte, Meduno, Montedoro, Pavia di Udine, Pozzuolo del Friuli nn. AQ220591, AQ116005, Sammardenchia n. AQ225152, S. Eliseo di Caporiacco, S. Stefano Aquileiese, S. Tomè (D’AMICO et al., 1996); Sammardenchia all the SAM samples (D’AMICO et al., 199�; PESSINA and D’AMICO, 1999); Gradisca di Provesano, Novacco, Mossa, Pozzuolo del Friuli n. AQ223080, Sammardenchia nn. AQ223082; AQ2251�0 (D’AMICO unpublished data); Deschmann pile dwellings (PELOI, 1996-199�); Istrian Peninsula, Nova Gorica, Opicina, Sandaya, S. Ivan pod Sterne (ALBERTI et al., 200�).

- 6�

tool collections, which are largely dominated by HP metaophiolites of northwestern Italian sources (D’AMICO et al., 2004).

CAtALogUE

n. 1: shaft-holed axe (fig. 4, n. 1). In frontal view it shows a bi-convex shape, a flat tapering heel, a symmetric biconvex cutting

edge and a medial perforation; in lateral view its shape is sub-rectangular with a curved cutting edge and a squared heel with blunt angles. Its transverse section is sub-rectangular. The surfaces are polished with hammering traces not completely eliminated by the polishing process. Length: 98 mm; width: 41 mm; thickness: 32 mm.

n. 2: axe blade (fig. 2, n. 2). It has a sub-rectangular massive body with a narrowing squared heel; an inclined cutting edge

with small use (?) fractures and a sub-rectangular transverse section. All the surfaces are well polished. Length: 131 mm; width: 58 mm; thickness: 31 mm.

n. 3: axe blade (fig. 2, n. 3). It has a sub-trapezoidal body with a rectilinear heel, scarcely distinct sides, an inclined, curved

cutting edge and a sub-rectangular transverse section. The surfaces are polished but in same ar-eas there are small concave roughnesses, not eliminated by polishing. Length: 134 mm; width: 49 mm; thickness: 25 mm.

n. 4: axe blade (fig. 2, n. 4). It has a massive sub-rectangular body with a rounded heel, convex sides, and a curved cutting

edge with small use (?) fractures. Only the cutting edge is well polished, while the other surfaces are covered with hammering traces. Length: 134 mm; width: 5� mm; thickness: 30 mm.

n. 5: shaft-holed axe, distal fragment (fig. 4, n. 5). Its original shape is impossible to reconstruct. In frontal view it shows curved lateral surfaces

and a symmetric biconvex cutting edge, curved in the lateral view. The transverse section is sub-rectangular. All the surfaces are very finely polished and there are no polishing stripes visible at macroscopic level. Length: (�8) mm; width: 39 mm; thickness: 38 mm.

n. 6: axe blade, proximal fragment (fig. 2, n. 6). The body shape was probably sub-rectangular. The sides are convex, the heel is a bit rounded, and

the transverse section is oval. The surfaces are polished, with hammering traces still visible on the sides. Length: (60) mm; width: 50 mm; thickness: 28 mm.

n. 7: axe blade (fig. 2, n. �). It has a flat, triangular body with a curved asymmetrical cutting edge, distinct sides and a sub-

rectangular transverse section. The surfaces are finely polished except for the lateral sides covered with dense hammering traces. The cutting edge is sharp and does not show wear traces; only a small fracture, 11 mm long, is visible. Length: 81 mm; width: 41 mm; thickness: 1� mm.

n. 8: shaft-holed axe, fragment (fig. 4, n. 8). In frontal view the body is sub-triangular with an irregular heel, curved lateral surfaces and a flat

symmetric cutting edge; the transverse section was probably sub-rectangular. The preserved sur-faces show a medium-fine polishing with macroscopic multidirectional smoothing stripes except near the cutting edge, where they are transversal; hammering traces are absent. It is not possible

68 -

to recognise technological traces on the heel because of thick use traces. Length: �2 mm; width: 44 mm; thickness: (24) mm.

n. 9: shaft-holed axe (fig. 4, n. 9). It has a narrow biconvex shape in frontal view and a sub-rectangular shape in the lateral one. The

cutting edge is curved; the hole is located in the proximal part of the axe. All the surfaces are very finely polished and polishing stripes in various directions are visible. Length: 125 mm; width: 22 mm; thickness: 56 mm.

n. 11: proximal fragment of a serpentinite shaft-holed axe (fig. 4, n. 11). It is too small to reconstruct its original shape. It was probably similar to that of axes 9 and 12.

The preserved surfaces are well polished. Length: (60) mm; width: 34 mm; thickness: (2�) mm.

n. 12: shaft-holed axe, distal fragment (fig. 4, n. 12). It has a narrow biconvex shape in frontal view, and a sub-rectangular shape in the lateral one.

The cutting edge was probably curved; the hole is located in the proximal part of the axe. All the surfaces are very finely polished and polishing stripes in various directions are visible. Length: (�8) mm; width: 2� mm; thickness: 55 mm.

n. 13: whetstone (?), fragment (fig. 5, n. 13). It has a bi-pyramidal hole and its body has a rectangular transverse section and distinct sides.

Its frontal surfaces are very polished while the sides are quite rough. Length: (60) mm; width: 3� mm; thickness: 21 mm.

n. 14: unfinished shaft-holed axe (?) made from carbonatic rock (fig. 4, n. 14). It is of triangular shape with a small shaft-holed located near its narrow extremity; the transverse

section is sub-rectangular. All the surfaces are covered with fine and dense hammering traces and are not polished. The largest extremity (in the position of the cutting edge) shows a rounded profile. Length: 141 mm; width: 35 mm; thickness: 69 mm.

n. 15: axe blade (?) (fig. 2, n. 15). It shows an irregular shape with a rectilinear heel and an inclined, curved cutting edge. The sur-

faces might have been weathered by dissolution processes. Length: �3 mm; width: 42 mm; thick-ness: 22 mm.

n. 44: unfinished axe blade (?) (fig. 2, n. 44). It has a flattened sub-trapezoidal body with distinct sides, an inclined curved cutting edge and a

curved hell. Length: 103 mm; width: 40 mm; thickness: 12 mm.

n. 45: chisel axe (fig. 2, n. 45). It has a massive sub-rectangular body with a squared heel, a narrowing distal part with a short and

rectilinear cutting edge with use fractures and an oval transverse section. The cutting edge is well polished while the axe body is covered with an accurate hammering. Also the heel is polished, perhaps for reasons connected to the handling or to the use of the tool. Length: 139 mm; width: 58 mm; thickness: 35 mm.

n. 46: axe blade (fig. 2, n. 46). It has a massive rectangular body with a squared heel, sub-rectangular transverse section and

an inclined curved cutting edge with use fractures. The sides are distinct and all the surfaces are polished, particularly the area near the cutting edge. Length: 112 mm; width: 50 mm; thickness: 28 mm.

- 69

n. 47: axe blade, distal fragment (fig. 2, n. 4�). The original shape was probably sub-rectangular; the cutting edge is curved and the transverse

section is planar-convex. The surfaces are well polished in the distal parte near the cutting edge, while in the middle part of the body hammering traces are visible. Length: (56) mm; width: 55 mm; thickness: 22 mm.

n. 48: axe blade, proximal fragment (fig. 2, n. 48). It shows a massive body, which probably was sub-rectangular; the transverse section is oval and

the heel is squared with blunt angles. The surfaces are polished with hammering traces. Length: (100) mm; width: 59 mm; thickness: 3� mm.

n. 49: axe blade, proximal fragment (fig. 2: n. 49). The body is massive with distinct sides and a sub-rectangular transverse section; the shape prob-

ably was sub-rectangular. The surfaces are weathered and it is difficult to identify technological traces. However it seems that it was well polished, as indicated by some small preserved areas. Length: (85) mm; width: 62 mm; thickness: 31 mm.

n. 50: axe blade, proximal fragment (fig. 2, n. 50). The body is massive with distinct sides, a rectilinear heel and a sub-rectangular transverse section.

The shape was probably sub-rectangular or sub-trapezoidal. The surfaces are very weathered. Length: (�4) mm; width: 61 mm; thickness: 25 mm.

n. 51: axe blade (fig. 3, n. 51). It shows a massive rectangular body with a rectilinear heel with blunt angles, convex edges, a

bit curved cutting edge and an oval transverse section. The surfaces are well polished. Length: 101 mm; width: 44 mm; thickness; 29 mm.

n. 52: axe blade, proximal fragment (fig. 3, n. 52). Its body was probably sub-rectangular. It has an oval transverse section, distinct edges and a

slightly curved heel. The width of the body rises from the heel to the distal part. The surfaces are well polished, in particular the heel. Length: (�1) mm; width: 42 mm; thickness: 22 mm.

n. 53: axe blade, distal fragment (fig. 3, n. 53). The body was probably sub-rectangular. It shows distinct sides, an inclined curved cutting edge

with use fractures, and an oval transverse section. The surfaces are finely polished although some hammering traces are visible, mainly in the central and in the proximal parts. Length: (65) mm; width: 58 mm; thickness: 26 mm.

n. 54: chisel axe (fig. 3, n. 54). It has a massive sub-trapezoidal body whose width decreases from the heel to the cutting edge.

The cutting edge is 25 cm long and slightly curved; the heel is sub-rectilinear, the transverse sec-tion is planar-convex. The surfaces are well polished except in one area of the flat, frontal surface. Length: 128 mm; width: 52 mm; thickness: 29 mm.

n. 55: coarse tuff hammer (?) made from a recycled axe blade (fig. 3, n. 55). It shows a rectangular body with a slightly curved heel, a sub-rectilinear distal edge, distinct

sides and a sub-rectangular transverse section. Polishing is visible mainly on the heel; hammering traces are present in the medial and distal parts. The surfaces are heavily weathered so that it is not possible to describe technological traces in detail. Length: 99 mm; width: 53 mm; thickness: 29 mm.

n. 56: axe blade (fig. 3, n. 56). It has a sub-rectangular massive body with a squared heel, a sub-rectilinear cutting edge with

�0 -

use fractures, slightly rounded sides and a sub-rectangular transverse section. The surfaces are polished although they still present some roughness. Length: 130 mm; width: 59 mm; thickness: 30 mm.

n. 57: axe blade (fig. 3, n. 5�). It has a massive sub-trapezoidal body with a rounded heel, a semicircular cutting edge with use

fractures, distinct sides and a sub-trapezoidal transverse section. The axe width rises from the heel to the cutting edge. All the surfaces are well polished. Length: 112 mm; width: 53 mm; thickness: 26 mm.

n. 59: axe blade (fig. 3, n. 59). It shows a sub-rectangular body with a rectilinear heel, a semicircular cutting edge, convex lateral

edges and a flat oval transverse section. The surfaces are not polished but covered with a uniform hammering. Length: 111 mm; width: 4� mm; thickness: 26 mm.

n. 60: weight (?) (fig. 5, n. 60). It has a triangular shape with slightly concave sides. In the central part of the sides there are, at

least in one case, 10 mm wide notches. The surfaces are covered with hammering traces. Length: 6� mm; thickness: 21 mm.

unnumbered: axe blade (fig. 3, unnumbered). It has a small and flat trapezoidal body with a curved cutting edge with use fractures, distinct sides

and a rectangular transverse section. The surfaces are well polished; polishing stripes in various directions are visible. Length: 51 mm; width: 44 mm; thickness: 13 mm.

AcknowledgementsThe authors are very grateful to Professors P. Biagi and C. D’Amico for the critical review of the manuscript.

Thanks are also due to L. Furlan, technician of the Dipartimento di Scienze della Terra, Trieste University, for the preparation of the samples.

- �1

R E F E R E n C E S

ALBERTI, A., BERNARDINI, F., BURELLI, G., CUCCHI, F., DEMARCHI, G., MONTAGNARI KOKELJ, E., PIANO, C., PRINCIVALLE, F. and Velušček, A. 200� - le materie prime litiche nelle Valli del Natisone e nelle aree limitrofe. In CHIABÀ, M., MAGGI, P. and MAGRINI, C. (eds.) le Valli del Natisone e dell’Isonzo tra Centro-Europa e Adriatico. Studi e ricerche sulla Gallia Cisalpina, 20: 189-208. Pasian di Prato (UD).

BALEN, J., BALEN, D. and KURTANJER, D. 2002 - Kamene alatke s nalazišta Samatovci iz fundusa Arheološkog muzeja u Zagrebu - Stone Artefacts from site Samatovci in the Archaeological Museum in Zagreb. opuscula archaeo-logica, 26: 19-3�. Zagreb.

Burić, M. 2000 - Finding of Nephrite and Jade Axes on the Territory of Croatia. Krystalinikum, 26: 39-43. Brno.

BUSER, S. 1980 - Kameninska sestava kamnitih sekir iz Prekmurja. Poročilo o raziskovanju paleolita, neolita in eneolita v Sloveniji, 8: �9-81. Ljubljana.

BUSER, S. 198� - Kameninska sestava neolitskega kamnitega orodja in Prekmurja. Poročilo paleolita, neolita in eneolita v Sloveniji, 15: 195-19�. Ljubljana.

D’AMICO, C. 2005 - Neolithic “greenstone” axe blades from northwestern Italy across Europe: a first petrographic comparison. Archaeometry, 4� (2): 235-252.

D’AMICO, C., FELICE, G., GASPAROTTO, G., GHEDINI, M., NANNETTI, M.C. and TRENTINI, P. 199� - La pietra levigata neolitica di Sammardenchia (Friuli). mineralogica et Petrographica Acta, 40: 385-426. Bologna.

D’AMICO, C., GASPAROTTO, G., GHEDINI M. and SABETTA, T. 2001 - Serpentiniti e metaultramafiti ad anfiboli e cloriti in asce-martello eneolitiche del NE Italiano. In geoItalia, 3rd forum, section 21, summaries Chieti, 5-8 Sep-tember 2001: 631-632.

D’AMICO, C., GHEDINI, M., MICHELI, R. and MONTAGNARI KOKELJ, E. 1996 - le asce forate del Friuli-Venezia giulia. In VENTURINO GAMBARI, M. (ed.) Le vie della pietra verde, l’industria litica levigata nella preistoria dell’Italia settentrionale. Omega, Torino: 229-238.

D’AMICO, C. and STARNINI, E. 2006 - Prehistoric polished stone artefacts in Italy: a petrographic and archaeological assessment. In MAGGETTI, M. and MESSIGA, B. (eds.) geomaterials in Cultural Heritage. Geological Society, 25�, Special Publications: 25�-2�2. London.

D’AMICO, C., STARNINI, E., GASPAROTTO, G. and GHEDINI, M. 2004 - Eclogite, Jades and other HP-metaophiolites em-ployed for prehistoric polished stone implements in Italy and Europe. Periodico di mineralogia, �3, Special Issue, 3: 1�-42. Roma.

DE MARINIS, R.C. 1996 - La pietra levigata nell’età del Rame dell’Italia settentrionale. In VENTURINO GAMBARI, M. (ed.) Le vie della pietra verde, l’industria litica levigata nella preistoria dell’Italia settentrionale. Omega, Torino: 1�4-1��.

DOZET, S. 2003 - Geologic structure of Kozjansko between Savinja river and Rudnica Mountain, part one: younger Paleozoic and Mesozoic. Rmz - materials and geoenviroment, 50 (3): 6�5-6��. Ljubljana.

EVREN, I. 19�2. - Die Serpentingesteine von Bernstein und Steinbach (Burgenland). Tschermaks mineralogische und Petrographische mitteilungen, 1�: 101-122. Wien.

HANFLAND, C., LäUFER, A.L., NEBELSICK, J.H. and MOSBRUGGER, V. 2004 - The Smrekovec Tertiary and related vol-canism (Slovenia): sedimentology, geochemistry and tectonic evolution. Neues Jahrbuch für geologie und Paläontologie Abhandlungen, 232 (1): ��-125. Stuttgart.

JANAK, M., FROITZHEIM, N., LUPTáK, B., VRABEC, M. and KROGH RAVNA, E.J. 2004 - First evidence for ultrahigh-pres-sure metamorphism of eclogites in Pohorje, Slovenia: tracing deep continental subduction in the Eastern Alps. Tectonics, 23, TC5014. Washington.

�2 -

JANAK, M., FROITZHEIM, N., VRABEC, M., KROGH RAVNA, E.J. and DE HOOG, J.C.M. 2006 - Ultrahigh-pressure meta-morphism and exhumation of garnet peridotite in Pohorje, Eastern Alps. Journal of metamorphic geology, 24: 19-31. Oxford.

KAVUR, B. 2005 - Kamnita orodja na najdišču Čatež-Sredno Polje. In GUšTIN, M. (ed.) Prvi poljedelci. Savska skupi-na lengyelske kulture. Annales Mediterranea: 131-144. Koper.

KOLLER, F. 1985 - Petrologie und Geochemie des Penninikums am Alpenostrand. Jahrbuch der geologischen Bundesanstalt, 128: 83-150. Wien.

KOMAC, M. 2005 - Statistics of the Geological Map of Slovenia at Scale 1 : 250.000. geologija, 48 (1): 11�-126. Ljubljana.

KOROšEC, P. and KOROšEC, J. 1969 - Najdbe s koliščarskih naselbin pri Igu na Ljubljanskem barju. Arheološki kata-logi Slovenije, 3. Ljubljana.

LUBšINA-TUšEK, M. 1993 - Kamnito orodje v severovzhodni Sloveniji. Ptujski Arheološki zbornik (ob 100-letnici Muzeja in Muzejskega Društva): 31-158. Ptuj.

MAJEROWICZ, A., WOJCIK, A., GUNIA, P. and CHOLEWA, P. 2000 - Comparative study of serpentinite textures and rock materials of Neolithic artefacts from Lower Silesia (SW Poland). Kristalinikum, 26: 111-11�. Brno.

MELCHER, F. and MEISEL, T. 2004 - A Metamorphosed Early Cambrian Crust-Mantle Transition in the Eastern Alps, Austria. Journal of Petrology, 45 (8): 1689-1�23. Oxford.

MELCHER, F., MEISEL, T., PUHL, J. and KOLLER, F. 2002 - Petrogenesis and geotectonic setting of ultramafic rocks in the Eastern Alps: constraints from geochemistry. lithos, 65: 69-112. Amsterdam.

Mioč, P. and Žnidarčič, M. 1983 - Osnovna geološka karta SFRJ, Ravne na Koroškem L 33-54, 1:100000. Zvezni geološki zavod, Beograd.

MONTAGNARI KOKELJ, E. 2001 - Pietra verde, Neolitico e post-Neolitico, Carso e Friuli (Italia nord-orientale): lo stato della questione. Atti e memorie della Commissione grotte ‘E. Boegan’, 38: �1-86. Trieste.

Montagnari kokelj, e., d’aMico, c., alBerti, a., Bernardini, F., Burić, M., koMšo, d., PrinciValle, F., teŽak-GREGL, T. and Velušček, a. 2006 - “Greenstone” shaft-hole axes of North-Eastern Italy, Slovenia, Croatia: a new research project. Atti della XXXIX Riunione Scientifica dell’Istituto Italiano di Preistoria e Protostoria: �13-�25. Firenze.

PEDROTTI, A.L. 1990 - L’insediamento di Kanzianiberg: rapporti fra Carinzia ed Italia settentrionale durante il Neo-litico. In BIAGI, P. (ed.) The neolithisation of the Alpine Region, Monografie di Natura Bresciana, 13: 213-226. Museo Civico di Scienze Naturali, Brescia.

PELOI, D. 1996-199� - Le asce-martello in pietra levigata: proposta di lettura analitica ed esempi applicativi a contesti del Friuli Venezia giulia e della Slovenia. Thesis submitted for Dott. Degree, Università degli Studi di Trieste (unpublished).

PESSINA, A., BASTIANI, G., DELLA BIANCA, B. and TONDELLA, L. 2006 - Nuove segnalazioni di industrie in pietra levi-gata dal Friuli. In PESSINA, A. and VISENTINI, P. (eds.) Preistoria dell’Italia settentrionale. Studi in ricordo di Bernardino Bagolini. Museo Friulano di Storia Naturale, Udine: 429-436.

PESSINA, A. and D’AMICO, C. 1999 - L’industria in pietra levigata del sito neolitico di Sammardenchia (Pozzuolo del Friuli, Udine). Aspetti archeologici e petroarcheometrici. In FERRARI, A. and PESSINA, A. (eds.) Sammarden-chia - Cueis. Contributi per la conoscenza di una comunità del primo Neolitico. Edizioni del Museo Friulano di Storia Naturale, 41: 23-92. Udine.

PÉTREQUIN, P., PÉTREQUIN, A.M., ERRERA, M., CASSEN, S., CROUTSCH, C., KLASSEN, L., ROSSy, M., GARIBALDI, P., ISETTI, E., ROSSI, G. and DELCARO, D. 2005 - Beigua, Monviso e Valais. All’origine delle grandi asce levigate di origi-ne alpina in Europa occidentale durante il V millennio. Rivista di Scienze Preistoriche, LV: 265-322.

- �3

PÉTREQUIN, P., ERRERA, M., PÉTREQUIN, A. M. and ALLARD, P. 200� - The Neolithic quarries of Mont Viso, Piedmont, Italy: initial radiocarbon dates. European Journal of Archaeology, 9 (1): �-30.

Petrić, n. 1995 - Sjekire od žadeita i nefrita u pretpovijesti Hrvatske. Histria Archaeologica, 26: 5-2�. Pula.

PICCOTTINI, G. 19�� - Töplitsch. Fundberichte aus Österreich, 16: 291. Wien.

PREMRU, U. 1983 - Osnovna geološka karta SFRJ, Ljubljana L 33-66, 1:100000. Zvezni geološki zavod, Beograd.

SASSI, R., MAZZOLI, C., MILLER, C. and KONZETT, J. 2004 - Geochemistry and metamorphic evolution of the Pohorje Mountain eclogites from easternmost Austroalpine basement of the Eastern Alps (Northern Slovenia). lithos, �8: 235-261. Amsterdam.

šIMEK, M., KURTANJER, D. and PaunoVić, M. 2002 - Eneolitičke glačane kamene alatke iz špilje Vindije (SZ Hrvat-ska) - Aeneolithic polished stone tools from the cave Vindija (NW Croatia). opuscula Archaeologica, 26: 39-55. Zagreb.

SKOCZyLAS, J., JOCHEMCZyK, E. and FOLTyN, E. 2000 - Neolithic serpentinite tools of west-central Poland and upper Silesia. Kristalinikum, 26: 15�-166. Brno.

VAHLKAMPF, G. 19�9 - Urgeschichtliche Funde aus Kärnten. Carinthia, 169: �-1. Klagenfurt.

Velušček, a. (ed.) 2004a - Hočevarica, eneolitsko kolišče na Ljubljanskem barju. Opera Instituti Archaeologici Sloveniae, 8. Ljubljana.

Velušček, a. 2004b - Past and present lake-dwelling studies in Slovenia: Ljubljansko barje (The Ljubljana Marsh). In MENOTTI, F. (ed.) Living on the lake in prehistoric Europe: 150 years of lake-dwelling research. Routledge, London - New york: 69-82.

VON USLAR, R. 1991 - Vorgeschichtliche Fundkarten der Alpen. Römisch-Germanische Forschungen, 48. Mainz.

VRABEC, M., DE HOOG, J.C.M. and JANAK, M. 200� - Origin of UHP garnet lherzolite and serpentinised harzburgites from Pohorje, Eastern Alps, Slovenia. Geochimica et Cosmochimica Acta, �1 (15S): a10�5. San Diego.

Authors’ Addresses:FEDERICO BERNARDINI and EMANUELA MONTAGNARI KOKELJ, Dipartimento di Scienze dell’Antichità, Università degli Studi di Trieste, Via Lazzaretto 6 – I - 34123 TRIESTEe-mails: bernardinifederico@hotmail.com; montagna@units.it

ANTONIO ALBERTI, GABRIELLA DEMARCHI and FRANCESCO PRINCIVALLE, Dipartimento di Scienze della Terra, Università degli Studi di Trieste, Via Weiss 8 – I - 3412� TRIESTEe-mails: alberti@units.it; demarchi@units.it; princiva@units.it

AnTOn VELUŠČEK, Inštitut za arheologijo, Znanstvenoraziskovalnega centra SAZU, Novi trg 2 – SI - 1000 LJUBLJANAe-mail: anton.veluscek@zrc-sazu.si

- �5

RENATO NISBET∗

wooD USE AnD AgRiCULtURE At ViLLAnDRo/ViLLAnDERS (boLZAno/boZEn, ALto ADigE/SoUtH tYRoL):

tHE CHARRED REMAinS FRoM MESoLitHiC to MiDDLE AgES

SUMMARY – Wood use and agriculture at Villandro/Villanders (Bolzano/Bozen, Alto Adige/South Tyrol): the charred remains from mesolithic to middle Ages. A long-term research at Villandro/Villanders, on the slope of the middle Isarco/Eisack Valley, has brought to light a stratigraphy with nine cultural phases, spanning from Mesolithic to Early Middle Ages. The study of the wood used on the site shows the continuity of two main environments, the mixed oak woodland on the lower valley and the conifer forest in the upper part of the slope. The oak association provided most of the wood burnt on the site until the 1st millennium BC but, since the Early Roman period, almost the totality of wood used as firewood or timber came from the spruce and pine forest. This implies a deep change in the local environment during the late 1st millennium BC, for climatic or human pressure, or both. Early Neolithic farming is documented by the presence of Emmer, naked Wheat and, mostly, Barley. During the Roman period, agriculture was largely practised, as shown by the occurrence of the previous mentioned species and, in addition, of Spelt, Rye, Horsebean, Lentil and a wide variety of fruits.

RiASSUnto – Uso delle foreste e agricoltura a Villandro/Villanders (Bolzano/Bozen, Alto Adige/Südtirol): i resti carbonizzati dal mesolitico al medioevo. Una serie di campagne di scavo tra la fine degli anni Ottanta e Novanta a Villandro/Villanders, sul versante della media Val d’Isarco/Eisack, ha messo in luce una complessa stratigrafia con nove fasi culturali che vanno dal Mesolitico all’Alto Medioevo. I carboni di legna studiati nell’articolo dimostrano che la composizione forestale fu sempre dominata da due ecosistemi principali, il querceto termofilo della bassa val-le e la foresta fredda a conifere dell’alto versante. Per tutto il periodo preistorico, la maggior parte del legno fu otte-nuta dal querceto, entro cui probabilmente si trovava l’insediamento. Tuttavia, a partire dai primi secoli AD la quasi totalità del legno proviene dalla foresta di conifere (soprattutto abete rosso e, in subordine, pini), ciò che suggerisce una importante trasformazione dell’ambiente forestale a partire almeno dagli ultimi secoli BC. I resti agricoli hanno messo in luce la presenza di una cerealicoltura sviluppata già a partire dal Neolitico Antico, con Dicocco, Frumento nudo e, soprattutto, Orzo. In età Romana i resti agricoli si arricchiscono di altre specie, come Spelta, Segale, Favino, Lenticchia e ampia varietà di frutti.

Keywords: South Tyrol, open-air site, archaeological sequence, archaeobotany, charred remains.

Atti Soc. Preist. Protost. Friuli-V.G., Trieste, XVI, 2006-200� (2008): �5-131

______________________________________ ∗∗  Torre Pellice, Torino, I

�6 -

1. intRoDUCtion

The open-air site Villandro/Villanders (Bolzano/Bozen) in Alto Adige/South Tyrol (northern Italy), was discovered in 19�9 on a terrace delimited by two deep stream courses, along the western side of the Isarco Valley, in the locality called Plunacker (Plum da Planum) at ca. �00 m (fig. 1), during the excavation of a drainage, some 200 m long and 4 m deep

(DAL Rì, 19�8). In the course of long-term researches started in 198�, a stratigraphy span-ning over seven millennia on three excavation areas, i.e. Trench A, B and C, was found. Most of the sediment was processed by flotation and by handpicking larger fragments visible to the naked eye, allowing the collection of the charred material the present paper deals with. These samples cover a time-span from Late Mesolithic (Castelnovian) to Early and Middle Neolithic, Copper Age, Late Iron Age, 2nd-3rd centuries AD, a later Roman period (end of 4th - beginning of 5th century AD) and two Early Medieval settlements. All these different phases bear evidence of dwelling-places, prehistoric huts and both Roman and Medieval stone houses. This persistence testifies to strategic interest of the site during the centuries. Such a favourable situation furnishes invaluable data concerning the agricultural history and the exploitation of the forest in this part of the Alps, given the particular geographical posi-tion of the site, connecting the two sides of the Alps through a network of passes of more or less easy accessibility.

Most of the researches on charred prehistoric and historical remains of vegetal origin (wood, seeds and fruits) in the Adige-Isarco Valleys was published from the 19�0’s onwards. It covers all the main cultural phases, although with varying frequency.

We must point out, however, the heterogeneity of the collected data, which arises from different sampling methods and frequently concerns only small sub-samples. In some cases the results suffer from the differences in recovery techniques, a large variety of taphonomic conditions (caves and rock-shelters, alluvial fans, peats, colluvial soils etc.), the absence of reference either to the volumes of the original sediment or to the drawn samples, or both.

Fig. 1 - Villandro/Villanders-Plunacker: the archaeo-logical site is located on the meadow in the foreground (photograph by courtesy of l. dal Rì).

- ��

In spite of these limits, the Adige Valley with its nearby surroundings is one of the best studied territories of the whole of Northern Italy with regard to wood [w], charred wood [c] seed (including cereal grains) [s] and fruits [f] studies (fig. 2). Pre-Neolithic sites (mo-stly Epigravettian and Mesolithic) are Bondone [c] (Trento) (KOFLER, 1992), Colbricon [c] (Trento) (BAGOLINI et al., 19�5), Grotta d’Ernesto [c] (NISBET, 1991a), Lago delle Stellune [c] (CASTELLETTI, 1984; CASTELLETTI and MASPERO, 1992), Plan de Frea [c] (Bolzano/Bozen) (AN-GELUCCI et al., 1995; 2002), Riparo Dalmeri [c] (Trento) (CASTELLETTI and MASPERO, 1992), Riparo Villabruna [c] (Belluno) (CASTELLETTI and MASPERO, 1992; AIMAR et al., 1992), Ter-lago [c, f] (Trento) (NISBET, 1983), and Val Lastari [c] (Vicenza) (CASTELLETTI and MASPERO, 1992; BROGLIO et al., 1992).

Fig. 2 - Location of some of the sites mentioned in the text with reference to archaeobotanical studies in Trentino-Alto Adige/South Tyrol. 1: Villandro/Villanders; 2: Velturno-Tanzgasse; 3: Barbiano; 4: Aica di Fiè/Völseraicha; 5: Bressanone/Brixen (a: Millan; b: Villa Dirce; c: Trattengasse); 6: Stufles; 7: Sotćiastel; 8: Plan de Frea; 9: Siebeneich; 10: Seeberg; 11. Similaun; 12: Ganglegg; 13: St. Walburg; 14: Sanzeno; 15: Colbricon; 16: Mezzocorona; 1�: Riparo Gaban; 18: Terlago; 19: Vela di Trento; 20: Santuario di Lasino; 21: Bondone; 22: Lago delle Stellune; 23: Riparo Dalmeri; 24: Grotta d’Ernesto; 25: Riparo Villabruna.

�8 -

Data from Neolithic sites come from Aica di Fiè/Völseraicha (Bolzano/Bozen) [c, s] (BAGOLINI et al., 1982), La Vela di Trento (Trento) [c, s, f] (CASTELLETTI, 19��; DEGASPERI et al., 2006; MOTTES and ROTTOLI, 2006) and further south from Fimòn-Molino Casarotto (Vicenza) [w], in the Berici Hills (CORONA et al., 19�4).

Copper Age sites are those of Bressanone-Millan [c, s, f] (Bolzano/Bozen) (CASTIGLIONI and COTTINI, 2005), Tolerait (Bolzano/Bozen) [s] (BIAGI and NISBET, 198�), and Velturno-Tanzgasse [c, s] (Bolzano/Bozen) (CASTIGLIONI and TECCHIATI, 2005).

More consistent data are from the Bronze Age, as Barbiano [s, f] (Bolzano/Bozen) (anal-yses E. CASTIGLIONI, in MARZATICO and TECCHIATI, 2002), Fiavé [s, f, w] (Trento) (Bronze Age) (JARMAN and GAMBLE, 19�5; JONES and ROWLEy-CONWy, 1984), Ganglegg [s, f] (Bolzano/Bozen) (SCHMIDL and OEGGL, 200�), Monte Croce [c] (Trento) (NISBET, 1999a), Riparo Gaban [c, s] (Trento) (NISBET, 1984), Santuario di Lasino rock-shelter [s, f] (Trento) (Early Bronze Age) (COSTANTINI et al., 2001), Seeberg (Bolzano/Bozen) [s] (OEGGL, 1992), and Sotćiastel [s, f] (Bolzano/Bozen) (SWIDRAK and OEGGL, 1998).

Seven sites belong to the Iron Age: Sanzeno [c] (Trento) (MADELLA and ROTTOLI, 1999), Stufles [c] (Bolzano/Bozen) (CASTELLETTI, 198�), St. Walburg [s] (Bolzano/Bozen) (RöSCH, 2002), Siebeneich [c] (Bolzano/Bozen), Ganglegg [c] (Bolzano/Bozen), Brixen-Trattengasse [s] and Brixen-Villa Dirce [s] (Bolzano/Bozen) (the last four sites are mentioned in SCHMIDL et al., 200�). Three more sites on Lessini Hills, not far from Verona (Veneto), yielded evi-dence of Iron Age farming, i.e. San Giorgio di Valpolicella [c, s] (NISBET, 1992), Monte Loffa [c, s] (NISBET, 1989-1990) and Castelrotto [c, s] (NISBET, 198�a).

Sites of Roman Age with charred remains are Maso Campaccio-Kompatscherhof (Bolzano/Bozen) [c] (NISBET, 1985) and Mezzocorona [c, s, f] (Trento) (CASTIGLIONI and ROTTOLI, 1994).

Previous unpublished work on carbonised materials (mostly seeds and fruits) of differ-ent ages has been carried out in recent years at Villandro, by other researchers. Mention of some of the results is given in a recent paper (OEGGL, 1999).

Pollen analyses were extensively carried out in Alto Adige/South Tyrol mountains (see, for instance, WAHLMüLLER, 1990; STUMBöCK, 1999; KOMPATSCHER and OEGGL, 2000). Some peat bogs located at different altitudes, between 8�0 and 2000 m in the mountains north of Bressanone/Brixen, not far from Villandro (SEIWALD, 1980) were also investigated. All pro-vide a basic reference to the history of vegetation in the middle Isarco/Eisack Valley.

2. MAtERiALS, CHRonoLogY AnD MEtHoDS

The present study has been carried out on 55 samples from nine different archaeological periods, listed in table 1. Most of the samples were extracted by flotation, but unfortunately de-tails on the original soil volume are not available. The carbonised samples show very different volumes, with regard to the period, the context, the extension and the thickness of the layers. Also the charred wood fragments are very different in size and shape. Most of them are less than one cm large, with the exception of some Roman houses (namely US 189, US 190), where the com-bustion of boards and ridgepoles produced the formation of charcoal fragment frequently as large as 10 cms. The analysis was made on sub-samples, apart from very small samples, which were studied in their totality. This regarded especially most of the prehistoric samples.

- �9

Archaeobotanical phases

Archaeological periods US n. Samples n.Sample volume (mL)

1 Recent Early Medieval � 14 1695

2 Intermediate Early Medieval 6 12 2660

3 Roman, late 4th-5th century AD � 13 3�25

4 Early Middle Empire 4 � 1800

5 Roman, 1st-2nd centuries AD 1 2 1185

6 Late Iron Age 1 2 840

7 Copper Age 1 1 350

8 Neolithic 3 12 680

9 Late Mesolithic 1 3 180

total 31 55 13,115

Table 1 - Villandro/Villanders-Plunacker. Chronology, quantity and volumes of the samples.

Charred wood was contained in all samples. They are most frequent, in number and size, in the historical layers, particularly in the US of rooms of Roman and Medieval houses. Some of these houses were used as stores and granaries, some showing evidence of ovens.

Seeds and fruits are present in almost all layers, but with strong differences in density. Apart from the Neolithic seeds, which represent up to now the earliest evidence of agriculture in this part of the Alps, a mass of cereal remains was contained inside some Early Middle Roman Empire buildings (US 190 and US 216). Besides cereal caryopses, crop weeds were recovered in many samples and, since the Roman Age, several edible fruits from domesti-cated trees are also present.

According to the available literature, personal information, radiocarbon dating and other unpublished data, the site has shown the existence of 9 cultural/archaeobotanical phases. The historical layers are dated on the basis of archaeological materials and structures, allowing the identification of five main periods (DAL Rì and RIZZI, 198�-1988; 1989-1990). The sub-division in nine phases is given here in order to arrange the archaeobotanical assemblages in a coherent frame, according to the published archaeological documentation. It must be considered, however, that the samples come from different areas and trenches.

Phase 1 is the Carolingian-Ottonian period, attributed to 9th-10th centuries AD. It is rep-resented by some badly preserved huts, some hearths and floors. The samples are labelled as Altomedioevo recente.

80 -

Phase 2 is attributed to the 6th century AD and consists of some buildings, repeatedly destroyed by fire. The samples from this unit have the indication Altomedioevo intermedio.

Phase 3 is Late Roman (4th-5th centuries AD), with houses and a possible granary. The samples are labelled as Tardo-romano, fine IV-V secolo.

Phase 4 is attributed to the 3rd century. The samples are indicated as Romano, primo medio Impero.

Phase 5, the earliest of the historical sequence, was dated by some coins to the early 2nd century AD. The samples were indicated as Romano, metà I-metà II secolo.

Phase 6 represents a Late Iron Age hut, with post-holes and hearths.Phase � belongs to the Copper Age, here represented by hut and fireplaces.Phase 8 is Neolithic and has some problematic aspects. Radiocarbon dates as well as

lithic and ceramic finds clearly show two different periods of occupation. The earliest period is present in the US 30 tg 9 and US 30 tg 8, and belongs to Early Neolithic Gaban Group. US 30 tg 1 to 5 have been dated to a middle phase of the Middle Neolithic Square-Mouthed Pottery Culture.

The problematic US 30 tg � has a radiocarbon date of Castelnovian Age (Late Meso-lithic) (ETH-30057: 6920±60 uncal BP; 5970-5680 cal BC at 2σ), but contains cereal grains evidently belonging to some upper more recent layers, of Middle Neolithic, but of uncertain provenance.

The distribution of pottery and lithic finds of the Trench C Neolithic sequence shows that also the stratigraphic position of the charred remains must be considered with caution. On the basis of the present information, it appears that some mixture of Early and Middle Neolithic occurred possibly in the past, and this could explain the discrepancies of the radio-carbon dates.

However, we can confidently ascribe the materials from US 80 and US 30 tg 9, carrying no evidence of later, Middle Neolithic artefacts, to the Early Neolithic, Gaban Group. In the upper layers, Early Neolithic chipped stone artefacts and pottery gradually reduces, and we would consider seeds and grains of these layers (tg 8 to 1) as belonging mostly to a middle phase of Middle Neolithic (Square-Mouthed Pottery Culture).

Phase 9 corresponds to the earliest occupation of the site by a group of Castelnovian Late Mesolithic hunter-gatherers.

Seed and fruit analyses were carried out by means of standard equipment that is 10x and 40x binocular microscope, reference collections and current available iconography. Charcoal fragments were examined on three planes of fracture with a darkfield equipment under reflected light. SEM was employed to confirm some doubtful determination for critical taxa.

3. wooD CHARCoAL

The distribution of taxa through the whole stratigraphy is presented in the following tables, both as absolute values (table 2) and percent (table 3), arranged according to archaeo-logical periods. As for the Neolithic, in which two cultural phases were detected, the data are summed up in one figure.

- 81

Phases Pc P PL Pi B A F C/O Co Q Ac U Fr M Pr Pa Po SP Cr J Σ1 9 20 159 �5 3 2 90 41 3 1 4 � 53 46�

2 3 25 �2 68 8 12 4 6 83 2 2 1 15 11 1 1 10 324

3 9 453 216 13 4 15 � 1 3 1 6 �28

4 1�1 92 5 2 5 1 3 2�9

5 5 48 20 22 1 1 35 1 2 135

6 9 42 2 2 3 45 2 1 11 2 119

7 1 1 3 3 8

8 20 2� �3 26 26 23� 2 1 4 2 2 4 424

9 2 16 10 28

Σ 26 146 918 531 49 95 5 26 140 466 5 8 11 35 22 2 4 3 1 �2 2512

Table 2 - Distribution and numbers of carbonised tree taxa. Phase 1: Carolingian/Ottonian period, 9th-10th centuries; Phase 2: Early Medieval, 6th century AD; Phase 3: Late Roman, end 4th-5th century AD; Phase 4: Early Middle Roman Empire; Phase 5: Roman, Middle 1st-middle 2nd century AD; Phase 6: Late Iron Age; Phase �: Copper Age; Phase 8: Neolithic; Phase 9: Late Mesolithic.Taxa: Pc: Pinus cf. cembra; P: Pinus sylvestris/mugo; PL: Picea/larix; Pi: Picea abies; B: Betula sp.; A: Alnus sp.; F: Fagus sylvatica; C/O: Carpinus betulus/ostrya carpinifolia; Co: Corylus avellana; Q: Quercus sp. (cad.); Ac: Acer sp.; U: Ulmus sp.; Fr: Fraxinus excelsior/ornus; M: Maloideae; Pr: Prunus sp; Pa: Prunus cf. avium; Po: Populus sp; SP: Salix/Populus; Cr: Cornus mas; J: Juglans regia.

Phases Pc P PL Pi B A F C/O Co Q Ac U Fr M Pr Pa Po SP Cr J

1 1.9 4.3 34.0 16.1 0.6 0.4 19.3 8.8 0.6 0.2 0.9 1.5 11.3

2 0.9 �.� 22.2 21.0 2.5 3.� 1.2 1.9 25.6 0.6 0.6 0.3 4.6 3.4 0.3 0.3 3.1

3 1.2 62.2 29.� 1.8 0.5 2.1 1.0 0.1 0.4 0.8

4 61.3 33.0 1.8 0.� 1.8 0.4 1.1

5 3.� 35.6 14.8 16.3 0.� 0.� 25.9 0.� 1.5

6 �.6 35.3 1.� 1.� 2.5 3�.8 1.� 0.8 9.2 1.�

7 12.5 12.5 3�.5 3�.5

8 4.� 6.3� 1�.2 6.1 6.1 55.9 0.5 0.2 0.9 0.5 0.5 0.9

9 �.1 5�.1 35.�

Table 3 - Percentages of carbonised tree taxa. Legend, see table 2.

82 -

As shown by the absolute values, the quantity of charcoal fragments exhibits strong differences through the sequence. The phases 1 to 3 (upper part of the historical sequence) are very well represented, as well as the Neolithic (phase 8), due to extensive sampling and flotation. On the contrary, the data referring to the Mesolithic and Copper Age are quite lim-ited and should be used cautiously in the reconstruction of the ancient environment. All the analytical data, with reference to the stratigraphy, are given in Appendix 1.

Altogether, of 2512 fragments, 20 taxa were recognised, � of which on a number of very limited fragments (<10). From this, it appears that the collection of wood was mostly made in a restricted number of wood macro-environments, and this is valid for all the considered archaeological periods.

3.1. ConiFERS

The morphology of the pits in the cross-fields generated by the contact between the cells of horizontal ray parenchyma and the vertical tracheids is an important diagnostic feature in the anatomy of conifers. The pits, present in the samples, are only of two kinds. The piceoid small pits, with strictly elliptical opening, are present only in clusters of two or three and are a distinctive feature of the Picea/larix group.

In most cases it has not been possible to get a more precise determination in this artificial group. When possible, this has been made taking into account the presence/absence of the Sanio bars in the vertical vessels, and a sharp/gradual transition from earlywood to latewood.

Fig. 3 - Spruce (Picea abies (L.) Karsten) needles from US 214, Early Middle Ages. Scale = 5 mm.

- 83

Another kind of wood is characterised by one or rarely two large, rectangular pinoid pits, with horizontal toothed tracheids. The transition from earlywood to latewood is gradual (figs. 4 and 5). This pattern belongs to the section sylvestris, and the fragments have been determined as Pinus sylvestris/mugo.

More doubtful is the identification of Arolla Pine wood (Pinus cembra) in some layers. This is based on the presence of smooth horizontal tracheids, but this characteristic is not easy to find on charred materials.

A further element useful in establishing the composition of the conifer forest around the site was found in US 214 (Phase 1), with 601 fragments of burnt needles with a rhomboid cross section (fig. 3), belonging to Picea abies (L.) Karsten. Their identification was made with regard to the shape of the cross section and to the cross diameter, in comparison with those of other Italian conifers (table 4).

P. uncinata P. cembra P. sylvestris Abies alba larix europaea Picea abies

Widtha 1.31 0.9� 1.4� 1.�8 0.5� 1.13

Thickness 0.48 0.64 0.49 0.43 0.19 0.43

T100/W 36.6 6�.1 34.2 24.2 33.3 39.2

Table 4 - Dimensions of needles of some conifers from Italian Alps: a) width and thickness of the needles are taken at a distance of 30-40 mm from the apex of the leave, on dry material, across the vascular bundle.

Apart from the dimension, the shape of the cross section is a diagnostic element to dif-ferentiate the Abies-Picea-larix group from Pines. Amongst the latter, the cross section of Pinus cembra needles has a characteristic T shape. Moreover, as usual in Pines, the needles of the Arolla Pine also frequently show a typical torsion, which is normally absent in Picea and Abies.

The US 214, where these leaves come from, is a sample of soil stretching between two Early Medieval buildings. One can suppose that the combustion occurred as consequence of the burning of the buildings or of a wood/branches-shed nearby, where fresh leaves and twigs were stored.

Conifer needles are frequently found north of Alps in charred samples, but are not often described in Italy. On the Italian side of Eastern Alps spruce needles are present at Sotćiastel (Bronze Age) (SWIDRACK and OEGGL, 1998) and at Ganglegg (Early Bronze Age) (SCHMIDL and OEGGL, 200�).

3.2. bRoAD-LEAVED tREES The taxon Prunus cf avium (US 30, tg �) has been identified on the basis of the scattered

distribution of pores in transversal section, the rays relatively large (up to 4-5 cells wide) and the presence of thin spiral thickenings in vessel.

The taxon Quercus (deciduous oak) (fig. 9) includes, at least in part, fragments with very numerous latewood pores (>100/mm2) and for this reason they are ascribed to the sec-tion Robur. The abundance of thin-walled tyloses in the vessels of earlywood, the elliptical form of the pores and mostly their small diameter (range between 140-205 μm) point to the

84 -

Fig. 4 - Pinus syl-vestris/mugo, radial plane, from US 25�.

Fig. 5 - Pinus syl-vestris/mugo, cross plane, with resin ca-nals, from US 82.

- 85

species Quercus petraea Liebl. (Durmast oak) according to the Italian literature (CAMBINI, 196�). However, only a part of the oak charcoal fits this identification, a larger number of fragments should be differentiated only as Quercus sp.

The taxon Fraxinus sp. numbers in Italy the species Fraxinus excelsior, F. ornus and F. angustifolia. On ecological ground, we exclude the last species, which grows in the riparian Mediterranean forests. Fraxinus charcoal is found in small percent (0.4% of the total) in almost all layers. Its high fragmentation made it impossible to distinguish between the two species.

The taxon Carpinus/ostrya is present only in the Neolithic layers, with less than 30 fragments. The distinction between the two species is based mostly on the presence (Carpi-nus)/absence (ostrya) of aggregate rays. Unfortunately this feature was not visible on any of the too small fragments belonging to this group. As known, the presence of ostrya carpini-folia pollen in ‘ötzi’s’ transverse colon has suggested his last journey was from the southern side of the alpine range (OEGGL, 1999; OEGGL et al., 200�).

The taxon Salix/Populus could not be differentiated, due to reduced dimensions of the fragments and to the impossibility of identifying the type of the rays (homogenous/heteroge-neous) which would allow the distinction between the two species.

No particular difficulties were found in the identification to a species level of Fagus sylvatica, Corylus avellana, Cornus mas and Juglans regia.

The Maloideae sub-family includes common Rosaceae trees, some of which cultivated, as Hawthorn (Crataegus sp.), Apple tree (malus sp.), Pear tree (Pyrus sp.), Service tree and White Beam tree (Sorbus sp.). The criteria for differentiating these genuses are not yet firmly established.

Several more taxa are present in the charcoal assemblages, such as Acer sp., Betula sp. (fig. 6 and fig. �), Prunus sp. (fig. 8), Alnus sp. (fig. 10) and Ulmus sp. (fig. 11). Also all these woods cannot be differentiated to species level on the ground of their anatomic structure (SCHWEINGRUBER, 1990).

4. SoME ContExtS oF PARtiCULAR intERESt

In some cases large fragments of charcoal show a strong reduction of the latewood rings. For instance (US 151, sector A6-B6, Trench C, Neolithic) in about half of oak fragments the summer rings are almost absent, and in a fragment of �.11 mm 22 rings are present. Such a strong reduction might be due to several factors, affecting the ring growth separately or synergically, but all with an inhibitory action on the cambial layer. The loss of a more or less large part of the foliage and the consequent interruption of the photosynthetic activity might be due to protracted unfavourable climatic conditions, or parasitic attacks, or pastoral activ-ity. At Villandro it was impossible to answer this question.

In US 190, sector QRS/0-1-2-3, Room N of building A (Phase 4) numerous large co-niferous charcoals (Picea/larix) are present. Their shape and size point to a probable use as boards or beams. Similarly, in the US 189, sector P-O/0-1, belonging to the same House A, Loom L, but to a later period (Late Roman, Phase 3) large Picea/larix fragments flat and wide were found, showing evidence of pre-combustion erosion and decay, imputable to a long-term use of wood exposed to trampling or weathering (fig. 14).

86 -

Fig. 6 - Betula sp., radial plane and scalariform perforation plate with over 10 bars. US 325.

Fig. � - Betula sp., cross plane with diffuse pore arrangement. US 325.

- 8�

Fig. 8 - Prunus sp., cross plane. Diffuse to semi-ring porous, multiseriate rays. US 214.

Fig. 9 - Quercus sp., cross plane. Ring porous, tyloses frequent. US 325.

88 -

Fig. 10 - Alnus sp., radial plane, scalariform plate with 15 bars. US 212/2.

Fig. 11 - Ulmus sp., cross plane. Pores in tangential bands in the latewood. US 25�.

- 89

The sample US 206, sector O-P/2, House A Room L (built probably in the Late Roman period, end 4th century AD) was part of a plastered hurdle belonging to a chimney-cowl, possibly collapsed during the first half of the 5th century. The burnt material adhering to the plaster is composed of an agglomerate of charcoal particles of vitreous aspect, without any tissue structure.

Only few very small fragments of Betula, Pinus and Picea/larix were identified. The process of vitrification is due to a long anaerobic heating of the wood/charcoal. Most of this porous material is filled with bubbles and inside the mass leaves and culms fragments of grass are barely recognisable. One can imagine vegetal detritus dragged by the smoke above the fire and laid on the surface during many years (fig. 12). A similar material, but of Phase 2 (Early Medieval), is present in US 125.

Sample US 59 comes also from a Phase 2 House A, Room N. It contains parts of a clay floor some of which clearly show the occurrence of small woody fragments (possibly bark) and coniferous leaves. This might have been done with the intention of obtaining a more resistant cob, in order to prevent fissures in the floor (fig. 13).

Only one sample (US 115, Phase 2), contains evidence of pruning on a branch of Elm (fig. 15). Whether this might be related to fodder procurement for cattle, or to obtain branches for other purposes, is impossible to demonstrate.

Fig. 12 - Soot particles encrusting a chimney-cowl of Late Roman period, 206 House A, Room L.

90 -

5. FoRESt HiStoRY

The analysis of the charred wood from the local stratigraphy allows the reconstruction of the use of the forest by the settlers over a time span of more than six millennia.

The identified taxa show the presence and use of few main woodland associations.

Fig. 13 - Clay floor fragments from US 59, Early Medieval house, filled with bark and conifer needles.

Fig. 14 - Fragments of spruce wood boards with pre-combustion decay, from US 189, House A.

- 91

1. The coniferous forest, as shown by Pines and Spruce/Larch, which was present since the Mesolithic occupation when, certainly due to the small size of the sam-ple, only Pine wood seems to have been used. However, its existence is continuous through the millennia up to the Middle Ages (and to the present time), and seems to be the most important environment where both firewood and timber wood was col-lected. Spruce is probably the most common conifer in the charcoal samples, with a lesser occurrence of larch. The phyto-climatic association with Scots pine, Spruce, Larch, Arolla pine, Mountain pine with underwood with rhododendron and dwarf juniper is characteristic of the Vaccinion-Piceenion, today well documented in the Alto Adige/South Tyrol mountains, and therefore on the slopes of the middle Isarco Valley (GAFTA and PEDROTTI, 1998). The absence of Fir (Abies alba) in the charcoal samples is noteworthy.

2. The oak-wood is the other main forest, largely used all over the archaeological periods. According to the charcoal spectrum this was formed, apart from a deciduous oak spe-cies, by broad-leaved species common at lower altitude, such as Carpinus/ostrya (only in Neolithic layers), a species of Ulmus (only in Neolithic and Iron Age layers) and, with a greater continuity, Fraxinus and Acer. This association could be part of the phyto-cli-matic unit Fraxino orni-ostryon carpinifoliae, in which can usually be found various oak species and, occasionally, Scots pine.

3. During several periods, at the edges of the oak wood, large clearances allowed the ex-pansion of Hazel thickets. At higher altitude, in the transitional belt between the Oak and coniferous wood, a heliophilous forest environment with Betula could develop.

4. At the upper limit of the oak wood the presence of Fagus is attested only sporadically during the historical periods.On the ground of charcoal analysis, we can observe a dramatic change in the forest

environment used for wood supply. Only two main ecosystems were systematically used, the meso-thermophilous broad-leaved forest, dominated by Oak, along the valley floor and the lower slopes, and the coniferous forest, covering the upper slopes. In the anthra-cology of the site, it appears that the relationship between these two main environments has deeply modified over the centuries, the main change occurring at the beginning of the historical period, when the coniferous forest becomes more and more important in the wood supply.

The data from the Mesolithic, though scarce, clearly show however that these two as-sociations were already settled on the slope, and both were close to the site. During the Neo-lithic, probably due to an increased spread uphill of the Oak wood, almost 90% of the wood used on the site comes from this association and from its openings (fig. 16). Four or five millennia later, the situation has completely changed (fig. 1�) and the thermophilous forest is present in the sample composition with less than 5%.

A more detailed study of the nine Neolithic layers (US 30) does not show important changes in the forest composition, where the wood of Oak and Alder, which are present in all samples, has constantly been collected. Conifer wood is present in six layer of nine, perhaps suggesting a lesser availability of that environment as, for instance, because of increased distance from the site. On the basis of frequency analysis for layer, it appears that only in the layer 6 and � wood collection interested a larger number of plants (fig. 18).

92 -

Subsequently the number of species decreases, with a minimum (two taxa only) in the layer 4. Whether this might be due to human choice, or to a casual selection of sample, or to a short period of settling, or to other reasons, is an open question.

Fig. 15 - Elm branch with evidence of pruning at an edge.

Neolithic

n = 424

Quercus

Corylus

Carpinus/Ostrya

Alnus sp

Picea excelsa

Picea/Larix

Populus

Prunus cf avium

Acer

Ulmus

Fraxinus

Prunus

Fig. 16 - Charcoal spectrum from the Neolithic layers, Trench C.

- 93

Fig. 1� - Charcoal spectrum from the Late Roman layers (4th-5th centuries AD).

Late Roman

n = 728

Picea/Larix

Picea excelsa

Pinus

sylvestris/mugo

Juglans

Fraxinus

Alnus

Corylus

QuercusAcer

Betula

Fig. 18 - Tree taxa frequency variation in the Neolithic layers.

0

1

2

3

4

5

6

7

tg 1 tg 2 tg 3 tg 4 tg 5 tg 6 tg 7 tg 8 tg 9

taxa n

.

94 -

Fig. 19 - Barley (Hordeum vulgare), US 216, oven M1, Early Roman Empire.

6. AgRiCULtURE AnD gAtHERing

Favourable preservation conditions and some aspects of harvesting and gathering, sub-sequent treatment and storage of seeds and fruits, allowed the recovery of a considerable quantity of data concerning the agriculture of Early Neolithic to Early Middle Ages.

Tables 5 and 6 show the data referring to the cultivated plant (cereals, legumes, fruit trees) and to wild plant and weeds. The analytical data are reported in Appendix 2.

Apart from the data here mentioned, we should add those from the Neolithic layers (without distinction between Early and Middle Neolithic) listed in a previous paper (OEGGL, 1999). Here Triticum cf. dicoccum, Triticum aestivo-compactum, Hordeum vulgare are re-ported and, amongst the wild fruits, Rubus fruticosus. The same cereals are also present in our samples. Moreover, in comparison with our materials, the interesting presence of lens culinaris, lentil, is mentioned, which in our record is absent in the Neolithic layers, and occurs only in Early Medieval layers.

Moreover, Evans (in DAL Rì and RIZZI, 1989) found Barley and possibly Wheat, together with Horsebean (absent in our samples), Wild Grape and Red-berried elder in Neolithic lay-ers (Trench C). The same author mentioned the presence of Barley, Wheat, Oat, Millet, Len-tils, Grape, Pears/Apples, Cherries and Nuts in the Early Medieval House B.

6.1. CEREALS Cereal grains are documented in all layers, with the exception of the Iron Age ones. The

greater quantity is present in Roman and, to a lesser extent, in Neolithic layers. A descrip-tion of these findings is given here, with some comments on their presence in northern Italy. Dimensions are in mm.

Hordeum vulgare L., Barley (fig. 19).Altogether 6820 grains of barley (almost �0% of the cereal assemblage, with the exclu-

sion of unidentified Cerealia) have been determined. This crop is present in all layers, apart

- 95

from the Iron Age period, which has no evidence of agriculture. Barley is therefore the out-standing crop at Villandro in all periods.

Some caryopses are already present in Early Neolithic layers. The lowest of these, US 30 tg 9, was dated as ETH-30059: 6105±60 uncal BP (5260-4840 cal BC at 2σ). Both exca-vated sectors (sector 44 and F/5) have barley caryopses, for a total of 61 as fragmented and entire grains. Their diameters are: L: 6.65; B: 2.83; T: 2.16. Mean indexes: L/B: 1.68; L/T: 2.23.

1 2 3 4 5 6 � 8 9 ΣCerealsTriticum dicoccum, c. 1132 46 10 1188Triticum dicoccum, one-grain spikelets 2 2Triticum dicoccum, gl. 445 445Triticum dicoccum, b. sp. 541 541Triticum cf. dicoccum, c. 1 1Triticum aestivum/durum/turgidum, c. 6 6 399 8 11 430Triticum dicoccum/spelta, c. 26 26Triticum durum, spikelets 4 4Triticum spelta, gl. 6 2 8Triticum spelta, sp. b. 25 3 28Triticum sp. 2 2 449 53 42 548Hordeum vulgare 3 4 9 6526 3 1 225 6��1Hordeum vulgare (hulled) 1 4 22 1 21 49Hordeum vulgare, sprouts �3 �3Cf. Hordeum 1 1Secale cereale 2 2 20 2 26Triticum/Secale 1 21 22Panicum miliaceum 1 2 3Cerealia identified 10,166Cerealia undeterminated 1 2 10 1046 295 1354

Σ 13 21 22 10�38 11� 4 605 11,520

PodsVicia faba 130 130Vicia sativaPisum sativum 2 2lens esculenta 1 1 2

Σ 1 133 134

treesVitis vinifera 2 1 10 2 5nc 15Vitis sp. 2 2Juglans regia 29 29Pyrus malus 1 1Prunus cf. avium 1 1Prunus spinosa 1 1

Σ 3 41 2 1 2 49

Table 5 - Cultivated plants. Legend: see table 2. c: caryopses; gl.: glumes; sp. b.: forks.

1

96 -

US 30 tg 8 sector 42, with date ETH-30058: 6050±60 uncal BP (5080-4�80 cal BC at 2σ), contained 21 hulled Barley grains, almost all fragmented. Only one grain could be meas-ured: L: 3.8; B: 2.5; T: 2.0.

In tg �, which has an unacceptable, Mesolithic radiocarbon date, 8� grains were found, almost all fragmented or very distorted by combustion. Three grains were measured: L: 4.23; B: 2.8; T: 2.13. Indexes: L/B: 1.51; L/T: 1.99.

Tg 6, sector 13, contained 8 caryopses, but only one could be measured. L: 4.�; B: 3.0; T: 1.8.In the upper two layers (tg 5 and tg 4) no cereal grains were recovered. Only one dis-

torted grain comes from tg 3, ETH-30053: 5425±60 uncal BP (4250-3950 cal BC at 2σ).The last appearance of Barley in Neolithic layers comes from tg 1 sect. D/2, ETH-

30051: 5225±60 uncal BP (4230-3940 cal BC at 2σ) with 8 grains. Mean diameters: L: 5.03; B: 2.63; T: 1.91. Indexes: L/B: 1.�1; L/T: 2.29.

1 2 3 4 5 6 � 8 9 ΣAvena sp. 1 9 1 11Bromus secalinus/hordaceus 1 4 3 8Carex sp. 10 10Cerastium arvense 3 3Cerastium/Stellaria 1 1Chenopodium album 18 18Chenopodium hybridum 1 1Chenopodium polyspermum 1 1Echinochloa crusgalli 3 1 1 5Euphorbia helioscopia 2 2Fragaria vesca 2 2galeopsis tetrahit 1 1 2galium sp. 2 6 9 8 6 31Undeterminated 1 3 4lolium temulentum 3 12 15medicago lupulina 5 5melilotus cf. dentata 1 1Papaver sp. 2 2Pisum sp. 2 2Polygonum convolvulus 1 1Polygonum dumetorum 3 3Polygonum lapatifolium 3 1 4Polygonum persicaria 2 2Quercus sp., acorn 5 5Ranunculus sp. 1 1Rubus idaeus 2 2Rubus sp. 2 2Rumex crispus 16 16Rumex sp. 1 1Sambucus ebulus/racemosa 113 113Sambucus nigra 63 63Setaria viridis/verticillata/italica 1 2 38 2 43Stachys annua/arvensis 1 1Veronica hederaefolia 3� 18 55Vicia cracca 3 3Vicia sativa 1 1Vicia sp. 1 18 4 23Vicia tetrasperma 1 4 1 6Vicia/lathyrus 11 1 12Viola arvensis/tricolor 2 2

Σ 3 5 11 229 12 2 221 483

Table 6 - Weeds and spontaneous vegetation.

- 9�

Apart from the US 30 series, two more Neolithic samples carry Barley grains. Us 80 tg 1 sector 42, the lowest layer of the whole Neolithic sequence, therefore belonging to the Early Neolithic, has 56 grains. L: 4.80; B: 2.�6; T: 1.98; indexes L/B: 1.98; L/T: 2.44. The Neolithic (uncertain cultural attribution and date) US 151 has 4 caryopses: L: 4.55; B: 2.90; T: 2.25. Indexes: L/B: 1.5�; L/T: 2.04.

In some cases it has been possible to observe a typical asymmetry, with longitudinal distortion, which characterises the many-rowed Barley; the L/B index, normally <1.8, would point to the dense-eared variety (H. vulgare var. hexastichum). Only sample US 80 shows a larger L/B value, thus it could contain the 4-rowed variety. Furthermore, the presence of some grains belonging to the hulled variety with fragments of palaea and lemma adhering to the base has been demonstrated. However most of them show a rather rounded apex with a plump cross-section, which is more similar to the naked variety. However, all these charac-ters are still doubtful, due to the distortion and fragmentation of grains.

Sample US 180 (Copper Age) shows two grains of Barley, one of which is broken. The other one (hulled form) has dimensions L: 4.8; B: 2.8; T: 2.2.

The only Iron Age sample has no evidence of cereals, which are instead well represented in the historical layers.

In the sample US 214 (Roman period, between 1st-2nd century AD) there are only three grains, with mean diameters L: 5.40; B: 3.10; T: 2.83; indexes L/B: 1.�4; L/T: 1.93.

The Early Middle Roman Empire layers show the best evidence of Barley through the entire archaeological sequence. In Us 82 there are 4 non-measurable grains and in US 212/2, inside a room of house A, there is only one broken grain. In US 216 (oven) together with an important quantity of other cereals there are 22 caryopses of H. vulgare partly represented by the hulled 6-rowed variety (L: 5.13; B: 3.16; T: 2.��; indexes: L/B: 1.62; L/T: 1.88). In room N, house A (US 190) which seems to be have been used as store, 6521 grains of H. vulgare and more than �3 sprouts have been found. The dimensions of �� grains are L: 6.29; B: 3.36,;T: 2.�3; indexes L/B: 1.86; L/T: 2.30. Only one grain showed fragments of lemma, with its characteristic basal horse-shoe shaped furrow, belonging to 4-rowed Barley.

Barley is poorly documented in later periods. In the late Roman layers (end 4th-5th cen-tury AD) only 9 grains were found (L: 6.30; B: 3.65; T: 3.15) in US 52, US 53 and US 208. Eight Barley grains, four of which belonging to the hulled variety (L: 4.93; B: 3.15; T: 2.50) come from the Intermediate Early Middle Ages. At the end of the Early Middle Ages there are 3 grains of H. vulgare (L: 5.33; B: 3.15; T: 2.50) and one of the hulled variety.

Table � shows the biometric values for cultural periods.

Table � - Hordeum vulgare L., dimensions and indexes.

L l s L/l L/sPhase 1 5.33 3.15 2.50 1.69 2.13Phase 2 4.93 3.15 2.50 1.5� 1.9�Phase 3 6.30 3.65 3.15 1.�3 2.00Phase 4 5.�1 3.26 2.�5 1.�5 2.0�Phase 5 5.40 3.10 2.83 1.�4 1.93Phase 7 4.8 2.8 2.2 1.�1 2.18Phase 8 4.54 2.�8 2.03 1.64 2.25

98 -

From these data it is possible to see an increase in dimensions through the ages, with higher values in the Late Roman period, even taking into account the differences in sample size, distortion of grains due to carbonization and taphonomy.

In northern Italy Barley cultivation is well documented since the Late Neolithic (CASTELLETTI et al., 2001). However, together with Barley, hulled Wheats (Triticum monococ-cum and T. dicoccum) are normally found, while naked Wheats (Triticum aestivum/durum) are rare (ROTTOLI and CASTIGLIONI, 2008). Barley becomes gradually the most frequent crop in the alpine region, perhaps due to its adaptability to unfavourable climate and thin soils. Only since the Roman period is it systematically grown with Rye and, in part, Oat.

The following pre-Iron Age sites of Trentino-South Tyrol show evidence of Barley (table 8).

La Vela Early and Middle NeolithicCASTELLETTI, 19��; MOTTES and ROTTOLI, 2006; DEGASPERI et al., 2006

Hauslabjoch Copper Age ‘ötzi’s’ dress OEGGL, 2000

tolerait Copper Age BIAGI and NISBET, 198�

bressanone-Millan Copper Age CASTIGLIONI and COTTINI, 2005

Velturno-tanzgasse Copper Age CASTIGLIONI and TECCHIATI, 2005

gaban Early Bronze Age NISBET, 1984

Lasino Early and Middle BA COSTANTINI et al., 2001

Fiavè Middle/Recent BA JARMAN and GAMBLE, 19�5

Sotćiastel Middle/Recent BA OEGGL, 1998

Fiavè Middle Bronze Age JONES and ROWLEy-CONWy, 1984

Molina di Ledro Bronze Age DALLA FIOR, 1940

Tab. 8 - Trentino and South Tyrol sites with evidence of barley cultivation.

Neolithic Barley grains from Villandro-Plunacker with calibrated radiocarbon dates be-tween 5260 and 3940 BC are of great interest in order to follow the diffusion of agriculture in the alpine region.

This is, in fact, the earliest evidence of cereal cultivation in inner areas of eastern Alps from the southern side. This could prove the hypothesis (KüSTER, 1995 quoted in OEGGL, 2000) that barley was introduced into the northern side of the Alps from the Mediterranean valleys, and the Adige Valley could well have been one of the natural paths into continental Europe.

In Trentino-South Tyrol Iron Age sites with evidence of Barley are more frequent than those of earlier periods. A recent analysis of alpine sites with evidence of early agriculture (SCHMIDL et al., 200�) has shown the presence of six Alto Adige/South Tyrol Iron Age sites with Barley, i.e. Ganglegg, St. Walburg, Sanzeno, Siebeneich, Trattengasse, Stufels and Villa Dirce. This finds a substantial symmetry with Austrian and Swiss alpine sites of the same age (SCHMIDL and OEGGL, 200�: 5�3), which demonstrates the rapid diffusion of Barley through the Alps during the late prehistory. At Villandro this cereal was still grown

- 99

extensively in Roman periods and only after several centuries does its presence diminish, perhaps because of a lasting abandonment of the slope.

Triticum dicoccumThe difficulty in determining tetra- and hexhaploid Wheats (Triticum dicoccum, T. spelta)

without spike elements is well known. These remains are present at Villandro only from the phase 5 (Roman period, 1st-2nd century AD). When absent, the more narrow and elongated shapes with rounded apex and dorsal longitudinal high profile have been determined as Triti-cum dicoccum, which cannot be considered as naked wheats. On this ground Emmer was deter-mined in two Neolithic layers. In US 30 tg 6 sector 13 there are � caryopses (L: 5.4; B: 2.8; T: 2.4). In US 80 tg 1 sector 42 there are 3 grains whose dimensions are L: 4.3; B: 2.2; T: 2.0.

In the Roman period (1st-2nd century AD) US 214 has 46 caryopses of Emmer (L: 5.63; B: 3.95; T: 2.6; indexes L/B: 1.83; L/T: 2.1�).

Fig. 20 - Naked Wheat (upper row) and Emmer (Triticum dicoccum), lower rows. US 216.

100 -

Later on (Middle Roman Empire, US 216, in an oven) 1018 grains of Emmer were found (fig. 20) having the following dimensions: L: 6.50; B: 3.32; T: 2.91; indexes: L/B: 2,0; L/T: 2,3. These indexes, rather larger than those of sample US 214, overlap those of Triticum dicoccum and T. spelta. There are also many spike fragments, such as glumes (US 414, fig. 21) and spikelet forks (US 541; fig. 22). The dimensions are as follows:

Spikelet forks width: 2.39 mm (1.22-3.31). Of 109 measurements, only 9 are >1.9.Upper disarticulation scar width: 0.89 mm.Glume base width: 0.88 mm (0.��-1.10).In US 190 belonging to the same archaeological period, dominated by Barley grains,

114 grains of Triticum dicoccum and 31 glumes were also found. Their mean width is 0.�� mm (0.55-0.93).

The determination as Triticum dicoccum of the main part of the wheats found at this site, is consistent with glume size and shape, as well as their morphology showing a robust main keel, lateral nerve hardly visible, and sometimes intermediate nerves standing out clearly.

After the 5th century AD, Emmer seems to disappear in the samples, giving place to other cereals.

Table 9 summarises the metrical data for Emmer.

Table 9 - Triticum dicoccum Schreib: measures and indexes of the caryopses.

Emmer, with some other wheat species and barley, is one of the commonest crops in the Italian prehistory, even in alpine areas. It is documented in the Adige Valley in Early and Middle Neolithic at La Vela (Trento) (CASTELLETTI, 19��; MOTTES and ROTTOLI, 2006; DEGASPERI et al., 2006) and is grown during the Copper Age at Tolerait (BIAGI and NIS-BET, 198�), at Brixen-Millan (CASTIGLIONI and COTTINI, 2005) and at Velturno-Tanzgasse (CASTIGLIONI and TECCHIATI, 2005). It is found at Riparo Gaban during the Early Bronze Age, at Fiavé and at Sotćiastel (OEGGL, 1998) during the Middle and Late Bronze Age, at Molina di Ledro (DALLA FIOR, 1940) during the Bronze Age (unspecified). At Villandro, the earliest phases are dominated by Barley, while Emmer is rarely documented. The latter species becomes very important during the Roman colonisation in particular during 4th-5th centuries AD.

Triticum spelta L., Spelt.A small quantity of spikelet forks, with a very short fragment of the upper internode,

belong to a hulled hexhaploid Wheat, Spelt. Spikelet forks more than 2.� mm large, with a maximum of 3.31 mm, are considered pertaining to this species (fig. 22). They represent no more than 5% of the total spikelets in US 216. In the same sample (Early Middle Roman Empire) there are also 26 grains of Spelt/Emmer.

Phases L l s L/l L/s

4 6.50 3.32 2.91 2.02 2.34

5 5.63 3.05 2.60 1.83 2.1�

8 4.85 2.50 2.20 1.95 2.20

- 101

The following mean diameters were obtained from 12 grains: L: �.3 (5.8-�.5); B: 3.1 (2.6-3.�); T: 2.6 (2.3-2.9). Indexes: L/B: 2.�, which is inside the range of the typical Spelt.

Spelt is present in South Tyrol, always in small quantities and in few sites, during the late prehistory. At the Iron Age site of St. Walburg (RöSCH, 2002) Spelt is associated with other cereals, apparently in a votive offering. In the Adige area the earliest Spelt presence is a doubtful determination of Copper Age at Brixen-Millan (CASTIGLIONI and COTTINI, 2005), another one of Early Bronze Age, comes from Riparo del Santuario at Lasino (COSTANTINI et al., 2001). To an earlier period (Middle and Late Bronze Age) belong the findings at Fiavé (JONES and ROWLEy-CONWy, 1984) and Sotćiastel (OEGGL, 1988).

Tetra/hexaploid naked wheat (fig. 20).To this taxon belong several different tetraploid species (Triticum durum Desf, T. turgi-

dum L.) and hexaploid (T. aestivum L., T. compactum Host), as they cannot be separated on the base of the morphology of the grains.

In the Neolithic US 30 tg 6, 10 grains of Triticum cf. aestivum and 1 in US 30 tg 2 were found.

Fig. 21 - Triticum dicoccum glumes. US 216. Scale = 5 mm.

102 -

This taxon appears again in Roman times (US 214, 1st-2nd century AD), with 8 grains, increasing its occurrence during the Early Middle Empire, with 399 grains (US 82, US190, US 216).

In US 216 three groups could be identified, following morphological criteria. Oval grains, with regularly curved lateral sides, with maximum breadth near the distal end, apex usually truncated and germ area deep were determined as Triticum aestivum/durum.

In addition to this prevalent taxon, a shorter and rounded form was observed, possibly belonging to T. compactum or T. turgidum; a third group, very short and thick, is also present. Dimensions of the three groups:

‘Normal’: L: 5.08; B: 3.34; T: 2.81. Indexes: L/B 1.53 (1.22-1.86); L/T: 1.82 (1.30-2.32).‘Short’: L: 4.09; B: 2.96; T: 2.49. Indexes: L/B: 1.40 (1.14-1.53); L/T: 1.�0 (1.24-2.20).‘Thick’: L: 4.8�; B: 3.55; T: 3.49. Indexes: L/B: 1.38 (1.11-1.6�); L/T: 1.39 (1.2�-1.53).How much these differences are due to the deformation during carbonisation or to the

presence of different wheat species, is hard to assess. However, it should be taken into account that from the Roman times onwards a large diversity of forms, arising by the previous millennia of cultivation, occurs and makes often difficult their determination (RöSCH et al., 1992).

Fig. 22 - Forks of Triticum dicoccum and (asterisk) Triticum spelta. US 216.

1

- 103

Few naked wheat grains were found also in intermediate (6) and recent (6) Medieval layers In the Middle Neolithic of Adige Valley naked wheat has been determined at La Vela di Trento as T. aestivum (CASTELLETTI, 19��) and Triticum aestivum/durum (MOTTES e ROT-TOLI, 2006). In later periods it is present at Riparo Gaban (Early Bronze Age) as T. aestivo-compactum (NISBET, 1984), as Triticum cf. compactum at Santuario di Lasino rock-shelter (COSTANTINI et al., 2001) and as cf. T. aestivo-compactum at Sotćiastel (OEGGL, 1988).

Secale cereale L., Rye (fig. 23).The occurrence of Rye is limited to the historical layer, from the 1st century AD onwards.

On a total of 2� grains, 20 belong to the early Middle Roman Empire layers, mostly coming from a granary (House A, Room N). Several grains (22) resembling naked wheat caryopses are determined as Triticum/Secale.

Dimensions of 10 grains from a Middle Roman Empire layer: L: 6.3�; B: 2.5�: T: 2.51. Indici: L/B: 2.48; L/T: 2.56; B/T: 1.03.

It seems that this cereal was of reduced importance, but its presence is not surprising in a farming system dominated by Barley, as normally happens in the Italian agrarian record of these centuries, particularly in the mountains. Rye is a winter-sown crop, well adapting to alpine climatic conditions and ideal to complement spring sowing crops. Medieval hilly sites with contemporaneous presence of Rye and Barley (and other cereals) are known at Monte Barro in Lombardy (CASTELLETTI and CASTIGLIONI, 1991) and Trino Vercellese in Piedmont (NISBET, 1999b). Earlier occurrence of both crops is attested in Roman times in the plain at Brescia (CASTIGLIONI et al., 1999) and Lomello (CASTELLETTI, 19�5; NISBET, 198�b); in Pied-mont at Montaldo di Mondovì, Iron Age (NISBET, 1991b), and at Alba-Borgo Moretta, Bronze Age (MOTELLA DE CARLO, 1995).

Fig. 23 - Oat, Secale cereale. US 216.

104 -

Avena sp., Oat. Only 11 grains of Oat have been recognised, ranging from 1st-2nd century AD to Early

Middle Ages. It is impossible to establish the presence of the cultivated or spontaneous spe-cies. The latter (Avena fatua, A. sterilis) are commonly found as weeds in cereal crops. Mean dimensions: L: 5.61; B: 1.95; T: 1.��.

Panicum miliaceum L., Setaria sp. pl. (Millets).Only 3 grains of Panicum miliaceum (Broomcorn millet) were found, two of them in

US 180 (Copper Age). The cultivation of Millet has been known in Italy, admittedly dis-continuously, since the Late Neolithic (CASTELLETTI et al., 2001). In the eastern Alps Millet is documented at Velturno-Tanzgasse (CASTIGLIONI and TECCHIATI, 2005) and, doubtfully, at Barbiano (NISBET, 1985) during the Copper Age, perhaps as weed in the main crops, and later at Ganglegg and several other Iron Age sites (quoted in SCHMIDL et al., 200�).

A larger quantity of Italian millet grains (Setaria sp. pl.) was found in the crop at Vil-landro, particularly in phase 4, probably as a weed. Grain dimensions are the following: L: 1.3�-2.00; B: 1.25-1.�1; T: 1.2�-1.35. The hilum on the ventral side is elliptical with its long axis about double that of the short one.

6.2. A CASE oF bARLEY FERMEntAtion?With regard to the tens of Barley sprouts found in House A, room N (early Middle Ro-

man Empire), we have taken into account the possibility of a local production of beer. It is well known that, in order to allow the hydrolysis of starch into fermentable mono- and disac-charides, wheat and barley grains must absorb a quantity of water, sufficient for germination, for a span of time of three or four days. Subsequently, grains must be dried in a suitable space, spreading them on a sieve or perforated vessels, so as to block the germination. At this phase the sprouts reach the length of the grain, which is exactly what we find in the case of the car-bonized samples from House A. Only now the grains are roasted, at a temperature of 80-100 °C. During this last process a casual contact with the flame can determine the combustion and the formation of the archaeological sample.

There are some examples of this process both in prehistoric and historical archaeol-ogy. Recently, evidence of beer production has been described in Bell Beakers Culture from Spain, together with the support of a wide experimental documentation (ROJO-GUERRA et al., 2006a; 2006b). In this case, the evidence was provided not by sprouts, but wheat phytoliths and chemical analysis of organic residues encrusting the pottery.

In Italy beer production seems documented by the pollen contained in organic crusts on pottery from an Iron Age cemetery at Pombia, northeast Piedmont (CASTELLETTI et al., 2001)

6.3. PULSE CRoPS

Apart from cereal fields, at Villandro horticulture was also practised, as shown by some legumes in the historical layers. As usual, the presence of legumes is much less important in comparison with cereals, due to the different kind of crop processing, the different harvesting time and, probably, the different storage places.

- 105

Vicia faba L., Horsebean (fig. 24).Only in US 190 (Early Middle Roman Empire) 130 seeds and several fragments of

Horsebean were found. The mean dimensions are L: 8.26; B: 6.01 mm.

Pisum sativum L., Pea.Only two seeds of Pea are present in US 190 (Early Middle Roman Empire). Dimen-

sions taken on the best-preserved one are L: �.4; B: �.0 mm. lens esculenta Moench, Lentil.

Two seeds of Lentil are contained in a sample from US 208 (Late Roman) and in US 221 (intermediate Early Middle Ages). Dimensions of the two seeds: L: 3.3; B: 3.0; T: 1.� and L: 3.5; B: 3.1; T: 2.6 mm.

6.4. wEEDS AnD wiLD HERbS

Wild taxa are common in the samples since the Neolithic layers. Many of these, as an-thropochores, are sown by chance with cereals. Others are present in the local environment and grow easily on tilled soils, as apophytes.

Bromus secalinus/hordaceus, Brome.

Caryopses of Brome were found in the Neolithic, Roman and Early Medieval layers. The dorsal side is convex, the ven-tral one concave in cross sec-tion. Dimensions: L: 3.�-6.3; B: 1.4-2.6; T: 1.1-1.6. This weed is common in the winter crop.

Cerastium arvense L., Field Mouse-ear.

Small seeds (L: 1.13; B: 0.8� mm) of this weed were found in two Neolithic layers. Cerastium arvense is common on dry pastures, at paths and fields margins, mostly on light, calcareous soils. It is less fre-quent on arable lands. It flow-ers in the late spring.

Cerastium/Stellaria. Only one unidentified

seed (L: 1.03; B: 0.88; T: 0.48) Fig. 24 - Horsebean, Vicia faba. US 190. Scale = 10 mm.

106 -

of this taxon is present in an early Middle Roman Empire layer. Cerastium/Stellaria group is formed by herbs common in gardens, along paths and at field margins.

Chenopodium album L., Fat Hen.This species numbers amongst the more important weeds, living in the gardens and

fields growing on all varieties of soil. According to some authors, its high nutrient content was well known to ancient peoples, who used Fat Hen as food. Its high seed production, its frequent presence in the prehistoric agriculture and recent experimental work (STOKES and ROWLEy-CONWy, 2002) point to a real possibility of the consumption of Fat Hen at least in periods of shortage. In Italy this species is known from the Neolithic (COSTANTINI and STAN-CANELLI, 1994). It flowers in summer and autumn. Only 18 seeds of it were found in few Neolithic layers at Villandro.

Chenopodium hybridum L., Maple-leaved Goosefoot.In the Late Roman layers one seed (L: 1.60 mm) of this nitrophilous herb was found.

Flowering in the summer, it is frequent in gardens and in crops.

Chenopodium polyspermum L., Many-seeded Goosefoot.Only one seed of this weed was found inside an oven (US 216), of Middle Roman Em-

pire Age, together with a large quantity of other seeds and cereal grains. This Goosefoot is very common on arable land, on light soils. It flowers in summer and early autumn, and is frequently found in winter crops. Dimensions of the seed: L: 1.1; B: 1.0; T: 0.5).

Echinochloa crus-galli (L.) P. Beauv., Cockspur grass.This grass is commonly found in vineyards, on dump soils and cultivated lands. It flow-

ers in summer. 4 seeds are present in Neolithic and Early Roman layers. Mean dimensions: L: 1.20; B: 1.12; T: 0.��.

Euphorbia helioscopia L., Sun Spurge.In a Neolithic layer two seeds of this herb were found, common in the gardens, on arable

soil and on path margins. It flowers from spring to autumn. Dimensions: L: 2.53; B: 2.02; T: 1.80 mm).

Fallopia convoluvulus Löve, Black Bindweed.One fruit of this species was found in an Early Middle Age layer. It is a typical weed,

common in the cereal fields, in vineyards and along paths and in courtyards, as it does not require nutrient-rich soils. It flowers in the late summer and is therefore commonly found in the winter cereals. Dimensions: L: 2.05; B: 1.83.

Fallopia dumetorum L., Copse-bindweed.This grass is commonly found in meadows and under hedges or bushes. In Middle Ro-

man Empire layers 3 fruits of this species have been found.

galeopsis tetrahit L., Common Hemp-nettle.This grass is commonly found in spring crops and grows mostly on lightly damp soils.

- 10�

Only 2 seeds are present, one in a Neolithic, and another in an Early Middle Empire layer. It is a ubiquitous species, growing in several different communities.

galium cf. aparine, cf. Cleavers (figs. 25 and 26).

This is an annual frequently found in the charred samples. Its seeds are small, rounded, bearing in the middle of the ventral side a deep hole. This is usually circu-lar, only in one case is somewhat squared. The elongated shape of the cells on the surface, barely visible, seems to characterise galium aparine, a grass growing commonly in the crop as weed. It flowers from spring to autumn. Of this species 31 seeds were found mostly in historical layers. The seed with a square hole (L: 2.33; B: 2.00; hole diameter 0.85) could be referred to galium spurium, which has a similar ecology.

lolium temulentum L., Darnel.This weed is found in the Neolithic and Middle Roman Empire layers. It grows in the

cereal fields and its grains, therefore, are common in the crop. Its poisonous properties were well known to the classical authors: infelix lolium was called by Virgil and Columella recom-mended the use of a particular sieve, cribrum loliarium, to part the small caryopses of Darnel from the larger ones of cereals. Mean dimensions: L: 3.3; B: 1.4; T: 1.1.

medicago lupulina L., Black Medick.

In a Neolithic layer 5 seeds of this grass were found. It grows at field margins or in uncultivated land and on moderately nitrog-enous soils. It flowers from spring to autumn.

Fig. 25 - galium cf. aparine, cleavers. US 216. Scale = 1 mm.

Fig. 26 - galium cf. aparine, details of the surface.

108 -

Papaver somniferum L., Opium Poppy.In the Middle Roman Empire layers (US 216) 2 Opium Poppy seeds were found. Their

dimensions are respectively: L: 1.05; B: 0.80; T: 0.68. hilum 0.3� and L: 1.00; B: 0.�5; T: 0.�0; hilum: 0.32.

Polygonum lapathifolium L., Pale Persicaria.This is one of the commonest weeds of any crop, growing mostly on rich and partially

damp soils. It is frequently found in winter cereals, because it flowers in late summer. Only 1 fruit was found in Neolithic layers, and 3 in Early Middle Ages.

Ranunculus sp., Buttercup.In an Early Medieval layer 2 flat, pear-shaped achenes, with a slightly papillous surface

could belong to R. acer, which is common on meadows and pastures. Dimensions: L: 2.25; B: 1.��; T: 0.93 and L: 2.25; B: 1.83; T: 0.8�.

Rumex crispus L. and Rumex sp., Curled Dock. In a Neolithic layer 16 trigonous seeds, somewhat pointed, small (1.9� mm) are referred

to the Curled Dock. This is a weed typical of the riparian communities. Similar seeds but more larger (2,95 mm) belong to the same genus but to an unidentified species found in a Late Roman layer.

Stachys annua/arvensis, Annual/Field Woundwort.In a Roman layer of the 1st-2nd century AD, one oval-shaped seed, with a keel on its

ventral side, was found, belonging to this group. This grass is common on cultivated land, vineyards, and gardens. Stachys annua tends to wind round the stalk of cereals. Dimensions: L: 1.9�; B: 1.3�.

Veronica hederaefolia L., Ivy-leaved Speedwell.The seeds of this weed are oval-shaped, with a dorsal side strongly convex and charac-

teristic transverse ribs, converging towards the deep elliptical hollow, in the ventral side. In the Neolithic layers 19 seeds were found (some not charred), in US 190 (early Middle Roman Empire) 3�. Mean dimensions: L: 2.85; B: 2.38; T: 1.33. Mean aperture diameter: 0.99 mm. The grass grows typically on light soils, and is frequent in the Agrostidion spicae-venti com-munity. It grows commonly in winter cereals.

Vicia cracca/villosa, Tufted/Fodder Vetch.The seeds are globular or sub-globular and are found in the Early Middle Ages layers.

L: 1.�5-2.35; B: 1.68-1.88; hilum mean wide 0.33 mm, mean length 1.90. Some unidentified but similar seeds are also present in the Neolithic layers.

Vicia sp., Vetch.Four sub-globular seeds of Vetch have the following mean dimensions: L: 2.�; B: 2.6.

Hilum width: 0.4.

- 109

Vicia tetrasperma (L.) Schreber, Smooth Tare.This grass has sub-globular seeds with narrow hilum. It grows on cultivated fields and

on wasteland. It flowers mostly in spring, sometimes even in autumn. It has been found (6 seeds) only in Roman and Early Medieval layers. Mean dimensions: L: 2.05; B: 2.03; hilum width: 0.3� and hilum length: 1.2�.

Vicia/lathyrus, Vetch/Vetchling.Eleven sub-globular seeds, with a mean diameter of 1-2 mm, were found in an early

Middle Roman Empire layer. This legume is frequent in the crops as weed.

Viola arvensis/tricolour, Field/Wild Pansy.These two species grow on arable fields and particularly V. arvensis is mostly found in

the cereal crops. Both species are frequent in fallow land and in autumn/winter cereals. Two seeds are present in US 216 (Early Middle Roman Empire). Dimensions: L: 1.5; B: 1.0 and L: 1.8; B: 1.05.

6.5. ECoLogY Amongst the numerous grasses found at Villandro, several belong to the Centauretalia

cyani phytosociological association, which is characteristic of cereal fields. Amongst these, are found Bromus secalinus/hordaceus, galium cf. aparine, lolium temulentum, Papaver som-niferum, Veronica hederaefolia, Vicia tetrasperma, Vicia villosa, Viola arvensis e V. tricolor. Elements of this association can be found in the Neolithic, Roman and Middle Age layers.

0

1

2

3

4

5

6

7

8

gardens field margins cereal fields wasteland grassland paths

sp

ecie

s n

°

Fig. 2� - Main environmental sources as shown by weeds from the Neolithic layers.

110 -

Another typical weed community of Villandro is the Chenopodietalia albi association, with some different varieties. It is formed by weeds frequent on uncultivated soils, or on tilled sandy and nitrogenous soils.

It frequently replaces in late summer the weeds of Centauretalia cyani association. Amongst the weeds belonging to this community Chenopodium album, Chenopodium hy-bridum, Polygonum lapathifolium and other species of arable fields, vineyards and paths margins like galeopsis tetrahit, Fallopia convolvulus, were found in both Neolithic and Ro-man layers.

The species Echinochloa crus-galli belongs to molinio-Arrhenatheretum, frequent in damp meadows with Ranunculus acer and Carex sp., from the first centuries AD.

Watersides and wet pastures not belonging to a crop field environment are represented by the Agropyron-Rumicion crispi association, with numerous seeds of Rumex crispus in the Neolithic layers. Polygonum persicaria, representing the Polygono-Chenopodion polysper-mi, grows on rich-nutrients soils and often on pastured land. It is present only in an Early/Middle Neolithic layer.

Weed associations belong mostly to cereal fields and gardens and were introduced in the final samples with the crops. As components of the local flora, they allow a reconstruction of the main grass environments throughout the ages (fig. 2�).

6.6. FRUitS

Fragaria vesca L., Strawberry.From two different Neolithic layers we have 2 achenes of Strawberry. They could

have been intentionally gathered in the surroundings, at the margins of wood. Dimensions: L: 1.12; B: 0.93; T: 0.�.

Rubus idaeus L., Raspberry.In US 30 tg �, with Mesolithic and Early Neolithic materials, 2 pips of this species

were found. In the same sample an achene of Strawberry was collected. Dimensions: L: 1.9; B: 0.93; T: 1.23.

Rubus sp. Two pips of Blackberry (probably R. caesius L. or R. fruticosus L. s.l.) come from one

of the lower Neolithic layers (US 30 tg 8), with a doubtful attribution to Early Neolithic. Dimensions: L: 2.20; B: 1.33; T: 1.25.

Pyrus malus L., Apple.An Apple seed is present in US 190 (early Middle Roman Empire). It is flattened obo-

vate with pointed base. Dimensions: L: 8.2; B: 4.0; T: 2.9.

Prunus avium L., Cherry. A partially fragmented Cherry stone was found in a Middle Roman Empire layer. The

shape is broadly oval, the wall is 0.55 thick. The ventral side is crossed by a large suture, partially destroyed, 2.40 mm wide, with distinct edges. Dimensions: L: �.0; B: 5.1.

- 111

Prunus spinosa L., Sloe.A whole stone of Sloe is the only palaeoethnobotanical presence for the Iron Age. The

stone is broadly oval, slightly pointed at the apex opposite to hilum. It is small and its surface is covered by a reticulum with distinct ribs. Dimensions: L: 6.5; B: 4.8; T: 4.2.

Sambucus ebulus/racemosa, Dwarf/Red-berried Elder.The seeds are oval-shaped, with a flattened side, the opposite slightly obtuse-carenated.

They are present in almost all Neolithic layers, and are numerous (113). The Dwarf Elder grows normally at path margins, in clearings or meadows, on nitrogenous-rich soils the Red-berried Elder is a small mountainous tree, growing at the edges of beech or conifer forests. Dimensions of the seeds: L: 2.25-2.40; B: 1.60-2.00; T: 1.20-1.40.

The abundance of Sambucus ebulus/racemosa at the site is not easy to explain. The fruits of Dwarf Elder are poisonous and its medicinal properties were known in the past, for instance against snakes poison. The fruits have also been employed as dye (grey-blue). The Red-berried Elder has edible berries, from which a fermented beverage can be produced.

Sambucus nigra L., Elderberry.Elderberry is present only in US 190, Early Roman Empire. Its seeds are oval, slight-

ly elongated and more flattened with respect to Sambucus ebulus. Dimensions: L: 2.5-3.�; B: 1.1-1.4; T: 1.0-1.3.

Quercus sp., deciduous Oak.In US 30 tg 8 (Neolithic) 5 fragmented acorns have been found, whose surface is crossed

by a few longitudinal, undulated furrows.

Juglans regia L., Walnut (fig. 28).From US 190 (Middle Roman Empire) 29 fragments of Walnut shell were extracted.

Evidence of the local cultivation of this tree is demonstrated by the contemporary presence of walnut-tree charcoal, though at a very low percent. Surely this plant was still grown in the following centuries, as its wood is documented in later charred samples.

Vitis vinifera L., Grape Vine (fig. 29).Apart from a few uncharred pips from two Neolithic and one Early Middle Ages lay-

ers, which should be considered as intrusive, Vine cultivation is documented by a small number of seeds (2) of Roman age (1st-2nd century AD), from a more consistent quan-tity (10) in Middle Roman Empire layers and one seed in an Early Medieval layer. Their dimensions are:

Early Middle Ages: L: 5.�; B: 3.5;Middle Roman Empire: L: 4.3-6.2; B: 2.5-3.3; T: 2.0-2.9 ;Roman Age, 1st-2nd centuries: L: 5.�; B: 2.� and L: 5.6; B: 2.0.

112 -

Fig. 29 - Pips of cultivated Grape, Vitis vinifera. US 190.

Fig. 28 - Walnut, Juglans regia. US 190. Scale = 2 cm

- 113

7. ConSiDERAtionS

According to charcoal and seed assemblage, the Mesolithic camp was placed where the morphology of the slope and the vegetal environment were most suitable for hunting. The site lay between the lower limit of the coniferous forest (mainly Scots pine and Dwarf moun-tain-pine.) and the upper margin of the broad-leaved forest (mainly deciduous oak), probably with clearings favourable to heliophilous trees, like Betula (birch).

Later on, the first Neolithic settlers used the site for a longer period of time, as shown by the presence of small huts during Early Neolithic, with well-structured functional areas, both indoor and outdoor (RIZZI and LARCHER RIZZI, 2002). Agriculture was certainly practised on site, as already demonstrated by the presence of flint sickles with their characteristic polish. Crops were possibly cultivated in pre-existent clearings, or by the swidden cultivation (slash and burn) technique. This latter seems to be a real possibility, as indicated by the presence of weed seeds, not specific of cereal crop.

Us 30 tg 9, with the oldest date of the Neolithic stratigraphy, demonstrates that the first settlers here grew Barley and naked Wheat, maybe using the surrounding environment for the collection of wild fruits like Strawberry and Elderberries. During the following occupation in the Middle Neolithic (Square-Mouthed Pottery Culture) the cereal assemblage does not change, but in US 30 tg 6 Emmer makes its first appearance. Meanwhile, many more wild fruits were gathered in the woodland, like Blackberries, Raspberries, Acorns and Wild Cherries. The large amount of charred Corylus wood suggests that also hazelnuts were commonly eaten in this period, but no shells were found in these layers. Meanwhile, the assemblage of apophytes and anthropochores becomes richer and shows the existence of several different environments, as a consequence of an increased human pressure and a prolonged period of permanence.

The slope settled by the Neolithic groups apparently remained unchanged during a long span of time but, probably as a consequence of an intensive exploitation of the nearby soils, new micro-environments are perceptible. The site, lying in the ecotone between the broad-leaved woodland with thermophilous species such as Quercus, Ulmus, Acer and Fraxinus on the lower slope and the coniferous forest on the upper slope, provided access to a broad spectrum of vegetal and animal resources.

However, the presence of poplar and alder charcoals point also to the presence of damp areas or running waters. The large quantity of Oak charcoal with regard to all other taxa, could suggest that the main woody structures of the huts were made of Oak, collected in the near oak woodland. Nevertheless, comparing Early Neolithic with Middle Neolithic samples, it seems that the Neo-lithic impact on the local environment was of very limited entity for more than ten centuries.

About 1500 years later the same area was settled, for a short time, by a small Copper Age group. A new hut was built and the radiocarbon age is R-601: 4660±�0 uncal BP (3650-3300 cal BC at 2σ) (ZANDò, 2002). Unfortunately, charcoal samples are too small and it is impossible to draw a detailed picture of the forest evolution at this time, though it seems that no particular changes occurred. The few carbonised wood fragments show the persistence of previous main associations, oak woodland and conifer forest. Farming is hardly represented, but the presence of Millet should be noted, besides that of Barley.

A new settlement rose in the late Iron Age, foreshadowing those, much more consistent, of historical age. The wood used in the fireplaces found inside a hut still show the use of coni-

114 -

fers and oaks, but provide no evidence of agriculture. The only gathering activity consists of a stone of Prunus spinosa, Sloe, probably collected for its edible properties which were known since earlier times, as well demonstrated by its presence on the mummified body of ‘ötzi’. The relative abundance (>9%) of charcoals of Maloideae in this period may point to a regular consumption of their fruits, apart from the Prunoideae ones.

Around the first half of the 1st century AD new buildings were erected, and there is also evidence of some alimentary plants. Rye was grown for the first time on the site, together with emmer, naked wheat and barley. The vine was certainly of the cultivated variety (Vitis vinifera). The charcoals found in these layers document, for the first time in the stratigraphy, an overwhelming presence (>�0% of the total) of conifers. This might be due to the spread of agriculture or pastoral activities, or to a progressive erosion of the mixed oak forest, partially used for providing timber. From this point onwards, however, the main wood source was to be the upper part of the slope with its pine and spruce forests, for the following millennium.

The following archaeological layer, early Middle Roman Empire, has by far the richest documentation of the whole sequence. The setting becomes permanent, with larger houses or farmhouses with ovens and storehouses. In a stone house (‘House A’), the ‘Room N’ was used, at least in part, as a crop store, for Barley, spikes of various Wheats (Triticum dicoc-cum, T. aestivum), and rye together with crop weeds such as galeopsis tetrahit, Vicia sp., Polygonum dumetorum, Setaria sp., Bromus sp., Avena sp. etc.

Moreover, in this room pulses such as horse beans and peas, and fruits, namely walnuts and bunches of grapes were stored. We have previously discussed the possibility that a part of this room was dedicated to the germination of barley grains, as part of the beer making process. Almost all wood used in the houses (94.3%) comes from the conifer forest. It is therefore possible that, around the 2nd century AD, the lower margin of the pine and spruce woodland was close to the site. After the burning of these buildings new houses were built at the end of 4th century AD (Late Roman period). The agricultural data become less abundant in respect to the previous occupation, but it is still considerable.

A very pronounced use of conifer wood (>93% of the total) is even now evident, together with walnut trees (walnuts had been eaten since the preceding period), which provides an excel-lent wood for carpentry. Barley is still widely grown and lentils make their first appearance.

However, dig reports (DAL Rì and RIZZI, 1989-1990) relate that, following the collapse of a room caused by fire, several kinds of charred fruits were preserved: bunches of grapes, apples, pears, plums and chestnuts. In another room a bowl filled with almonds and chestnuts was found (interestingly, the wood of this latter species was not found in any of the archaeo-logical layers). These materials were not included in the present report. The Early Middle Age layers provided very few palaeoethnobotanical remains. Besides Barley, oat and naked Wheat were found. Some Vine pips still testify the presence of cultivated grapes.

8. ConCLUSionS

Multi-disciplinary researches on the prehistoric ecology in the eastern Alps had a strong increase following the recovery of the frozen Copper Age mummy (BORTENSCHLAGER and OEGGL, 2000; DICKSON et al., 2000; OEGGL et al., 200�). However, a comprehensive picture of

- 115

farming techniques and forest evolution had already been reached by means of earlier studies of alpine sites, some of which in the inner alpine chain, thanks to a renewed interest in man-mountain relationships. The Alps are no longer considered marginal areas, but as a primary nuclei of radiation towards new environments and as rich stores of animal, floral and mineral resources, some of which indispensable to the development of new technologies. However, it must be recognised that most of the palaeoagricultural records from northern Italy come almost exclusively from the lowlands or hilly regions, as shown in recent reviews (ROTTOLI and CASTIGLIONI, 2008). For this reason, the research carried out at Villandro/Villanders-Plunacker can contribute to a better understanding of some aspects of the penetration and the strengthening of farming over a time span of about seven millennia.

First of all, the primary importance of Barley since the Early Neolithic, as an economic and alimentary resource, is confirmed.

Secondly, and differently to what was found in the lowlands, at Villandro the presence of Emmer and naked wheat is quite limited. In the Adige Valley, at La Vela (Trento), Barley is associated with wheat in Early Neolithic (CASTELLETTI, and ROTTOLI, 1998-1999) and, fur-ther south, a single Barley grain was extracted in a pollen core from a peat-bog at Isera, and radiocarbon dated to the second half of the 6th millennium BC (CALDERONI et al., 1996, from PEDROTTI, 2002).

Thirdly, the radiocarbon dates show that since the end of the 6th millennium BC groups of farmers spread deep into the alpine chain, where they could adapt to mountain environ-ment and resources. These small communities fit chronologically in the Gaban Group, with radiocarbon dates (BAGOLINI and BIAGI, 1990) well matching those from the deepest layers of Trench C at Villandro.

With regard to the recent debate rising from the occurrence of single pollen grains of Cerealia-type of Early Holocene date, from South Tyrol, Tyrol and Swiss Alps, which are considered by several authors as an indication of Late Mesolithic proto-agriculture, we refer to the recent criticism from BEHRE (200�). At the moment, the earliest consistent evidence of cereal spread near or close to the southern side of Alps, based on macro-remains, are to be found at Villandro in South Tyrol, at Sammardenchia in Frioul (ROTTOLI, 2000) and pos-sibly at Isolino di Varese in Lombardy (CASTELLETTI, 1990), all sites belonging to the Early Neolithic age. More consistent evidence comes from the plain or the Apennines, as listed in ROTTOLI and CASTIGLIONI (2008).

AcknowledgementsI owe my thanks to Drs. L. Dal Rì and U. Tecchiati (Ufficio Beni Archeologici, Soprintendenza Provinciale

ai Beni Culturali di Bolzano-Alto Adige) who charged me in 2005 for the study of the charred materials from Villandro/Villanders. They, and Dr. G. Ricci (Società Ricerche Archeologiche, Bressanone/Brixen), have also pro-vided the indispensable documentation on stratigraphy, radiocarbon dating and references. Prof. P. Biagi and Dr. F. Fuolega (Università Ca’ Foscari di Venezia, Dipartimento di Scienze dell’Antichità e del Vicino Oriente. Sezione Archeologia) gave valuable advice on the chronology of the Trench C (Early and Middle Neolithic). Thanks are due to Prof. U. Casellato and Dr. F. Fenzi (CNR-ICIS, Padova), and Prof. D. Mazza (Dipartimento di Scienza dei Materiali, Politecnico di Torino) for the use of SEM, and Prof. K. Oeggl, who provided some important references on the archaeobotany carried out at the Institut für Botanik, Universität Innsbruck. I am particularly indebted with J. Robinson for the revision of the original English text.

116 -

R E F E R E n C E S

AIMAR, A., ALCIATI, G., BROGLIO, A., CASTELLETTI, L., CATTANI., D’AMICO, C., GIACOBINI, G., MASPERO, A. and PERESA-NI, M. 1992 - Les Abris Villabruna dans la Vallée du Cismon. Preistoria Alpina, 28 (1): 22�-254. Trento

ANGELUCCI, D.E., ALESSIO, M., BARTOLOMEI, G., CASSOLI, P.F., IMPROTA, S., MASPERO, A. and TAGLIACOZZO, A. 2002 - Il Riparo Frea IV (Selva Val Gardena, BZ). Atti della XXXIII Riunione Scientifica dell’IIPP: 145-160.

ANGELUCCI, D.E., BARTOLOMEI, G., BROGLIO, A., BARTOLOMEI, P.F., IMPROTA, S., MASPERO, A. and TAGLIACOZZO, A. 1995 - Plan de Frea Site 4 (Val Gardena, Dolomites): Mobility. Actes du 5ème Colloque International UISPP. Grenoble: 443-448.

BAGOLINI, B., BARBACOVI, F., CASTELLETTI, L. and LANZINGER, M. 19�5 - Colbricon (scavi 19�3-19�4). Preistoria Alpina, 11: 201-235. Trento

BAGOLINI, B. and BIAGI, P. 1990 - The Radiocarbon Chronology of the Neolithic and Copper Age of Northern Italy. oxford Journal of Archaeology, 9 (1): 1-24.

BAGOLINI, B., BIAGI, P. and NISBET, R. 1982 - Ricerche negli insediamenti di Fingerhof presso Aica di Fiè (Völserai-cha-BZ). Rapporto preliminare sugli scavi 1980-1981. Rivista di Archeologia, VI: 11-2�. Venezia.

BEHRE, K.E. 200� - Evidence for Mesolithic agriculture in and around central Europe? Vegetation History and Archaeobotany, 16: 203-219.

BIAGI, P. and NISBET, R. 198� - Ursprung der Landwirschaft in Norditalien. zeitschrift für Archäologie, 21: 11-24. Berlin.

BORTENSCHLAGER, S. and OEGGL, K. 2000 - The Iceman and his Natural Environment. Springer, Wien and New york.

BROGLIO, A., CASTELLETTI, L., FRIGO, G., MARTELLO, G.V., MASPERO, A. and PERESANI, M. 1992 - Le site épigravettien de Val Lastari sur l’Haut Plateau d’Asiago (Préalpes de la Vénétie). Preistoria Alpina, 28 (1): 20�-225. Trento

CALDERONI, G., FINOTTI, F., ILICETO, V., LEOPARDI, D. and PAGANELLI, A. 1996 - Analisi polliniche e radiometriche nella torbiera di Isera. In TECCHIATI, U. (ed.) dalle radici della storia. Archeologia del Comun Comunale lagarino: storia e forme dell’insediamento dalla preistoria al medio Evo. Museo Civico di Rovereto, Rovereto: 2�-34.

CAMBINI, A. 196� - Riconoscimento microscopico del legno delle querce italiane. In GIORDANO, G. (ed.) Contributi scientifico-pratici per una migliore conoscenza ed utilizzazione del legno. CNR, Roma: 51-�9.

CASTELLETTI, L. 19�5 - Segale (Secale cereale L.) subfossile a Lomello. Centro Studi e documentazioni sull’Italia Romana Atti, VI: 55-�1. Como.

CASTELLETTI, L. 19�� - Resti vegetali macroscopici da “La Vela”. In BAGOLINI, B. (ed.) l’ambiente neolitico de “la Vela” (Trento). Il momento meandrospiralico nella Cultura dei vasi a bocca quadrata. Museo Tridentino di Scienze Naturali, Trento: 59.

CASTELLETTI, L. 1984 - Il combustibile legnoso negli insediamenti Mesolitici dell’Italia Settentrionale. Preistoria Alpina, 19: 235-242. Trento

CASTELLETTI, L. 198� - Resti della struttura lignea della “Casa Retica” di Stufles. denkmalpflege in Südtirol/Tutela dei Beni Culturali in Alto Adige: 195-216. Athesia, Bolzano.

CASTELLETTI, L. 1990 - Relazione preliminare sui resti macroscopici vegetali dell’Isolino di Varese: scavi 19��-1985. In BIAGI, P. (ed.) The neolithisation of the Alpine Region. Monografie di Natura Bresciana, 13: 20�-212. Brescia.

CASTELLETTI, L. and CASTIGLIONI, E. 1991 - Resti vegetali. In BROGIOLO, G.P. and CASTELLETTI, L. (eds.) Archeologia a monte Barro. I: Il grande edificio e le torri. Stefanoni, Lecco: 169-203.

CASTELLETTI, L., CASTIGLIONI, E. and ROTTOLI, M. 2001 - L’agricoltura dell’Italia settentrionale dal Neolitico al Me-dioevo. In FAILLA, O. and FORNI, G. (eds.) le piante coltivate e la loro storia: 33-84. Franco Angeli, Milano.

- 11�

CASTELLETTI, L. and MASPERO, A. 1992 - Dati antracologici relativi all’ultima fase della glaciazione nelle Alpi Orien-tali italiane. Preistoria Alpina, 28 (1): 105-114. Trento.

CASTELLETTI, L., MASPERO, A., MOTELLA DE CARO, S., PINI, R. and RAVAZZI, C. 2001 - Il contenuto del bicchiere della t. 11. In GAMBARI, F.M. (ed.) la birra e il fiume. Pombia e le vie dell’ovest Ticino tra VI e V secolo a.C. Celid, Torino: 10�-109.

CASTELLETTI, L. and ROTTOLI, M. 1998-99 - L’agricoltura neolitica italiana. Una sintesi delle conoscenze attuali. In PESSINA, A. and MUSCIO, G. (ed.) Settemila anni fa… il primo pane. Catalogo della mostra. Comune di Udine e Museo Friulano di Storia Naturale, Udine: 15-24.

CASTIGLIONI, E. and COTTINI, M. 2005 - I resti botanici carbonizzati dalla struttura a scorie di Millan. Indagini preli-minari. In DAL Rì, L. and TECCHIATI, U. (eds.) der spätkupferzeitliche Schmelzplatz von milland bei Brixen im Rahmen der beginnenden metallurgie in alpinen Raum. Internationale Tagung. Bozen: 13-15.

CASTIGLIONI, E., COTTINI, M. and ROTTOLI, M. 1999 - I resti botanici di Santa Giulia a Brescia. In BROGIOLO, G.P. (ed.) S. giulia di Brescia, gli scavi dal 1980 al 1992. Reperti preromani, romani e alto medievali. All’Insegna del Giglio, Firenze: 401-424.

CASTIGLIONI, E. and ROTTOLI, M. 1994 - Resti vegetali: carboni, semi e frutti. Ricostruzione dell’ambiente vegetale. In CAVADA, E. (ed.) Archeologia a mezzocorona. documenti per la storia del popolamento rustico di età ro-mana nell’area atesina. Provincia Autonoma di Trento, Trento: 205-231.

CASTIGLIONI, E. and TECCHIATI, U. 2005 - Ricerche archeobotaniche nei livelli della tarda età del Rame di Velturno-Tanzgasse in val d’Isarco (BZ). Atti dell’Accademia Roveretana degli Agiati, 255 (5): 505-528, Rovereto.

COSTANTINI, L., LAURIA, M. and TECCHIATI, U. 2001 - I resti carpologici dell’antica e media età del Bronzo del Riparo del Santuario di Lasino (Trento) - Scavi 1996. Annali del museo Civico di Rovereto, sezione Archeologia, Storia e Scienze Naturali, 1�: 3-40. Rovereto.

COSTANTINI, L. and STANCANELLI, M. 1994 - La preistoria agricola dell’Italia centro-meridionale: il contributo delle indagini archeobotaniche. origini, XVIII: 149-244.

DALLA FIOR, G. 1940 - Analisi polliniche di torbe e depositi lacustri della Venezia Tridentina. la palafitta di molina di ledro (655 m.). Memorie del Museo di Storia Naturale della Venezia Tridentina, V: 9�-148. Trento.

DAL Rì, L. 19�8 - Villandro-Villanders (Bolzano). Preistoria Alpina, 14: 243-244. Trento.

DAL Rì, L. and RIZZI, G. 198�-1988 - Villandro. Plunacker 1988. denkmalpflege in Südtirol/Tutela dei Beni Culturali in Alto Adige: 59-61. Athesia, Bolzano.

DAL Rì, L., and RIZZI, G. 1989 - Archäologische Ausgrabungen auf dem Plunacker in Villanders. der Schlern, April 1989: 201-224. Athesia, Bozen.

DAL Rì, L. and RIZZI, G. 1989-1990 - Villandro. Plunacker, Zona A-B. denkmalpflege in Südtirol/Tutela dei Beni Culturali in Alto Adige: 81-86. Athesia Bolzano.

DEGASPERI, N., MOTTES, E. and ROTTOLI, M. 2006 - Recenti indagini nel sito neolitico de La Vela di Trento. In PES-SINA, A. and VISENTINI, P. (eds.) Preistoria dell’Italia settentrionale. Studi in ricordo di Bernardino Bagolini. Pubblicazioni del Museo Friulano di Storia Naturale, 53: 143-168. Udine.

DICKSON, J.H., OEGGL, K., HOLDEN, T.G., HANDLEy, L.L., O’CONNELL, T.C. and PRESTON, T. 2000 - The omnivorous Tyrolean Iceman: colon contents (meat, cereals, pollen, moss and whipworm) and stable isotope analyses. Philosophical Transactions of the Royal Society of london, B, 355: 1843-1849.

GAFTA, D. and PEDROTTI, F. 1998 - Fitoclima del Trentino-Alto Adige. Studi Trentini di Scienze Naturali, Acta Bio-logica, �3: 55-111. Trento.

JARMAN, H.N. and GAMBLE, C. 19�5 - Plant remains from Fiavè: Preliminary report. Preistoria Alpina, 11: �5-�6. Trento.

JONES, G. and ROWLEy-CONWy, P. 1984 - Plant Remains from the North Italian Lake Dwelling of Fiavé (1400-1200 BC). In PERINI, R. (ed.) Scavi archeologici nella zona palafitticola di Fiavé-Carera, parte I. Trento: 323-355.

118 -

KOFLER, W. 1992 - Die Vegetationentwicklung im Spätpaläolithikum und Mesolithikum im Raume Trient. Preistoria Alpina, 28 (1): 83-103. Trento.

KOMPATSCHER, K. and OEGGL, K, 2000 - Late-glacial and Holocene Vegetation Development of the überetsch (Südti-rol). In WICK, L. (ed.) Southern Alps. Excursion Guide of the XXIV Bernese Mire excursion: 124-129. Insti-tute of Plant Sciences: University of Universität Bern, Bern.

KüSTER, H. 1995 - Postglaziale Vegetationsgeschichte Südbayerns. geobotanische Studien zur prähistorischen landschaftskunde. Akademie-Verlag, Berlin.

MADELLA, M. and ROTTOLI, M. 1999 - Resti lignei carbonizzati da uno scavo a Sanzeno. In CIURLETTI, G. and MAR-ZATICO, F. (eds.) I Reti/die Räter. Archeologia delle Alpi, 5: 99-105, Trento.

MARZATICO, F. and TECCHIATI, U. 2002 - L’età del Bronzo in Trentino e Alto Adige/Südtirol. Atti della XXXIII Riu-nione Scientifica dell’IIPP: 45-92.

MOTELLA DE CARO, S. 1995 - Paleoecologia ad Alba nella Preistoria. Indagine sui macroresti vegetali. In VENTURINO GAM-BARI, M. (ed.) Navigatori e contadini. Alba e la valle del Tanaro nella Preistoria. Camedda, Torino: 245-255.

MOTTES, E. and ROTTOLI, M. 2006 - I resti carpologici del sito neolitico de La Vela di Trento (campagne di scavo 19�5 e 19�6). In PESSINA, A. and VISENTINI, P. (eds.) Preistoria dell’Italia settentrionale. Studi in ricordo di Bernardino Bagolini. Pubblicazioni del Museo Friulano di Storia Naturale, 53: 129-142. Udine.

NISBET, R. 1983 - Analisi macrobotaniche preliminari del deposito di Terlago (Trento). Preistoria Alpina, 19: 243. Trento.

NISBET, R. 1984 - Vegetazione e agricoltura durante l’età del Bronzo al Riparo Gaban (Trento). Preistoria Alpina, 20: 301-310. Trento.

NISBET, R. 1985 - Aspetti cronologici dell’agricoltura neolitica dell’Italia settentrionale. In MORONI, A., ANELLI, A. and RAVERA, O. (eds.) Atti del secondo Congresso Nazionale della Società Italiana di Ecologia. Zara, Parma: 943-945.

NISBET, R. 198�a - I vegetali carbonizzati nell’insediamento dell’età del Ferro di Castelrotto (Verona). In ASPES, A. and SALZANI, L. (eds.) Prima della storia. Inediti di 10 anni di ricerche a Verona. Museo Civico di Storia Naturale, Verona: 121-130.

NISBET, R. 198�b - Ricerche archeobotaniche. In BLAKE, H. and MACCABRUNI, C. (eds.) Dallo scavo a Villa Maria di Lomello (Pavia), 1984: la buca tardo-antica 203. Archeologia medievale, XIV: 1��-182. Firenze.

NISBET, R. 1989-1990 - I grani carbonizzati dell’età del Ferro a Monte Loffa. Annuario Storico della Valpolicella: 21-28. Fumane.

NISBET, R. 1991a - Le analisi dei carboni. In AWSIUK, R., BARTOLOMEI, G., CATTANI, L., CAVALLO, C., DALMERI, G., d’ERRICO, F., GIACOBINI, G., GIROD, A., HERCMAN, H., JARDON-GINER, P., NISBET, R., PAZDUR, M.F., PERESANI, M. and RIEDEL, A., La Grotta d’Ernesto (Trento): frequentazione umana e paleoambiente. Preistoria Alpina, 2�: 65-66. Trento.

NISBET, R. 1991b - Storia forestale e agricoltura a Montaldo tra età del Ferro e XVI secolo. In MICHELETTO, E. and VENTURINO GAMBARI, M. (eds.) montaldo di mondovì. Un insediamento protostorico. Un castello. De Luca, Roma: 24�-251.

NISBET, R. 1992 - Una nota sull’agricoltura veneta dell’età del Ferro. In BRUGNOLI, P and SALZANI, L. (eds.) San gior-gio di Valpolicella. Scavi archeologici e sistemazioni museali. Banca Popolare di Verona, Vago di Lavagno: 88-92.

NISBET, R. 1999a - Nota sui carboni di Monte Croce. In CIURLETTI, G. and MARZATICO, F. (eds.) I Reti/die Räter. Archeologia delle Alpi, 5: 221-222. Trento.

NISBET, R. 1999b - I macroresti vegetali. In NEGRO PONZI MANCINI, M.M. (ed.) San michele di Trino (VC). dal vil-laggio romano al castello medievale. Ricerche di Archeologia Altomedievale e Medievale, 25-26: 621-630. All’Insegna del Giglio, Firenze.

- 119

OEGGL, K. 1992 - Zur Besiedlung des mittleres Alpenraumes während der Bronze- und Eisenzeit: Die Vegetati-onsverhältnisse. In KOVAR-EDER, J. (ed.) Palaeovegetational development in Europe and regions relevant to ist palaeofloristic evolution. Proceedings of the Pan-European Palaeobotanical Conference (PEPC): 4�-5�, Naturhistorisches Museum, Vienna.

OEGGL, K. 1999 - Die letzte Mahlzeit des Mannes aus dem Eis. In FLECKINGER A. (ed.), die gletschermumie aus der Kupferzeit. Neue Forschungsergebnisse zum mann aus dem Eis/la mummia dell’età del rame. Schriften des Südtiroler Archäologiemuseums, 1: 9�-110. Folio, Bozen.

OEGGL, K. 2000 - The diet of the Iceman. In BORTENSCHLAGER, S. and OEGGL, K. (eds.) The man in the ice. IV. The iceman and his natural environment. Springer, Wien: 89-116.

OEGGL, K., KOFLER, W., SCHMIDL, A., DICKSON, J.H., EGARTER-VIGL, E. and GAGER, O. 200� - The reconstruction of the last itinerary of “ötzi“, the Neolithic Iceman, by pollen analyses from sequentially sampled gut extracts. Quaternary Science Reviews, 26: 853-861.

PEDROTTI, A. 2002 - Il Neolitico in Trentino Alto Adige. Atti della XXXIII Riunione Scientifica dell’IIPP: 33-43.

RIZZI, G. and LARCHER RIZZI, G. 2002 - Ricostruzione ipotetica di una capanna neolitica di Villandro-Plunacker (Bolzano). Atti della XXXIII Riunione Scientifica dell’IIPP: 225-22�.

ROJO-GUERRA, M.A., GARRIDO-PENA, R. and GARCíA-MARTíNEZ-DE-LAGRáN, í. 2006a - Un Brindis con el pasado: la Cerveza hace 4500 años en la Península Ibérica. Universidad de Valladolid, serie Arte y Arqueología, 22. Valladolid.

ROJO-GUERRA, M.A., GARRIDO-PENA, R., GARCíA-MARTíNEZ-DE-LAGRáN, í., JUAN-TRESSERAS, J. and MATAMALA, J.C. 2006b - Beer and Bell Beakers: Drinking Rituals in Copper Age Inner Iberia. Proceedings of the Prehistoric Society, �1: 243-265.

RöSCH, M. 2002 - Botanische Untersuchung einer Bodenprobe des eisenzeitlichen Brandopferplatzes von St. Walburg im Ultental, Südtirol. In SöLDER, W. (ed.) Kult der Vorzeit in den Alpen. Athesia, Bozen: 95�-959.

RöSCH, M., JACOMET, S. and KARG, S. 1992 - The history of cereals in the region of the former Duchy of Swa-bia (Herzogtum Schwaben) from the Roman to the Post-medieval period; results of archaeological research. Vegetation History and Archaeobotany, 1 (4): 193-231.

ROTTOLI, M. 2000 - Nascita e sviluppo dell’agricoltura in Italia settentrionale. I nuovi dati archeobotanici. In PESSINA, A. and MUSCIO, G. (eds.) la neolitizzazione tra oriente e occidente. Museo Friulano di Storia Naturale, Udine: 91-98.

ROTTOLI, M. and CASTIGLIONI, E. 2008 - Prehistory of plant growing and collecting in northern Italy, based on seed remains from the early Neolithic to the Chalcolithic (c. 5600-2100 cal B.C.). Vegetation History and Archaeo-botany, in press (electronic version).

SCHMIDL, A., JACOMET, S. and OEGGL, K. 200� - Distribution patterns of cultivated plants in the Eastern Alps (Central Europe) during the Iron Age. Journal of Archaeological Science, 34: 243-254.

SCHMIDL, A. and OEGGL, K. 200� - Ernährung und Wirtschaftweise der Siedler am Ganglegg während der Bronze- und Eisenzeit. Paläoethnobotanische Untersuchungen. In STEINER, H. (ed.) die Befestigte Siedlung am gang-legg im Vinschgau - Südtirol. Forschungen zur denkmalpflege in Südtirol, 3: 511-589. Bozen.

SCHWEINGRUBER, F.H. 1990 - Anatomie europäischer Hölzer. Haupt, Stuttgart.

SEIWALD, A. 1980 - Beiträge zur Vegetationsgeschichte Tirols IV: Natzer Plateau - Villanderer Alm. Berichte des Naturwissenschaftlich-medizinischen Vereines in Innsbruck, 6�: 31-�2.

STOKES, P. and ROWLEy-CONWy, P. 2002 - Iron Age Cultigen? Experimental Return Rates for Fat Hen (Chenopodium album L.). Environmental Archaeology, �: 95-99.

STUMBöCK, M. 1999 - die spät und postglaziale Vegetationsgeschichte des nordwestlichen Südtirols, mit einem Beitrag zur Verküpfung natürlicher Archive mit historischen Quellen. Dissertationes Botanicae, 299. Berlin-Stuttgart.

120 -

SWIDRAK, I and OEGGL, K. 1998 - Palaeoethnobotanische Untersuchungen von Bodenproben aus der bronzezeitli-chen Siedlung von Sotćiastel. In TECCHIATI, U. (ed.) Sotćiastel. Un abitato fortificato dell’età del bronzo in Val Badia. Istitut Cultural Ladin Micurá de Rü e Soprintendenza Provinciale ai Beni Culturali di Bolzano, Bolzano: 335-3�1.

WAHLMüLLER, N. 1990 - Spät und postglaziale Vegetationsgeschichte des Tschöggelberges (Südtirol). Berichte des Naturwissenschaftlich-medizinischen Vereines in Innsbruck, ��: �-16.

ZANDò, N. 2002 - La capanna dell’età del Rame di Villandro-Plunacker. Atti della XXXIII Riunione Scientifica dell’IIPP: 339-342.

Author’s Address:RENATO NISBET, Viale Rimembranza � – I - 10066 TORRE PELLICE (TO)e-mail: renisbetin@tin.it

- 121

Appendix 1. Charcoal analysis US 22, sq. D-Q/8-9, wall collapse Recent Early Medieval Picea/larix 20 Picea abies 5

US 22, House A, outside room H Recent Early Medieval Picea/larix 34 Picea abies 11

US 59, sq. QRS/0-1, House A, room N, floor room N Recent Early Medieval Pinus cembra 6 Pinus sylvestris/mugo 20 Picea/larix 9 Picea abies 8 Alnus sp. 2 Corylus avellana 9 Juglans regia 1

US �8, sq. S-T/6-�-8, outside SW House A, room A Recent Early Medieval Picea/larix 21 Picea abies 1� Quercus cad. 26 Juglans regia 5

US 115, lower cut, House A, room F Recent Early Medieval Pinus cembra 3 Picea/larix 2 Betula sp. 2 Quercus (sect. Robur) 1 Ulmus sp. 1 Prunus sp. 4

US 115, I-H/0-3 House A, room F, stone foundation Recent Early Medieval Picea/larix 8 Quercus cad. 2 Juglans regia 43

US 121, House A, room D, floor Recent Early Medieval Picea/larix 12 Corylus avellana 4 Quercus cad. 10 Ulmus sp. 1 Maloideae 2 Prunus sp. 3 Juglans regia 4

US 182, sq. M-L/19-20, House B Recent Early Medieval Picea/larix 44 Picea abies 28 Corylus avellana 26 Quercus cad. 1 Ulmus sp. 1 Fraxinus sp. 1 Maloideae 2 US 204, sq. L-M/1�-18, House B, hut B3, floor base, Recent Early Medieval Picea/larix 9large pit with pallet Picea abies 6 Betula sp. 1 Corylus avellana 51 Quercus cad. 1

subtotal 467

122 -

US 125, sq. L-M/0, House A, room L (hearth US 194) Intermediate Early Medieval Pinus sylvestris/mugo 4 US 186, sq. R/0-1, House A, room N, hearth B Intermediate Early Medieval Pinus cembra 1 Pinus sylvestris/mugo 6 Picea/larix 11 Picea abies 22 Betula sp. 5 Alnus sp. 6 Fagus sylvatica 1 Corylus avellana 1 Quercus cad. 6 Fraxinus sp. 1 Maloideae 6 US 194, sq. M/1, House A, room L, hearth Intermediate Early Medieval Pinus sylvestris/mugo 2 Picea/larix 2 Betula sp. 1 Alnus sp. 6 Corylus avellana 3 Quercus cad. 65 Juglans regia 6 US 220, sq. V-Z/6-8, House D, floor Intermediate Early Medieval Picea/larix 32 Picea abies 28 Betula sp. 1 Fagus sylvatica 1 Corylus avellana 2 Quercus cad. 4 Salix/Populus 1 Ulmus 2 Maloideae 4 Prunus sp. 9 Juglans regia 4 US 221, lower cut, sq. AE-AD-AC/�, room Q Intermediate Early Medieval Pinus sylvestris/mugo 1 Picea/larix 11 Picea abies � Quercus cad. 1 Cornus mas 1 US 221, sq. AA-AE, 16/18 House D under the Trench, Intermediate Early Medieval Picea/larix 6floor phase B, room Q Picea abies 3 Betula sp. 1 Fagus 1 Quercus cad. 3 Maloideae 5 Prunus sp. 2 US 231, sq. z/6, House D, room y, hearth near the wall Intermediate Early Medieval Pinus cembra 2 Pinus sylvestris/mugo 12 Picea/larix 10 Picea abies 8

- 123

Fagus sylvatica 1 Quercus cad. 4 Acer sp. 2 subtotal 324

US 26, House A, floor room A Late Roman, end 4th-5th centuries Picea/larix 2�6 US 52, sq. P-Q/19-20, Trench A Late Roman, end 4th-5th centuries Picea excelsa 85 US 53, sq. Q-R/20, House B Late Roman, end 4th-5th centuries Picea/larix 2� Betula sp. 6 Alnus sp. 1 Corylus avellana 14 Quercus cad. 1 Juglans regia 6 US 189, sq. P-O 0/1, House A, room L (under floor phase C) Late Roman, end 4th-5th centuries Picea/larix 13� US 189, sq. 0/1, West House A, room L, earthen floor Late Roman, end 4th-5th centuries Picea excelsa 110 Alnus sp. 3 US 206, sq. O-P/2, House A, room L Late Roman, end 4th-5th centuries Pinus sylvestris/mugo 2 Picea/larix 1 Betula 3 US 208, sq. N/2-3, House A, room L Late Roman, end 4th-5th centuries Pinus sylvestris/mugo � Picea/larix 9 Picea excelsa 21 Betula 4 Corylus avellana 1 Quercus cad. 2 Acer sp. 1 Fraxinus sp. 3 Maloideae 1 US 210, sq.ore Z/1, zone M Late Roman, end 4th-5th centuries Picea/larix 3 Quercus cad. 4 subtotal 728

124 -

US 82, U-T/9, zone R, pit bottom between House D and House A Roman-Early Middle Empire Picea/larix 40 Betula 5 US 82, sq. U-T/9, zone R, between House D and House A Roman-Early Middle Empire Picea/larix 14 US 190, sq. QRS/0-1-2-3, House A, room N Roman-Early Middle Empire Picea/larix 92 Picea excelsa 92 US 21 2/2, sq. O/1-2, House A, room L, hearth surface Roman-Early Middle Empire Picea/larix 15 Quercus cad. 5 Fraxinus sp. 1 US 212/2, sq. O/1-2, House A, room L Roman-Early Middle Empire Picea/larix 4 Alnus sp. 2 Juglans regia 3 US 216, sq. AB/1, zone M1 Roman-Early Middle Empire Picea/larix 6 subtotal 279

US 214, cut I, sq. T-AE/2-3, zone M, House F Roman middle 1st-middle 2nd century Pinus cembra 5 Pinus sylvestris/mugo 48 Picea/larix 20 Picea abies 22 Betula sp. 1 Fagus sylvatica 1 Quercus cad. 35 Prunus sp. 2 Maloideae 1 subtotal 135

US 25�, Iron age hut Late Iron Age, 1st century BC Pinus cembra 9 Pinus sylvestris/mugo 42 Betula sp. 2 Alnus sp. 2 Corylus avellana 3 Quercus cad. 45 Ulmus sp. 2 Fraxinus sp. 1 Maloideae 11 Salix/Populus 2 subtotal 119

- 125

US 180, P-Q/13-14, Trench A, hut floor Copper Age Picea/larix 1 Betula sp. 1 Quercus cad. 3 Maloideae 3 subtotal 8

US 30, cut 1, sq. D/2, Trench C Neolithic Picea excelsa 1 Alnus sp. 1 Quercus cad. 10 US 30, cut 2, sq. E/4, Trench C Neolithic Picea excelsa 3 Carpinus/ostrya 3 Quercus cad. 5 US 30, cut 3, sq. 39, Trench C Neolithic Picea/larix 1 Picea excelsa 3 Alnus sp. � Carpinus/ostrya 1 Quercus cad. 34 Prunus sp. 2 US 30, cut 4, sq. 40, Trench C Neolithic Alnus sp. 4 Quercus cad. 29 Populus sp. 4 US 30, cut 5, sq. 41, Trench C Neolithic Alnus sp. 3 Carpinus/ostrya 1 Quercus cad. 9 Acer sp. 2 US 30, cut 6, sq. 13, Trench C Neolithic Picea/larix 2 Picea excelsa 2 Alnus sp. 2 Carpinus/ostrya 10 Quercus cad. 8 Fraxinus sp. 1 US 30, cut �, sq. 40, Trench C Neolithic Alnus sp. 15 Carpinus/ostrya 6 Quercus cad. 18 Fraxinus sp. 3 Prunus cf. avium 2

126 -

US 30 cut 8, sq. 42, Trench C Neolithic Picea/larix 6 Picea excelsa � Alnus sp. 9 Corylus avellana 2 Quercus cad. 40 US 30, cut 9, sq. F/5, Trench C Neolithic Picea/larix 1 Picea excelsa 2 Alnus sp. 2 Quercus cad. 4 Corylus avellana 2 Carpinus/ostrya 1 US 30, cut 9, sq. 44, Trench C Neolithic Alnus sp. 8 Carpinus/ostrya 1 Corylus avellana 22 Quercus cad. 12 US 80, cut 1, sq. 42, Trench C Neolithic Picea/larix 5 Picea excelsa 6 Alnus sp. 13 Carpinus/ostrya 1 Quercus cad. 15 Ulmus sp. 1 US 80, sq. 15-1�, Trench C Neolithic Picea/larix 4 Picea excelsa 2 Alnus sp. � Carpinus/ostrya 2 Quercus cad. 10 US 151, sq. A6-B6, Trench C, hearth Neolithic Picea/larix 1 Picea excelsa 1 Alnus sp. 2 Quercus cad. 43 subtotal 424

US 325, sq. AB/B4, plinth � Mesolithic Pinus sylvestris/mugo 1 Betula 4 Quercus cad. 6 US 325, zona M, plinth � Mesolithic Pinus sylvestris/mugo 1 Betula sp. 12 Quercus cad. 4

subtotal 28

- 12�

Appendix 2. Seeds and fruits US 59 Recent Early Medieval Triticum aestivum s.l. 2 Hordeum vulgare 1 Avena sp. 1 US 115 Recent Early Medieval Hordeum vulgare (hulled) 1 Vitis vinifera 2 galium cf. aparine 2 US 121 Recent Early Medieval Triticum aestivum s.l. 2 Hordeum vulgare 2 Secale cereale 2 US 182 Intermediate Early Medieval Hordeum vulgare 2 Secale cereale 1 Setaria viridis/vert/italica 1 Vicia sp. 1 Vitis vinifera 1 US 204 Intermediate Early Medieval Secale cereale 1

US 186 Intermediate Early Medieval Triticum aestivum s.l. 4 Hordeum vulgare (hulled) 3 Secale cereale 1 Triticum/Secale 1 Bromus secalinus/hordaceus 1 Fallopia convolvulus 1 Ranunculus sp. 2

US 125 Intermediate Early Medieval Hordeum vulgare (hulled) 1

US 221 Intermediate Early Medieval Triticum aestivum s.l. 2 Hordeum vulgare hexastichum 2 lens esculenta 1 Indeterminated 2 US 52, sq. P-Q/19-20, Trench A Late Roman, end 4th-5th centuries Hordeum vulgare 2 Cerealia 3

US 53, sq. Q-R/20, House B Late Roman, end 4th-5th centuries Hordeum vulgare hexastichum 1 cf. Hordeum vulgare 1 Cerealia � Setaria viridis/verticillata/italica 2 Chenopodium hybridum 1 galium cf. aparine 5 Rumex sp. 1

128 -

US 208, sq. N/2-3, House A, room L Late Roman, end 4th-5th centuries Triticum sp. 2 Hordeum vulgare hexastichum 6 lens esculenta 1 galium sp. 1 Vicia tetrasperma 1 US 82, u-T/9, zone R, pit bottom Early Middle Roman Empire Triticum aestivum s.l. 45 Hordeum vulgare 4 Secale cereale 2 US 190, sq. QRS/0-1-2-3, House A, room N Early Middle Roman Empire Triticum dicoccum, grain 114 Triticum dicoccum, glumes 31 Triticum aestivum s.l. 90 Triticum sp. 24 Hordeum vulgare 6504 Secale cereale 8 Avena sp. 5 Vicia faba 130 Pisum sp. 3 Vitis vinifera 10 Juglans regia 29 Prunus cf. avium 1 Pyrus communis 1 Bromus secalinus/hordaceus 1 Carex sp. 10 Echinochloa crus-galli 1 galeopsis tetrahit 1 lolium temulentum 3 melilotus cf. dentata 1 Polygonum dumetorum 3 Polygonum lapathifolium 3 Sambucus nigra 63 Setaria viridis/verticillata/italica 32 Veronica hederaefolia 3� Vicia sp. 18 Vicia tetrasperma 4 Vicia/lathyrus 11 US 21, 2/2 sq. O/1-2, House A, room L Early Middle Roman Empire Hordeum vulgare 1 US 216, sq. AB/1, Oven M1 Early Middle Roman Empire Triticum dicoccum (one grain spikelet) 2 Triticum dicoccum 1018 Triticum dicoccum, glumes 460 Triticum dicoccum, forks 510 Triticum aestivum s.l./durum/turgidum 264 Triticum spelta, grain 26 Triticum spelta, forks 25 Triticum spelta, glume 5 Triticum sp. 424 Hordeum vulgare 22 Hordeum vulgare (hulled) 22 Secale cereale 10

- 129

Secale/Triticum 21 Panicum miliaceum 1 Avena sp. 4 Cerealia 1046 Picea excelsa (needles) 3 Bromus secalinus/hordaceus 3 Cerastium/Stellaria 1 Chenopodium polyspermum 1 Composite, achenes 1 Echinochloa crus-galli 2 galium cf. aparine 9 Papaver sp. 2 Setaria viridis/verticillata/italica 6 Vicia cracca/villosa 3 Viola arvensis/tricolor 2 US 214 middle 1st-middle 2nd centuries AD Triticum dicoccum 46 Triticum aestivum s.l. 8 Triticum spelta, glumes 3 Triticum spelta, glume-base 2 Triticum sp. 53 Hordeum vulgare 3 Secale cereale 2 Avena sp. 1 Vitis vinifera 2 Echinochloa crusgalli 1 galium cf. aparine 8 Stachys annua/arvensis 1 Vicia tetrasperma 1 US 25� Late Iron Age, 1st century BC Prunus spinosa 1 US 180 Copper Age Hordeum vulgare 2 Panicum miliaceum 2 Setaria viridis/verticillata/italica 2 US 30, cut 1, sq. D/2, Trench C Neolithic Hordeum vulgare 8 Chenopodium album � galium cf. aparine 1 US 30, cut 2, sq. E/4, Trench C Neolithic Triticum aestivum s.l. 1 Chenopodium album 6 Veronica hederaefolia 1 US 30, cut 3, sq. 39, Trench C Neolithic Hordeum vulgare 1 Cerealia 1 Chenopodium album 1 Sambucus ebulus 3 Veronica hederaefolia 1 US 30, cut 4, sq. 40, Trench C Neolithic Cerealia 1 Cerastium arvense 1

130 -

Sambucus ebulus 1 Indeterminated 2 US 30, cut 5, sq. 41, Trench C Neolithic Cerealia 2 Cerastium arvense 2 Sambucus ebulus 3 Veronica hederaefolia 1 US 30, cut 6, sq. 13, Trench C Neolithic Triticum dicoccum 1 Triticum aestivum s.l. � Hordeum vulgare 8 Cerealia 49 lolium temulentum 3 Polygonum persicaria 2 Rumex crispus 16 Sambucus ebulus 13 Veronica hederaefolia 2 Vicia sativa 1 Vicia/lathyrus 1 Indeterminated 3 US 30, cut �, sq. 40, Trench C Neolithic Triticum aestivum s.l. 2 Hordeum vulgare 8� Fragaria vesca 1 Rubus idaeus 2 Sambucus ebulus 28 Veronica hederaefolia 2 US 30, cut 8, sq. 42, Trench C Neolithic Triticum aestivum s.l. 2 Hordeum vulgare 21 Cerealia 65 Chenopodium album 4 Echinocloa crusgalli 1 Euphorbia helioscopia 2 galium cf. aparine 4 lolium temulentum 4 medicago lupulina 5 Quercus sp. (acorn) 5 Rubus sp. 2 Sambucus ebulus 42 Veronica hederaefolia 8 Vicia sp. 3 US 30, cut 9, sq. F/5 Neolithic Triticum aestivum s.l. 1 Hordeum vulgare 52 Cerealia 82 Bromus secalinus v. hordaceus 3 Sambucus ebulus 3 Veronica hederaefolia 2 US 30, cut 9, sq. 44, Trench C Neolithic Hordeum vulgare 9 Sambucus ebulus 14

- 131

Fragaria vesca 1 US 80, cut 1, sq. 42, Trench C Neolithic Vitis vinifera 3 Veronica hederaefolia 1 Triticum dicoccum 1 Hordeum vulgare 53 Cerealia �5 galeopsis tetrahit 1 galium cf. aparine 1 lolium temulentum 5 Polygonum lapathifolium 1 Sambucus ebulus 9 Vicia sp. 1 US 80, sq. 15-1�, Trench C, witness Neolithic Hordeum vulgare 3 Cerealia 15 Veronica hederaefolia 1 US 151, sq. A6-B6, Trench C, hearth 151 Neolithic Triticum dicoccum 8 Triticum sp. 42 Hordeum vulgare 4 Cerealia 5 Vitis vinifera 2

- 133

ALBERTO GIROD* e GIULIA RICCARDINO**

VALUtAZionE DEL ContRibUto Di Cerastoderma glaucum (PoiREt, 1789) ALLA DiEtA DELLE PoPoLAZioni PREiStoRiCHE DEL Sito Di MontE D’ACCoDDi

(PoRto toRRES, SARDEgnA)

la mesure précise d’un fait imprécis ne se traduit jamaispar un gain en précision, mais en imprécision

CHENORKIAN, 1992

SUMMARY – Evaluation of the contribution of Cerastoderma glaucum (PoIRET, 1789) to the diet of the pre-historic populations of monte d’Accoddi (Porto Torres, Sardinia). The excavations carried out between 19�9 and 2000 in the area that surrounds the prehistoric shrine at Monte d’Accoddi have always brought to light large quantities of marine shells. The dominant species is Cerastoderma glaucum. On the basis of the relationships between dimensions and fresh weight of the individuals of this species living in five Italian lagoons, the weight of the flesh contained in the valves of the archaeological site was estimated. This method, which is rapid, although not necessarily precise, and useful only for defining general trends, has in any case favoured an average possible evaluation of the availability of protein and energy that C. glaucum offered to the prehistoric populations around the shrine. A modest nutritional content has emerged. This indicates that both C. glaucum, and the other sea mol-luscs, represented only an integrative part of the diet.

RiASSUnto – Valutazione del contributo di Cerastoderma glaucum (PoIRET, 1789) alla dieta delle popolazioni preistoriche del sito di monte d’Accoddi (Porto Torres, Sardegna). Gli scavi eseguiti fra il 19�9 e il 2000 nell’area che circonda il santuario preistorico di Monte d’Accoddi, hanno messo in luce delle grandi quantità di conchiglie marine. La specie dominante è Cerastoderma glaucum. In base alla relazione che intercorre tra le dimensioni e il peso a fresco degli individui della stessa specie viventi in cinque lagune italiane, è stato stimato il peso della carne contenuta nelle valve del sito archeologico. Questo metodo, rapido ma non necessariamente preciso, utile a definire delle tendenze generali, ha comunque consentito una valutazione di massima della disponibilità proteica ed ener-getica che C. glaucum poteva offrire ai frequentatori preistorici del santuario. Sono risultati degli apporti nutrizionali molto modesti, che fanno supporre che C. glaucum, e gli altri molluschi marini, costituissero solo un elemento integrativo della dieta alimentare degli abitanti.

Keywords: Sardinia, Monte d’Accoddi, marine molluscs, prehistoric diet, nutritional content.

Atti Soc. Preist. Protost. Friuli-V.G., Trieste, XVI, 2006-200� (2008): 133-146

______________________________________

** Laboratorio di Malacologia Applicata, Milano, I** Monza, Milano, I

134 -

iL Sito

Il santuario di Monte d’Accoddi si trova nella pianu-ra Turritana, circa 11 km a N-W di Sassari, su di un altipiano, a circa �0 m di altezza, poco a valle della cima del Monte La Corra (�5 m), nei pressi del Riu d’Ottava (fig. 1).

Le campagne di scavo che si sono susseguite dal 1952 al 2000 hanno esposto, in tutta l’area d’interesse archeologico, estesi accumuli conchigliari risalenti al periodo di frequen-tazione del sito, nei periodi culturali sotto indicati, che rap-presentano resti di pasto. Tra questi, la specie predominante, è Cerastoderma glaucum (POIRET, 1�89), un mollusco bivalve lagunare raccolto, mol-

to probabilmente, nella vicina laguna di Platamona che si trova ad est del sito (GIROD e NO-VATI, 2004).

LA RiCERCA

Dei quesiti sorti nel corso degli ultimi scavi, restava irrisolto quello sul valore alimentare e sulla disponibilità di carne che questi molluschi potevano fornire ai frequentatori del sito.

Per tentare una risposta ci si è innanzi tutto avvalsi delle centinaia di biometrie effettu-ate sulle valve di C. glaucum raccolte nelle campagne di scavo del 1995 e 2000 (BALZARINI, 2001-2002; GIROD e NOVATI, 2004). In seguito si sono misurate l’altezza, un parametro indica-tivo della crescita dei bivalvi (CHENORKIAN, 1992, 1996; MORETEAU, 1995) già usato in prec-edenti lavori su detta specie di Monte d’Accoddi, il peso fresco e il peso secco di individui di C. glaucum provenienti da cinque diverse lagune: Palude Maggiore (Venezia), Orbetello (Grosseto), Sacca di Goro (Chioggia), Santa Gilla (Cagliari) e Calich (Sassari) (fig. 2). Con un’analisi di regressione lineare1 si è stabilita l’esistenza di una relazione in grado di legare tra loro i parametri rilevati. Nel caso di C. glaucum, tale relazione è del tipo:

peso fresco / secco = intercetta + (P1*altezza) + [P2*(altezza)2]

______________________________________

1 Il programma statistico utilizzato per l’analisi di regressione lineare è S-Plus 2000, Copyright Math Soft.

Fig. 1 - Ubicazione di Monte d’Accoddi nella regione nord occidentale della Sardegna.

- 135

Fig. 2 - A) regressione lineare tra altezza e peso fresco degli individui di C. glaucum raccolti nelle cinque lagune; B) regressione lineare tra altezza e peso secco degli individui di C. glaucum raccolti nelle cinque lagune.

136 -

(in cui intercetta, P1 e P2 sono i coefficienti derivati dall’analisi di regressione lineare, come da tab. 1); relazioni di questo tipo non sempre si riscontrano in altre specie marine, quali ad esempio, Rangia cuneata (CLAASSEN, 1998).

Peso fresco

Coefficienti P - value R - quadrato

P1 P2 Intercetta

-0,063 0,004 0,368 0 0,844

Peso secco

Coefficienti P - value R - quadrato

P1 P2 Intercetta

-0,020 0,001 0,18� 0 0,424

Tab. 1 - Quadro riassuntivo della regressione lineare tra altezza e peso fresco e secco. P – value 0: presenza di correlazione tra due variabili: R quadrato 1: errore commesso piccolo. Il programma statistico indica 0 quando ci sono più di quattro zeri dopo la virgola.

AREE Di SCAVo DA CUi PRoVEngono i REPERti MALACoLogiCi

Durante la campagna del 1995 è stato esteso lo scavo aperto nel 1989 a SW del santuario,

corrispondente alla zona dell’abitato (TINÉ e TRAVERSO, 1992; TRAVERSO, 1996-199�; GIROD e BALZARINI, 2004-2005), dove sono stati individuati i quattro orizzonti corrispondenti ai quattro principali periodi di frequentazione: 1: storico, Bronzo, Rame; 2: Cultura di Filigosa; 3: Cultura di Ozieri; 4: Neolitico recente; al di sotto del quale è stato rinvenuto il substrato roccioso.

Nel 2000 lo scavo SW è stato ulteriormente ampliato portandolo, dai 36m2 originari, a 45m2; è stata poi aperta una nuova trincea di 8,5m2 nella zona a NE del santuario. La strati-grafia in questo punto risulta parzialmente danneggiata a causa della presenza di buche, tal-volta estese e profonde, di origine antropica, aperte probabilmente dai frequentatori del sito allo scopo di gettarvi resti di pasto (fig. 3).

Stima dei pesi dei molluschi contenuti nelle valve di epoca preistoricaConoscendo la relazione che intercorre, negli individui raccolti vivi, tra l’altezza e il

peso fresco e l’altezza delle conchiglie raccolte durante le campagne di scavi del 1995 e del 2000, è stato possibile stimare il peso della carne contenuta all’interno delle valve recuperate nel sito preistorico. Questa stima tiene conto anche della percentuale di valve che non è stata possibile analizzare, perché danneggiate, e che rappresentano il 25% dei campioni raccolti nel 1995, e il 10% di quelli raccolti nel 2000.

Il calcolo effettuato riguarda gli strati dello scavo SW e i tagli e le buche considerate nel loro contenuto totale dello scavo NE; si riferisce pertanto a periodi d’accumulo molto lunghi, probabilmente di due millenni circa, a partire da 49�0±100 uncal BP (Utc-146�) (TINÉ e TRAVERSO, 1992; TRAVERSO, 1996-199�). Come si può notare dalla tab. 2, l’insieme di valve di C. glaucum finora recuperate a Monte d’Accoddi, su di una superficie comples-

- 13�

Fig. 3 - Perimetro del santuario di Monte d’Accoddi, posizione degli scavi SW e NE con le coordinate dei quadrati di scavo, e la sequenza delle fasi di frequentazione.

siva di 53,5m2, esprime una modestissima quantità di carne e quindi di valore nutrizionale, nonostante sia il risultato di attività alimentari protrattesi per lungo tempo, durante i quattro periodi di frequentazione sopra elencati.

Stima della disponibilità proteica ed energetica che C. glaucum poteva offrire ai fre-quentatori del sito

Sulla base di quanto si conosce dei valori nutrizionali di alcuni molluschi bivalvi marini (GEIGy, 1983; SECCHI, 1990; HOLLAND et al., 1993; CLASSEN, 1998; SOUCI et al., 2000) (tab. 3), è stato possibile calcolare una stima della disponibilità proteica ed ener-getica che i molluschi contenuti nelle valve rinvenute a Monte d’Accoddi potevano offrire ad un individuo (maschio o femmina) (tab. 4).

I molluschi, rispetto alle altre specie animali, sono un alimento povero di grassi (circa 2%), vitamine e oligoelementi; tuttavia sono ricchi di acqua (circa 80%), proteine (circa 5-15%), minerali (soprattutto Ca, P, Fe, Na, K) e idrati di carbonio (assenti in altri cibi animali) (tab. 3).

In mancanza di dati sul C. glaucum, i calcoli sono stati eseguiti prendendo come riferi-mento i valori nutrizionali di una specie molto affine, Cerastoderma edule (LINNÉ, 1�58). Sapendo che la quantità di proteine e di energia contenute in 100 gr di parte edule di C. edule è rispettivamente di 11,0-13,2 gr e 48,0 kcal (CLAASSEN, 1998) si sono fatte le stime utiliz-zando le seguenti proporzioni:

11,0 - 13,2 gr di proteine : 100 gr di parte edule = X gr di proteine : X gr di parte edule(X gr di parte edule = gr di C. glaucum stimati in ogni singolo taglio e livello);

48,0 kcal : 100 gr di parte edule = X kcal : X gr di parte edule.(X gr di parte edule = g di C. glaucum stimati in ogni singolo taglio e livello).

138 -

biometrie 1995 Numero di valve (100%)

Individui / m2 Stima dei pesi totali (100%)

Livelli (SW)

2 2�12 80 3156,5 g

3 2596 45 2440,3 g

4 269� �1 2018, g

biometrie 2000 Tagli (NE)

I 1�3 44 113,3 g

Buca I 221 111 1�0,9 g

II 494 62 298,0 g

Buca 1230 238 30 180,2 g

Buca 1241 220 110 151,9 g

Buca 1245 248 62 162,2 g

Buca 1246 218 109 149,9 g

IV 204 102 205,1 g

VI 123 62 113,0 g

Buca VI 160 80 129,2 g

VII 126 63 104,4 g

Buca VII 86 43 60,1 g

VIII 213 10� 153,3 g

Livelli (SW)

4 �93 66 1024,6 g

5 934 �8 949,5 g

6 91 23 4�,4 g

Tabella 2 - Stima dei pesi totali al 100% dei C. glaucum in ogni livello del settore SW e taglio del settore NE. La media di individui al m2 è calcolata sui quadrati di 1 m2 da cui realmente provengono le valve, poiché non tutti hanno fornito materiale.

L’apporto proteico ed energetico risulta piuttosto esiguo; questo dato riflette il modestis-simo valore nutrizionale della carne contenuta in C. glaucum.

Stima della durata della disponibilità proteica ed energetica che C. glaucum offriva ai frequentatori del sito

In base alle richieste energetiche e proteiche giornaliere degli individui riportate nella tabella 5, mediante il seguente calcolo:

r.e.g. : 24 ore = d.e.o. : X ore(r.e.g. = richiesta energetica giornaliera)

(d.e.o. = disponibilità energetica offerta da Cerastoderma in ogni taglio e livello)si è valutata la durata della disponibilità proteica ed energetica messa a disposizione dei frequentatori del sito (tab. 6). Dai calcoli è emerso che l’apporto energetico, da parte di

- 139

C. glaucum, poteva consentire ad una donna di svolgere un’attività fisica moderata per un massimo di 23 ore, mentre l’apporto energetico per un’attività intensa è di minore durata, 19 ore al massimo.

Per un uomo, invece, la durata della disponibilità energetica tende ad esaurirsi più in fretta: sedici ore per un’attività moderata e 14 ore per un’attività intensa. Prendendo invece come raffronto degli agricoltori bulgari del 5000 BC (CHENORKIAN, 1989), l’apporto ener-getico sarebbe stato tale da soddisfarne i bisogni energetici giornalieri per poco più di mezza giornata (tab. �).

È ragionevole pensare che, nell’area del santuario, si radunassero più persone dello stesso gruppo familiare. Pertanto si sono considerati gruppi ipotetici costituiti da 3 e da 10 individui (tab. 8 e 9). I due gruppi sono così costituiti: il primo da 1 maschio, 1 femmina e 1 bambino di �-8 anni; il secondo da 4 maschi, 4 femmine e 2 bambini di �-8 anni.

Specie AutoriAcqua

(g)Proteine

(g)grassi

(g)Carboi-drati (g)

Energia (kcal)

Cardidi HOLLAND et al., 1993 83 12 0,6 Tracce 53

Cerastoderma edule CLAASSEN, 1998 11,0 – 13,2 0,3 3,4 48

Mitili HOLLAND et al., 1993 80,9 12,1 1,8 2,5 �4

mya arenaria SOUCI et al., 2000 83,1 10,5 1,3 54

mytilus edulis CLAASSEN, 1998 8,9 – 12,2 1,� – 2,0 2,9 8�

mytilus edulis GEIGy, 1983 82,5 12 1,� 2,2 �6

mytilus edulis SOUCI et al., 2000 83,2 10,2 1,96 2,3� 68

mytilus galloprovin-cialis

SECCHI, 1990 83 10 1,4 53

mytilus sp. CLAASSEN, 1998 14,4 2,2 3,3 95

ostrea edulis CLAASSEN, 1998 8,6 – 13,1 0,9 – 1,9 5,9 59

ostea edulis SOUCI et al., 2000 83 9 1,2 4,8 66

ostrea sp. HOLLAND et al., 1993 85,� 10,8 1,3 2,� 65

ostrea sp. GEIGy, 1983 83 9 1,2 4,8 68

ostrea sp. SECCHI, 1990 83 9 1,2 5 6�

Pecten jacobaeus SECCHI, 1990 80 15 0,1 60

Pecten maximus CLAASSEN, 1998 1�,5 0,1 105

Pecten sp. GEIGy, 1983 �9,8 15,3 0,2 3,3 �9

Pecten sp. SOUCI et al., 2000 80 15,6 0,1 63

Vongole SECCHI, 1990 84 10,1� 2,53 2,24 68,4

Tabella 3 - Valori nutrizionali di molluschi bivalvi marini per 100 g di parte edule.

140 -

La durata della disponibilità proteica varia da un minimo di un’ora ad un massimo di 2-3 giorni per un gruppo costituito da 3 individui; mentre è più breve per un gruppo formato da 10 individui: da un minimo di 13-16 minuti ad un massimo di 15-18 ore. La durata della disponibilità energetica è più breve per entrambe i gruppi: da 5 minuti a 5 ore per un gruppo di 3 individui e da 1 minuto a 2 ore per un gruppo di 10 individui (tab. 8). Le durate delle disponibilità proteiche ed energetiche sono tali da non riuscire a soddisfare i fabbisogni proteici ed energetici di entrambe i gruppi in caso di attività moderata o intensa (tab. 9).

biometrie 1995

Stima dei pesi totali

(100%) (g)

Disponibilità proteica (g)

Disponibilitàenergetica (kcal)

Livelli (SW) Maschio-Femmina Maschio-Femmina

2 3156,5 34�,2 – 416,6 1515,1

3 2440,3 268,4 – 322,1 11�1,3

4 2018 222 – 266,4 968,9

biometrie 2000

Tagli (NE)

I 113,3 12,4 – 14,9 54,4

I Buca 1�0,9 18,8 – 22,5 82

II 298 32,� – 39,3 143

Buca 1230 180,2 19,8 – 23,� 86,5

Buca 1241 151,9 16,� – 20 �2,9

Buca 1245 162,2 1�,8 – 21,4 ��,8

Buca 1246 149,9 16,5 – 19,8 �1,9

IV 205,1 22,5 – 2� 98,4

VI 113 12,4 – 14,9 54,2

VI Buca 129,2 14,2 – 1� 62

VII 104,4 11,4 – 13,� 50,1

VII Buca 60,1 6,2 – �,9 28,8

VIII 153,3 16,8 – 20,2 �3,6

Livelli (SW)

4 1024,6 112,� – 135,2 491,8

5 949,5 104,4 – 125,3 455,8

6 4�,4 5,2 – 6,26 22,�

Tabella 4 - Stima rapportata a C. edule della disponibilità proteica ed energetica per un maschio e una femmina.

- 141

ConCLUSioni

La metodologia adottata è stata rapida, e non necessariamente precisa, ma ha consentito una valutazione di massima del fenomeno, studiato fornendo informazioni sulle tendenze generali.

Si può affermare che, nel caso di Monte d’Accoddi e in considerazione dei modesti apporti proteici nonché calorici, C. glaucum veniva utilizzato come elemento integrativo della dieta alimentare. Sembra inoltre che il consumo non avvenisse in modo regolare e con-tinuativo, bensì solo in determinati periodi, come risulta dallo studio di GIROD e BALZARINI (2005). Occorre anche ricordare che, nel caso di Monte d’Accoddi, pur trattandosi di residui di pasto, la quantità di conchiglie accumulate è certamente esigua se rapportata ai tempi di deposizione molto lunghi (due millenni circa). In questo caso è corretto parlare di semplici accumuli conchigliari, residui di pasto.

bisogni nutrizionali umani

bambini Maschi Femmine

Proteine (g) 40 56 44

Grassi (g) 25 25

Energia (kcal) 2200 2�00 2000

Attività (kcal / giorno) *

Moderata * 2300 * 1600 *

Intensa * 2�00 * 1880 *

Agricoltori bulgari 5000 BP **

2500 ** 2500 **

Vitamine solubili (mg)

Tiamina 1,4 1

Riboflavina 1,6 1,2

Niacina 18 13

Acido ascorbico 60 60

Minerali (mg)

Ca 800 800

P 800 800

Fe 10 18

Na 10,3 10

Tab. 5 - Necessità caloriche umane (di individui attuali) (da CHENoRKIAN, 1989, con modificazioni).

142 -

biometrie 1995

Durata disponibilità proteica

Durata disponibilità energetica

Maschio Femmina Maschio Femmina

Livelli (SW) giorni ore . minuti

2 6 – � 8 – 9 13:4� 18:18

3 5 – 6 6 – � 10:41 14:06

4 4 – 5 5 – 6 9:01 12:03

biometrie 2000

Tagli (NE) ore . minuti

I 5.34 – 6.41 �.20 – 8.6 0:29 0:39

I Buca 8.06 – 10.0� 10.26 – 12.31 0:44 0:59

II 14.05 – 1�.26 18.29 – 21.46 1:2� 2:11

Buca 1230 8.50 – 10.20 11.22 – 13.38 0:46 1:04

Buca 1241 �.16 – 8.59 9.11 – 11.34 0:39 0:52

Buca 1245 8.50 – 9.18 10.14 – 12.08 0:41 0:56

Buca 1246 �.0� – 8.49 9.00 – 11.20 0:38 0:52

IV 10.0� – 12.01 12.31 – 15.1� 0:53 1:18

VI 5.33 – 6.39 �.18 – 8.14 0:29 0:39

VI Buca 6.09 – �.31 8.16 – 9.31 0:33 0:44

VII 5.32 – 6.31 6.2� – �.52 0:2� 0:36

VII Buca 3.24 – 3.41 4.01 – 4.32 0:16 0:21

VIII �.23 – 9.0� 9.20 – 11.04 0:39 0:53

Livelli (SW) giorni ore . minuti

4 2 3 4:3� 6:30

5 2 2 – 3 4:05 5:4�

ore . minuti

6 2.24 – 3.10 3.25 – 3.41 0:12 0:16

Tab. 6 - Stima rapportata a C. edule della durata (in ore . minuti oppure giorni) della disponibilità proteica ed ener-getica in ogni livello del settore SW e taglio del settore NE.

RingraziamentiSiamo grati ai Prof. A. Cau e Sig. M. Angius, dell’Università di Cagliari; al Sig. F. Cavolo, col-

laboratore presso l’Università di Venezia; al Dr. M. Lenzi, direttore delle Peschiere di Orbetello; al Prof. Serra dell’Università di Sassari e al Dr. E. Turolla, ricercatore presso l’istituto CRIM di ricerca sui molluschi di Chioggia, senza la cui collaborazione non sarebbe stato possibile ottenere esemplari di C. glaucum freschi. Un prezioso aiuto per l’analisi statistica ci è stato fornito dal Prof. G. Guerra, dell’Università degli Studi di Milano Bicocca. Si ringraziano inoltre il Dr. M. Mariani, Direttore della Stazione di Idrobiologia, Civico Acquario di Milano, per aver messo a nostra disposizione una parte delle attrezzature del laboratorio, e l’Istituto Italiano di Archeologia Sperimentale (IIAS) di Genova, per le informazioni riguardanti il sito e lo scavo del 2000.

- 143

biometrie 1995

Durata della dispo-nibilità energetica

in attività moderata

Durata della dispo-nibilità energetica in attività intensa

Agricol-tori

bulgariMaschio

Fem-mina

MaschioFem-mina

Livelli (SW) ore . minuti

2 16:21 23:13 13:4� 19:34 14:55

3 12:22 1�:5� 10:41 15:35 11:24

4 10:11 14:53 9:01 12:3� 9:30

biometrie 2000 Tagli (NE) ore . minuti

I 0:32 0:49 0:29 0:41 0:31

I Buca 0:52 1:23 0:44 1:05 0:4�

II 1:49 2:15 1:2� 2:23 1:3�

Buca 1230 0:54 1:30 0:46 1:10 0:50

Buca 1241 0:45 1:09 0:39 0:56 0:42

Buca 1245 0:49 1:1� 0:41 1:05 0:45

Buca 1246 0:45 1:08 0:38 0:55 0:41

IV 1:03 1:48 0:53 1:26 0:5�

VI 0:34 0:49 0:29 0:41 0:31

VI Buca 0:39 0:56 0:33 0:4� 0:36

VII 0:31 0:45 0:2� 0:38 0:29

VII Buca 0:18 0:26 0:16 0:22 0:1�

VIII 0:46 1:10 0:39 0:56 0:43

Livelli (SW) ore . minuti

4 5:13 �:38 4:3� 6:28 5:12

5 5:16 �:24 4:05 6:21 4:38

6 0:14 0:20 0:12 0:1� 0:13

Tab. � - Stima rapportata a C. edule della durata (in ore . minuti oppure giorni) della disponibilità proteica ed ener-getica per un maschio e una femmina di età compresa tra i 23 e 50 anni, di peso medio e che svolgano un’attività moderata o intensa, o che siano agricoltori.

144 -

biometrie 1995

Durata della disponibilità proteica

Durata della disponibilità energetica

tre individui

Dieci individui

treindividui

Dieci individui

Livelli (SW) giorni ore . minuti

2 2 – 3 15.28 – 18.26 5:2� 1:5�

3 2 11.0 – 14.21 4:0� 1:21

4 2 9.52 – 11.42 3:3� 1:00

biometrie 2000

Tagli (NE) ore . minuti

I 2.13 – 2.56 0.32 – 0.38 0:11 0:03

I Buca 3.22 – 4.26 0.48 – 0.58 0:1� 0:05

II 6.02 – �.14 1.40 – 2.09 0:30 0:09

Buca 1230 3.40 – 4.08 1.00 – 1.02 0:18 0:05

Buca 1241 3.26 – 3.44 0.43 – 0.51 0:15 0:04

Buca 1245 3.06 – 4.1� 0.46 – 0.55 0:16 0:05

Buca 1246 3.23 – 3.39 0.42 – 0.51 0:15 0:04

IV 4.2� – 5.04 0.58 – 1.00 0:21 0:06

VI 2.13 – 2.56 0.32 – 0.38 0:11 0:03

VI Buca 2.44 – 3.33 0.3� – 0.44 0:13 0:04

VII 2.3� – 2.36 0.30 – 0.35 0:10 0:03

VII Buca 1.13 – 1.36 0.1� – 0.20 0:06 0:02

VIII 3.29 – 3.4� 0.43 – 0.52 0:15 0:04

Livelli (SW) ore . minuti

4 19.32 – 23.19 5.23 – 6.20 2:11 0:31

5 18.31 – 21.49 4.48 – 5.3� 2:00 0:28

6 0.54 –1.00 0.13 – 0.16 0:05 0:01

Tab. 8 - Stima rapportata a C. edule della durata (in ore . minuti, oppure giorni) della disponibilità proteica ed ener-getica per due gruppi familiari di tre e dieci individui.

- 145

biometrie 1995

Durata della disponibilità proteica in attività

moderata

Durata della disponibilità energetica in attività

intensa

tre individui

Dieciindividui

treindividui

Dieciindividui

Livelli (SW) ore . minuti

2 6:36 2:21 5:36 2:00

3 5:00 1:41 4:15 1:24

4 4:21 1:16 3:43 1:02

biometrie 2000

Tagli (NE) ore . minuti

I 0:13 0:04 0:12 0:03

I Buca 0:19 0:06 0:1� 0:05

II 0:34 0:10 0:30 0:09

Buca 1230 0:20 0:06 0:18 0:05

Buca 1241 0:1� 0:05 0:15 0:05

Buca 1245 0:18 0:06 0:16 0:05

Buca 1246 0:1� 0:05 0:15 0:04

IV 0:23 0:0� 0:21 0:06

VI 0:13 0:04 0:11 0:03

VI Buca 0:15 0:04 0:13 0:04

VII 0:12 0:04 0:11 0:05

VII Buca 0:0� 0:02 0:06 0:02

VIII 0:1� 0:05 0:16 0:05

Livelli (SW) ore . minuti

4 2:34 0:35 2:14 0:31

5 2:20 0:32 2:02 0:29

6 0:05 0:02 0:05 0:01

Tab. 9 - Stima rapportata a C. edule della durata (in ore . minuti) della disponibilità proteica ed energetica per due gruppi familiari di tre e dieci individui che svolgano un’attività moderata o intensa.

146 -

b i b L i o g R A F i A

BALZARINI, A. 2001-2002 - Indicatori morfometrici di impatto ambientale: il caso delle paleopopolazioni di Cera-stoderma glaucum (Poiret, 1789) del sito di monte d’Accoddi (Porto Torres - Sassari). Tesi di Laurea discussa all’Università degli Studi di Milano. Facoltà di Scienze Matematiche, Fisiche e Naturali, Corso di Laurea in Scienze Biologiche. Anno Accademico 2001-2002.

CHENORKIAN, R. 1989 - Mollusques testacés et diètes préhistoriques. Travaux du lAPmo, 1989: 29-58. Aix-en-Provence.

CHENORKIAN, R. 1992 - Saisonnalisation et malacofaune. Méthodes et possibilités. Préhistoire et anthropologie méditerranéennes, 1: 143-168.

CHENORKIAN, R. 1996 - Pratique archéologique statistique et graphique. Errance, Paris: �3-83.

CLAASSEN, C. 1998 - Shells. Cambridge Manuals in Archaeology. Cambridge University Press, Cambridge.

GEIGy TAVOLE SCIENTIFICHE, 1983 - Volume 1. Unità di misura, liquidi corporei, composizione del corpo, nutrizione. Ciba-Geigy Edizioni, Basilea.

GIROD, A. e BALZARINI, A. 2004-2005 - Indicatori morfometrici di impatto ambientale: le paleopopolazioni di Cerastoderma glaucum (Poiret, 1�89) di Monte d’Accoddi (Porto Torres). Atti della Società per la Prei-storia e Protostoria della Regione Friuli-Venezia giulia, XV: 135-153.

GIROD, A. e NOVATI, M. 2004 - Contributo alla conoscenza della malacofauna del sito di Monte d’Accoddi, Porto Torres (SS), Sardegna. Atti della Società per la Preistoria e Protostoria della Regione Friuli-Venezia giulia, XIV: 15�-1�9.

HOLLAND, B., BROWN, J. e BUSS, D.H. 1993 - Fish and fish products. The supplement to: MCCANCE, R. A. e WIDDOW-SON, E. M. (eds.) The composition of foods. Royal Society of Chemistry, Cambridge: �8-85 (5a edizione).

MORETEAU, J.C. 1995 - Méthodes d’étude de la croissance chez les mollusques. In VICENTE, N. (ed.) Vingt ans de malacologie. Mémoires de l’Institut océanographique Paul Ricard, 199: 65-�6.

NOVATI, M. 199� - les malacofaunes archéologiques de la Sardaigne et le problème de monte d’Accoddi. Diplôme d’Etudes Approfondies, 96. Univ. D’Aix-Marseille I.

SECCHI, G. 1990 - I nostri alimenti. Hoepli, Milano.

SIEGEL, S. e CASTELLAN, N.J. 1992 - Statistica non parametrica. MacGraw-Hill Libri Italia srl, Milano (2 a edizione).

SOUCI, S.W., FACHMANN, W. e KRAUT, H. 2000 - Food Composition and Nutritional Tables die zusammensetzung der lebensmittel Nährwert-Tabellen la composition des aliments Tableaux des valeurs nutritives. Medpharm Scientific Publishers, Stuttgart (6a edizione).

TINÉ, S. e TRAVERSO, A. 1992 - monte d’Accoddi. dieci anni di nuovi scavi. Istituto Italiano Archeologia Sperimen-tale, Genova.

TRAVERSO, A. 1996-199� - Il santuario di monte d’Accoddi (Sassari) materiali e tecniche costruttive. Tesi di Laurea discussa all’Università degli Studi di Genova. Facoltà di Lettere e Filosofia. Anno Accademico 1996-199�.

Indirizzo degli Autori:ALBERTO GIROD, Laboratorio di Malacologia Applicata, Via Savona 94/A – I - 20144 MILANO e-mail: fraberto.girod@fastwebnet.it

GIULIA RICCARDINO, Via Ambrosini 3 – I - 20052 MONZA (MI)e-mail: july.r@libero.it