Eurasian Prehistory, I(/): 83- 106.

TRAPEZOIDS AND DOUBLE TRUNCATIONS IN THE EPIGRAVETTIAN ASSEMBLAGES OF NORTHEASTERN ITALY

Silvia Ferrari and Marco Peresani

University ofFerrara, Dipartimento delle Risorse Nalura/i e Culturali, Corso Ercole I d'Este 32, 1-44100 Ferrara. Italy; [email protected] (Marco Peresani)

Abstract Trapezoids represent a significant category of tools among the innovative geometric implements in the lithic assem­ blages of European late glacial complexes, particularly those late Epigravettian industries from the Mediterranean region to the southern Ukrainian plain. The recent recovery of several artifacts from two excavlllions in the Venetian Pre-Alps (northern Italy) prompted a large-scale examination of the chronological, and geographical distribution of trapezoids as well as the first study of their techno-morphological features. Results show an evident 'ariability in those morphological and dimensional parameters investigated and seem to suggest that the prepar-.ttion of different shapes of blanks to obtain microliths might have occurred in the ambit of economical behaviors. Specifically, these economic choices involve an in­ vestment in the manufacture ofmicroliths and this is evident in the intensi"e retouch of the tools.

position of the so-called hi-truncated bladelets. INTRODUCTION These were included in a group of geometric mi­ Trapezoids are well known from the Meso­ croliths known as Divers (Daniel and Yignard, lithic and Early Neolithic technocomplexcs. 1953), and distinct from the trapezes in their ra­ These geometric implements, produced by double tios of length to breadth (lamelles a deux lronca­ transverse truncation from laminar blanks, were tures, G.E.E.M .. 1969; jleche tranchante, Bar­ identified at the end of nineteenth century, when riere, 1956; Escalon de Fonton, 1953; Trapeze G. Chierici ( 1875) and A. de Mortillet ( 1896), de­ etire, Daniel and Yignard, 1953), or by the or­ scribed the Trapezes and Rhomboi'des. In the thogonal disposition of both the truncations in re­ twentieth century, many scholars addressed the spect to the blank axis (Rectangles, lleinzelin de typological definition and classification criteria of Braucourt, 1962; Daniel and Yignard, 1953) and these artifacts (Heinzelin de Braucourt, 1962; were therefore considered as belonging to the Daniel and Vignard, 1953; Barriere, 1956; Boh­ truncated-pieces category (Tixier, 1963; Laplace, mers and Wouters, 1956; Tixier, 1963; Laplace, 1964a). 1964a; Rozoy, 1967; G.E.E.M., 1969; Hinout, At the end of the nineteenth century the mi­ 1973; Broglie and Kozlowski, 1983), giving rise croburin method had been identified among the to a more and more detailed spectra of types and lithic complexes (Siret, 1893; Breuil, 1921) and varieties that, in some cases, were obviously was considered implicit in typological classifica­ linked to the regional context While recent stud­ tions in the following years (Heinzelin de Brau­ ies seem to support the morphological distinction court, 1962). Truncation, by contrast, does not re­ of principal types from the in classifying lithic ar­ quire systematic blade fragmentation through tifacts, some scientists are at variance about the such a method, since truncation can be produced 84 S. Ferrari & M. Pcrcsani

by transverse fractures obtained by means of flex­ ger Dryas period, excavations of two mountain ion or percussion. sites conducted by the Trento Natural Science Initially classified as points, and successively Museum and the Ferrara University both devoted defined as armatures microlithiques, arrowheads, to investigating human occupation at the preal­ or projectile points, trapezoids were the subject of pine fringe provided insights. Since the lithic as­ hypotheses concerning their functional signifi­ semblages recovered from the extensive excava­ cance (de Mortillct, 1883; Dechclette. 1924; Frii­ tion at the Bus de Ia Lum site on the Cansiglio Johanscn, 1919). In the last few decades interest Plateau (Peresani et a!., 1999-2000) and from a in improving traccological analyses and experi ­ test trench at La Cogola Rockshelter on the Car­ mentation peaked, as did debates on whether bonare Pass (Dalmcri eta/., 1995) yielded tens of these were multifunctional tools, or transverse, trapezoidal pieces, we considered these assem­ oblique, or piercing arrowheads, with or without blages viable for assessing the general spread of various barbs and edges hafted on projectile this particular type of a1tifact. weapons (Barton, 1992; ance, 1971; Rozoy, As a basis for further research devoted to the 1978 and, recently, uzhnyj, 1989; 1993; 2000). functional significance of trapezoids among the Regard ing these two latter possibilities, the finds late glacial technocomplexes and for the special recovered in European Paleolithic and Mesolithic purposes of thi article, we conducted a large­ sites like Stellmoor, Holmegaard IV, Loshult, scale re-examination of the evidence reported in Tvaerrnose, Vinkel Mose, and Shikaevka II in the literature about the occurrence of similar western Siberia, provided an essential contribu­ pieces. Cultural and chronological extra-regional tion to the investigation of hafting methods. Fi­ affinities, techno-typological features, chronol­ nally, traccological evidence tends to indicate use ogy, and the environmental contexts of northern in hunting tasks for these pieces (Tringham eta/., Italian implements are presented below and con­ 1974; Odell , 1978; Lcmorini, 1997; Winiarska­ sidered in relation with the site settlement system Kabacinska, 1992) but revealed improper use too ofthe eastern Italian Alps. (Odell, 1978). Widespread in Europe from the second half TRAPEZOIDS IN THE LATE of ninth millennia cal. B.C. during Late Mesoli­ GLACIAL EUROPEAN CULTURES thic, the presence of these geometries among the A broad, preliminary overview of the litera­ late glacial cultures has always been neglected ture concerning Europe and its neighboring re­ due to their very low frequency and geographical gions revealed a wide distribution of trapezoids discontinuity. Both the sn1dies of lithic assem­ from the end of Last Glacial Maximum up to the blages and the detailed fieldwork repo1ts count, beginning of Holocene. By focusing on the sites describe, and classify trapezoids as rare or occa­ that have been 14C dated or the li thic industries sional pieces but have elicited minimal interest for with undoubted cultural attribution. it is possible a large-scale examination in defining Late Upper to show how the utilization of this implement con­ Paleolithic cultural phases. cerned both a large temporal interval and geogra­ In northeastern Italy, few trapezoids have phic area, from the Near East to the Great Euro­ been recovered during research in the last forty pean Plain. We arbitrarily subdivided the overall years from late glacial Epigravettian sites exca­ evidence into two main chronological steps, the vated at the foot of the Pre-Alps, or in the moun­ older one coinciding with the Oldest Dryas inclu­ tain region (Ferrari, 2000-200 I). Such a scarce ding the end of the Last Glacial Maximum, and number of implements is usually associated with the later one including the late glacial interstadials very rare geometric pieces, like triangles and lu­ up to the Pleistocene/ Holocene boundary, in order nates, and with the imperceptible appearance of to assess the chronological occurrence of these morphologically defined microburins. This me­ geometric mjcroliths. thod was systematically adopted at the end of the Epigravettian, when a dramatic increase in the End of the Last Glacial Maximum a nd the frequency ofmicroliths occurred. Concerning the Oldest Dr yas Recent Epigravettian and in particular the Youn- Trapezoids were documented during this pe- Trapezoids and double truncations in the Epigravettian 85 riod in some parts of southern Ew-ope, like the blages (i.e. at La Gare de Couze, layers from G to Balkans, the Italian peninsula, and Crimea, and B, layer C dated to 11.730±3 I 0 B.P.; Bordes, De outside of Europe in the Near East (Syria. Jordan). Sonneville-Bordes, 1979), while in Aquitania, ln the Near East, the Kebaran culture shows a sig­ some bitruncated pieces were found within the nificant increase in trapezes a dos (Cauvin and Laborien complex (between I 0,350 B.P. and Coqueugniot, 1988; Kaufman, 1988), as well as 9870 B.P. Lc Tensorer, 1981 ). the presence of trapezoids at Umm el Tiel (layer An important area for their occurrence is 6: 16,320:1:200 B.P.; layer 5: 14,650±70 B.P.), El Middle Europe, where Ahresburgian sites pro­ Kown (Syria), and eve-David (13,400± 150; vided evidence of use for this microlithic imple­ 12,610±130 B.P.). Northward, in Transcaucasia, ment later on at ca. I 0,320±250, 981 0± I 00 B. P. such implements arc part of the Shan-Koba Cul­ (Bohmers, I 960; i.e. Stell moor). Complexes from ture that occupies the ecological zone of the Belgium to western Poland show a step of devel­ northern slopes of Caucasia and Crimea (Ko­ opment with microliths (Dewez, 1979; Kobusic­ zlowski, I 977). The data is scarce for the Balkan wicz and Kabacinski, I 992; Ottc, 1983). Peninsula, and only the site of Kadar I in Bosnia, which dates back to the end of the Last Glacial Italy Maximum (TL dates: layer IE/2b, 17,200± I ,400; ln Italy, trapezoids were noted at several Late layer IE/2a, 16,200± I ,050), yielded some pieces. Upper Paleolithic sites in Sicily and along the During this first chronological step, the fre­ Adriatic and Tyrrenian coastal belts (Fig. l). The quency of trapezoids in assemblages appears oldest evidence was recorded in southern Italy rather uniform, numbering two or three pieces per (Table 1): Sicily, and the Apulian and Latium Re­ site, except for the Near East assemblages, where gions. Some sites in western Sicily, Nebrodi 1 such occurrences tend to be more important • Mounts, and Egadi Islands (Acqua Fitusa Cave, San Teodoro Cave, Levanzo Cave) show a cul­ Late glacial interstadials a nd the beginning tural succession that includes a bipartite Late Epi­ Holocene gravettian, from the Oldest Dryas to the B0lling During the late glacial interstadials and espe­ chronozones. Drawing on one of the main aspects cially after the fourteenth and fifteenth millennia in the oldest phase, the geometric implements cal. B.P., there is an increasing number of assem­ seem to disappear at one of the northernmost blages containing trapezoids, as well as a spread sites. Rare isocele, scalene, and rectangular trape­ of this implement in many regions of Europe. The zoids have been recovered at the Acqua Fitusa presence of this artifact is characterized by high Cave, levels 3 to 5 (Bianchini and Gambassini, numerical and morphological variability. 1973), Giovanna Cave in the Siracusa Province In the Near East, sites like Kharaneh IV rec­ (Cardini, I 971 ), and San Teodoro Cave, levels D I ord geometries at 10,620±125 B.P. (layer D) with and C (Lo Vetro and Martini, I 999-2000). the development of the Kcbaran Culture and an Concerning the Apulian Region and the increasing microlith production that was a prelude lower-middle Adriatic coastal belt, Paglicci Cave to the appearance of the Natufian. By comparison, in the Gargano Region (Mczzena and Palma di the evidence for Crimea (Buran-Kaya Ill, layers Cesnola, 1967: Palma di Cesnola et a/.. 1983) 6.8 and 6.9, dated to I 1,900± 150 B.P., I 1,950± shows a detailed record. The thick and culturally 130 B.P.), and the central and southern Ukraine, rich stratigraphic succession explored at this site show a different world represented by Postgravet­ yielded the oldest trapezoid from level 8, dated to tian cultures that, for Osokorivka and Shan-Koba, the end of the Last Glacial Maximum. Specifi­ includes trapezes and some large triangles in their cally in the Apulian region, the late glacial evi­ lithic tool sets (Zaliznyak, 1999). dence is represented by the very rare implements In southern Europe, we see the production of found at Santa Croce Cave (Dryas lll Allemd) and these geometric implements during the final Epi­ at Zinzulusa Cave, whil e in the Romanellian gravettian, both in the Balkan and Italian peninsu­ phase at the Youger Dryas-Preboreal transition, a lae. Sicily included. Westward in Dordogne, tra­ few pieces were found at Romanelli (Laplace, pezoids are repo11ed in Late Magdalenian assem- 1964b) and Cavallo (Laplace, l964b) caves. 86 S. Ferrari & M. Pcrcsani

taincd trapezoids in their lithic assemblages. At Polesini Cave at Ponte Lucano near Tivoli (La­ tium; dated to the end of Younger Dryas; Laplace, 1964b; Bietti et a/., 1985) one isoscelc trapeze and one rhomboidal piece were recovered from layer 7. Attributed to the same final cultural phase, the assemblages recovered in layers l l and 12 at the Salvini Rockshelter in Latium, included few double-truncated bladelets (Bietri, 1984). Similar artifacts have been documented at Pali­ doro Cave in sections I and 8 within layer B (dated to the Oldest Dryas; Bietti, 1976 77). Fur­ ther north in inland Tuscany, sites like Fredian Rockshelter (Boschian et a/., 1995), Pianali di Montalbano (Martini, 1991 ), and Castagnina (Ga­ liberti, 1997) contained evidence dated to the Younger Dryas (Fredian), or has been culturally attributed to the end ofthe Evolved Epigravettian. Finally, among the numerous late glacial sites of the upper Tyn·enian coastal belt, only one rectan­ gle trapeze was found at the Arene Candide Cave Fig. I. Sketch map of the Italian Epigravcttian sites in layers 1-2 (B ietti, 1994; Laplace, 1964b). that include trapezoids in lithic assemblages: I - The spread of these implements is clearly Giovanna Cave; 2 - Acqua Fitusa Cave: 3 - San Teo­ characterized by very low frequencies in the ear­ doro Cave; 4 U1uzzo Cave; S- Romanelli and Zinzu­ lusa caves; 6 - Santa Croce Cave; 7 - Paglicci Cave; 8 - lier sets and, in some cases, by their difference in Prete and Ferrovia caves; 9 - Salvini Shelter; I0 - size relative to those recorded for the younger Po1esini Cave; II - Palidoro Cave; 12-Castagnina; 13 contexts. For this reason, we point out that the two - Pianali di Monte Albano; 14 - Fredian Shelter; IS ­ trapezoids found in the Oldest Dryas/B0lling A rene Candide Cave; 16 - Soman Shelter; 17-Tag! i­ level 10 at Tagliente Rockshelter (Fig. 4: 37, 38) ente Shelter; 18 - La Cogola Shelter; 19 - Val Lastari; because of their notable size, have been excluded 20 - Villabruna A Shelter; 21 - Bus de La Lum from the analysis of the Recent Epigravcttian set.

NORTHEASTERN ITALY AND THE RECENT EPIGRA VETTIAN Moving northward, the numerous Bertonian in­ dustries in the Abruzzi Region (Bisi eta/., 1983) The context and its modifications: chronology, lack double-truncated pieces that, on the contrary, environment, site distribution, main cultural were recovered in the Marchc Region (Broglio outline and Lollini, 1981 ). The Grotta del Prete record The late glacial period in northeastern Italy marks the last step of the cultural evolution that corresponds approximately to the Recent Epi­ occurred during the Aller0d- Preboreal interval gravettian as A. Broglio ( I 980), and Bisi et a/. and that also seems to have afTected the techno­ ( 1983) stated when they presented the chronologi­ logical background. While the oldest phase is cal and cultural layout of the Tagliente Rockshel­ characterized by the blade and microblade pro­ ter succession and those of other sites previously duction at Ferrovia Cave with geometric pieces, discovered in this region. After more than 20 during the latest one it is possible to observe a de­ years of research in the Eastern Italian Alps, this crease in the microgravette content and the ap­ definition is sti ll largely acceptable, even if some pearance of trapezoids. aspects regarding the internal subdivision and the Moving northward along the Tyrrenian position of the upper limit are under discussion coastal belt and inland, few late glacial sites con- and remain to be clarified (Bartolomei et a!., Trapezoids and double truncations in the Epigravettian 87

Table 1 Occurrence of trapezoidal implements and their frequency in the totality ofmicro li ths recovered at the Recent Epigravettian sites of northeastern Italy r--s· -- I Altitude r I -:--1=-Tra2ezoids M;orohtl>• % _ -----=- ue -4 1 1 ~'"";'U.T.B. 195 2.6 Som'" :0 Sholto ~ L.T.B. 473 --· _Q2__ Tagliente Shelter 250 529 0.3 1--!0c ---- 2 Vi labruna Shelter 500 ---5-4 75 4 ---5.3 Bus de Ia Lum 950 253 26 10.3 3 Va l Lastari 1060 ____§!_9 - - 3 0.4 t= ,:. 73 3 4. 1 ILa Cogola Shelter 1070 191 21 11.0 9 313 13 4.2 -- - - Dalmeri Shelter• 1240 h±I /'26 2566 4 - 0.2------Notes about the provenance of the pieces: I - lower thermoelastic breccia, unit 160; upper thermoelastic breccia. units I IOA , 4/J 24, 133, 5/21 , 158; 2-pieces equall y distributed between layer 5 and 4; 3 - two pieces from unit 3, one from its arbitrary cut 3E: 4 - pieces recovered from units 14a, 14b, 22, 26c (data on La Cogola and Dalmeri shelters ha ve been provided by A.Cusinato and G.Dalmeri)

1977; Broglio, 1992; 1997; Guerreschi, 1983; late glacial beginning is marked by the occurrence Palma di Ccsnola, 1993; Cusinato eta/., in press). of open larch stands (Casadoro eta/., 1976), from Currently, a technological study of some lithic as­ which the conifer forests expanded during the semblages recovered at both valley bottom and Belling and the tree line reached about 1,500 m mountain sites (Montoya et a/., in press) is pro­ altitude or even higher in the inner Italian Alps viding new perspectives for investigating Epi­ (Schneider, 1985; Kofler, 1994). Consistently, a gravettian behavior and culture, above all whether recently discovered late glacial-Earl y Holocene these are correlated with the geographical and paleobotanical and archaeological record, the biological modifications that occurred in the al­ Palughetto basin on the Cansiglio Plateau (Avigli­ pine southern margin during the late glacial. ln ano et a/., 2000), shows that tree line was well this region, the history of human colonization is over 1,050 mat the end of the B0lling. Many pol­ supported by numerous sites, some of which have len records at the southern alpine margin suggest been systematicall y investigated, and by various a mass tree expansion at 12,300/ 12,200 B.P. and interdiscipli nary data that ill ustrate the ecological provide evidence that the start of peat accumula­ contexts in which the Epigravettians li ved from tion at low and middle altitudes is close to the age Last Glacial Maximum to the beginning Holo­ of forest expansion during a short time interval cene. during the second part of B01ling (Caval! in et a/., The studies conducted in the Belluno Area 1997; Gehrig, 1997; Wick, 1996). demonstrate that after the Last Glacial Maximum Forest expansion and the stabilization of (ca. 17,000 B.P. in the northern Treviso area; mountain slopes during the 80lling- Ailemd per­ Bondesan, 1999) the glacier withdrawal from the mitted the peopling of the eastern Pre-Alps and prealpine fringe was accompanied by intermo­ the Dolomites. This process is evident in the rainic lake formation and influenced indirectly the number of camps (more than 40) lying on the occurrence of several landslide events responsible karst plateaux (Lessini, Tonezza, Asiago, Cansi­ for the fonnation of other lacustrine basins (Pelle­ glio, Piancavallo) at altitudes ranging from I ,000 grini and Surian 1994, 1996). In some sites, the to l ,500 m, and in the main valleys crossing the 88 S. Ferrari & M. Pcresani

co tr? ti mti ,.._

10

11 I

0.: !Xi >. 12 ~ ~ 13

14

15 +--

16 ~

Fig. 2. Conventional radiocarbon dates for the Italian Epigravcnian assemblages with trapezoids, except for the northeastern region. Sites are arranged from left to right according to the geographic critcra used in the text: Sicilian sites, AFC-Acqua Fitusa Cave (layers 3-5, Azzi, 1973) and GC Giovanna Cave (Alessio eta/., 1978): the Apulian region and the lower-middle Adriatic coastal belt and inland, PC Paglicci Cave (layer 8, Azzi et at., 1974). RC ­ Romanelli Cave (layer C, Vogel, Waterbolk, 1963), PrC- Prete Cave (layer 6, Alessio et at., 1978); the tyrrenian coastal belt, PaC - Palidoro Cave (layer B, arbitrary cuts l and 8, Alessio ef at., 1976-- 1977), PoC - Polcsin i Cave (layer 7, Belluomini. 1980), FS - Fred ian Shelter (layer 5. Bosch ian et at.. 1995), AC - A rene Candide Cave (layers I, 2, Bieni, 1994) prealpine fringe (Broglio, 1992; Broglio and close to the Trento Basin (Viotte del Bondone, Lanzinger, 1996) and peaking at I ,870 m in the Andalo, Terlago; Bagolini and Guerreschi, 1978; Dolomitic Region (Avanzini et al., 1998), near Bagolini and Dalmeri, 1983; Guerreschi, 1984) rock cliffs, peat-bogs, sheets of water, or under must receive a more reliable chronological attri­ roekshelters sometimes very close to flint sources. bution going beyond that suggested by their Sites at valley bottoms usually lie near lakes or pede-stratigraphic layouts (Cremaschi and Lao­ under rockshelters close to a stream. zinger, 1983; 1987), the dates provided at Bus de Regarding the Younger Dryas, the impact of La Lum (Peresani eta/., 1999- 2000) and La Co­ this climatic event on the prealpine landscape is gola shelter (Dalmeri et a/., 1995) indicate that known through palynological diagrams (Schnei­ mountain human occupation certainly occurred der and Tobolski, 1983; Wick, 1996) that show a during the Younger Dryas, and that it fits the evi­ moderate retreat of forest vegetation at the preal­ dence from Soman shelter (Battaglia eta/., 1992), pine border (300-500 m). While at higher altitude although its position is in a valley bottom at the ( 1,000 m) this main environmental change has not prealpine fringe (Fig. 2). been macroscopically observed, in the way that The main cultural outline of the Recent Epi­ the Palughetto conifer record documents (A vigli­ gravettian has been illustrated traditionally on the ano et a/., 2000). basis of typological features of the lithic assem­ From this perspective, while open ai r sites blages recovered in open air and sheltered sites. Trapezoids and double truncations in the Epigravcttian 89

Other aspects regarding subsistence, site organi­ trated flints, charcoal, and other remains both ver­ zation, production, utilization of non-flint tools, tically and horizontally, and weathered artifact and art, are less well known due to the strong dis­ surfaces (Angelucci and Peresani, in press; Ange­ homogeneity and general dearth of archaeological lucci eta/., 1995). evertheless, important mate­ records. According to Broglio and Lanzinger rial was obtained from a few recently discovered ( 1990), in the settlement system different site sites: La Cogola shelter at the Astico Va lley head types can be distinguished. (Dalmeri eta/., 1995), Dalmeri shelter on the Asi­ ago Plateau (Bassetti et a/., 1998; Fiore et a/., Residential sites in valley bottoms at the 1998), and Yillabruna shelters (A, B, C) in the prealpine fringe Cismon Valley (Aimar eta/., 1992). All of them, These are sheltered sites at the base of cliffs, and in particular Dalmeri, show well-preserved in cave openings, or on steep slopes; a few have occupation layers and subsistence economies dur­ been found in the open too. Currently, the main ing the Allemd- Younger Dryas interval. Evi­ bulk of information comes from the excavations dence of seasonal ibex hunting was uncovered at conducted at Soman shelter in the Adigc Valley Dalmeri, where bone remains were discarded and (Battaglia et a/. , 1992; Tagliacozzo and Cassoli, accumulated on the ground both inside and out­ 1992); Taglientc shelter at the foot of the Lessini side a dwelling structure show the time of slaugh­ Mountains (Bartolomei et a/., 1982; 1992; Cilli ter between summer and autumn (Curci and Ta­ and Guerreschi, 2000; Cilli eta!., in press; Liagre, gliacozzo, 2000; Fiore et a/., l998). The occa­ 2000- 200 I); Biarzo shelter at the base of an allu­ siona l presence of other ungulates as well as vari­ vial terrace in the Natisone Valley, eastern Friuli ous carnivores has also been reported. Fishing is (Guerreschi, 1996). All of these sites show evi­ documented too. At a lower altitude, the Villa­ dence of repeated human occupation, paleo living bruna A shelter revealed vari ed game that might floor organization, and subsistence economy and reflect the swTounding environment and its evolu­ show that a variety of activities took place in the tion: in lower levels ( 17-1 0) ibex comprises more inhabited area including: flint tool manufacture, than 50% of total ungulates (red deer and cham­ bone tool manufacture, butchering, hearth con­ ois), while in upper levels (9-4) the dominance struction and maintenance, modification of living turns in favor of red deer and roe-buck (Aimar et areas, cleaning tasks in the sheltered area, and ref­ a/., 1992). use evacuation. Fishing was a significant task too: The functional significance of these open air at Biarzo it occurred all year long (Cassoli and sites has yet to be analyzed for use-wear and lithic Tagliacozzo, in Guerreschi, 1996). The spectra of technology. The lithics are characterized by an hunted mammals reflects the paleoecological evo­ equilibrated rate between retouched tools and lution at site surroundings: at Tagliente, ibex and various implements used to make arrowheads or bovids appear to be dominant during the Oldest barbs. Excavations perfonned at the Val Lastari Dryas (Bartolomei et a/., 1982), while during the site (Broglio et a/., 1992) exposed flint work­ Bolling red deer, wild boar, roe-buck, and cham­ shops, a flint tool cache, and thousands of flaking ois predominate. Similar prey was hunted at So­ by-products that were the result of various reduc­ man and Biarzo too, during the Allerod, but dur­ tion sequences aimed at obtaining blades, blade­ ing the Younger Dryas at Soman there is a de­ lets, and laminar flakes to be used, shaped or ex­ creasing frequency of wild boar, red deer, and ported outside the investigated area (Peresani, roe-buck. 1992; Montoya and Peresani, in press).

Middle-low mountain seasonal camps. mostly Seasonal hunting camps at middle-high altitude selll ed in the open Lithic assemblages recovered at these sites Except for Dahneri and La Cogola shelters, are both scarce and dominated by types intended the archaeological records of these sites were af­ for hafting onto weapons: backed points, backed fected by several postdepositional agents that dis­ bladelets, truncated backed bladelets, and some mantled the living floors, dissolved bones and types of geometric microliths. At present, only other perishable remains, dispersed or concen- two sites show this type of evidence, Pian dei 90 S. Ferrari & M. Peresani

Laghetti (an open air site at I ,488 m; Bagolini et ment of the Early Epigravettian, because this a/., 1984), and Tschonstoan Rockshelter (at 1.870 technological tradition seems to be discemable in m in the Siusi Alps; Avanzini eta/., 1998). Due to the flaking procedure, typology, and size of the a dearth of archaeological remains in both sites lithic implements. On the other hand, the two fol­ they remain radiometrically undated and their lowing phases mark the appearance of technologi­ chrono-cultural status has been assigned based on cal and typological innovations giving a signature the typological features of the lithic artifacts. to the lithic assemblages that, in the author's opin­ For retouched implements, the Tagliente suc­ ion, unexclusively come from the Epigravettian cession is uniquely useful for typologically defin­ entity. In fact, the spread of particular types of ing the first of three phases in which the Recent short end-scrapers, backed knives, and geometric Epigravettian has been divided (Bisi el a/., 1983) implements obtained through the microburin because this succession covers a chronological method as well as trapezoids, is a complex phe­ span from the Older Dryas to the beginning Al­ nomenon that goes beyond the Epigravettian area lered2. In fact, the tool sets recovered from levels and points to the establishment of connections 16-1 I are dominant in elongated frontal endscrap­ among different cultural entities. It seems likely ers on short types that, inversely, predominate in that this situation was favored by the ameliorated the overlying levels {l 0-4) during the second environmental conditions. phase. The second phase is well represented in many other sites in addition to Tagliente levels Sites with trapezoids in a multi-differentiated I 0-4 (Broglie, 1992): endscrapers prevail as very landscape: stratigraphic layout and basic short, thumbnail, sub-circular, semicircular, and information fan shapes; truncations show variable frequency, The Recent Epigravettian industries that con­ sometimes higher at mountain sites than at low al­ tain trapezoids are from some prealpine sites that titude sites; backed knives are present only in the vary in location, environment, and site preserva­ Allemd; like the microgravettes, the backed tion. In accordance with their geography and alti­ points form a large typological variety including tude - the mountain foot to the middle altitude implements with truncated bases and proximal plateaux - information is presented about the stra­ points with natural base; rare lunates, triangles, tigraphy, environmental data, archaeological lay­ and trapezoids occur. Except for trapezoids which out, and archaeozoological data (when available), disappear, these microlithic geometric imple­ and techno-typological data extracted from the ments abruptly increase during the third, final Tagliente, Soman, Villabruna, La Cogola and Epigravettian phase, concurrently with the mas­ Dalmcri shelters, and the open air sites of Bus de sive adoption of the microburin method and the La Lum and Val Lastari. All of these sites have appearance of truncation-points and double been radiocarbon dated (Fig. 3) and, the shelters backed points. in particular, have preserved stratigraphic succes­ Chronologically, these phases approximately sions that provide feasibility for analyzing their fall into the late glacial chronozones in accor­ lithic assemblages. The open air Epigravcttian site dance with the 14C dates obtained from many dif­ ofTerlago, settled at 414 min altitude (Bagolini ferent sites (Broglio and lmprota, 1994- 1995): and Dalmeri, 1983), was disregarded for the pur­ the first phase being included in the Older Dryas, poses this study, due to the suspicion that the se­ the second one covering the late glacial interstadi­ quences have been polluted with culturally als and the main part of the Younger Dryas, while younger geometric implements. the third one is of uncertain position, and might be placed around the Younger Dryas- Preboreal bou­ Tagliente Rockshe/ter ndary. The attribution of sites to one or other of Tagliente Rockshclter lies on the right bank these phases, nevertheless differs among scholars of the Valpantena river, at the foot of a small (Broglio, 1992; Guerreschi, 1996). rocky wall at 250 m in the Lessini Mountains. According to A. Broglio (Broglio and Im­ This shelter was initially investigated from 1962 prota, 1994-1995; Broglio, 1997), the lithic in­ to 1964 by the Verona Natural History Museum, dustry of the first phase derives from a develop- and thereafter the Epigravettian deposits were Trapezoids and double truncations in the Epigravettian 91

Q) Q) ~ N en en 0 Cll en ~ J:l J:l (j) Cll Cll Cll (j) Cll 1- 1- 1- 1- en > > 0 0 0 > 0 ....J en en (/) Cll ....J en ....J > 9 +-

10 ~------

11

12 1------~·~~--~------~ r I

13 ~--4,------~l

14

Fig. 3. Conventional radiocarbon dates for the Epigravettian assemblages with trapezoids in northeastern Jtaly, arranged in chronological order: VL - Val Lastari (unit 3, arbitrary cuts 3D and 38; 30 including structures Sill, SlY, SV; Broglio et al. , 1992); TS - Tagliente shelter (layer 10 and 8-10; Alessio et al., 1970; Hedges eta/., 1994); SS Soman shelter (lower and upper thermoelastic breccia; Battaglia eta/., 1992); DS - Dalmeri shelter (units 14/26e, 26b; Bassetti eta/., 1995); LCS - La Cogola shelter (units 19 and 18, Dalmeri; pers. comm.) largely excavated by the University of Ferrara (P. artifacts, bones, and other remains. Moving out­ Leonardi) from 1967 to the present (A.Broglio, side, the parallel bedding shifts from horizontal to A.Guerreschi, C. Peretto). Partially disturbed in inclined. the medieval period, its thick stratigraphic succes­ This episode of Epigravettian human occupa­ sion dates back to the Upper Pleistocene and is tion lasted about 3,000 years from the Oldest formed by two sequences. The lowermost se­ Dryas to the Allemd (Capuzzi and Sala, 1980) quence, which has been attributed to the Middle during a considerable environmental change to­ and Early Wi.irm, groups several lithological units ward a more temperate and wooded landscape in of cryogenic clast-supported breccia with blocks level 12 (Cremaschi, 1990; ). Presently, research of rock-fall; from the base, a partially rubified is focused on the archaeozoology (Cilli eta/., in clayey matrix is progressively substituted by aeo­ press), lithic workshops (Guerreschi and Squin­ lian loams toward the top (Cremaschi, 1990). This tani, 1995), blade and bladelet manufacture (Lia­ sequence includes Mousterian assemblages (lev­ gre, 2001 - 2002; Montoya et a/. , in press), bone els 52-31) and a few reworked Aurignacian re­ implements, and ornamental objects (Fiocchi, mains (level 25). Above an erosional surface, the 1998). Regarding the lithic technology, Montoya uppermost sequence dates back to the late glacial. eta/. (in press) point out the relationship between It thickens considerably from inside to outside the the laminar end-products and the use of soft ham­ shelter. From bottom to top, it includes conglom­ mer throughout the entire sequence. Typological erates, gelifluction and solifluction deposits, and information was provided by A. Gucrreschi (Bar­ alternating planar living fl oors (levels 18-4) con­ tolomei eta/., 1982) who studied levels 16 to 4. taining structures, flint workshops, a burial, lithic Among the tools, the longer endscrapers prevail 92 S. Ferrari & M. Pcrcsani in levels 16 to 11 with a majority being the shorter the lower units, the two phases share the same ty­ types such as thumbnail, semicircular, circular, pological features in their lithic assemblages. sub-circular, and fan sha- pes. Burins frequently Tools account for ca. 30% of the whole set: short appear on truncated blades and bladelets, borers, thumbnail, semicircular, circular, sub-circular, retouched blades, and scrapers. Backed points are and fan-shaped endscrapers are more frequent generally small and straight, while various types than truncations and burins. Among the backed of truncated backed bladelets prevail in levels I 0 items, various types of truncated backed bladelets to 4. Rare triangles, lunates, and two trapezoids largely prevail on points, on a very few backed were recovered from level I 0 to 4. bladelets, and rare triangles, lu nates, and trape­ zoids. Some microburins occur too. Truncated Soman Rockshelter backed bladelets show a large typological variety Soman Rockshelter lies at I 00 m in the Adige mostly produced by single or double truncation: Valley, three kilometers north of the Ceraino normally oriented or symmetrically or asymmetri­ Gorge, and was investigated by A. Broglie and M. cally disposed, with oblique or obtuse angle. Lanzinger from 1984 to 1988 (Battaglia et a/., Points, appear mostly as microgravettes and small 1992). Laid upon a gravelly bar and overbank de­ types with the base tapered by accessory bipolar posits of the Adige River, the stratigraphic se­ retouch. Trapezoids are six: one from the lower quence consists from bottom to top of thermoelas­ complex, five from the upper. tic breccia including Late Upper Paleolithic lay­ ers, and of mainly colluvial and anthropogenic Villabruna Rockshelters fine deposits with Mesolithic and Neolithic re­ Filled by a glacier ablation-tongue during the mains. Excavations took place in two sectors, the Last Glacial Maximum, the Cismon Valley in the smaller excavation investigated the deposit par­ Venetian Dolomites was inhabited by Epigravet­ tially disturbed by recent quarry activity, and the tians at the end of the B0lling interstadial. This larger excavation revealed an Epigravettian living evidence is provided by the Yillabruna rockshel­ floor with bone accumulation. Stratigraphy, fau­ ters, positioned at ca. 500 m at the foot of a lime­ nal remains, and 14C dates reveal the existence of stone cliff and excavated in 1988- 1989 by A. two main Epigravettian occupation phases corre­ Broglie (Aimar eta/., 1992). Explored at the top lated with some environmental modifications: the of a debris cone, the stratigraphic succession of lower one dates back to the Aller0d, the upper one Villabruna A consists of clast-supported cryo­ to the Younger Dryas (Battaglia eta!., 1992). genic breccia with organic matrix accumulated According to Tagliacozzo and Cassoli during the human occupations. Units from 17 up ( 1992), Epigravettians hunted adult and young­ to I OA are radiometrically dated back to the Epi­ adult ungulates in the ecologicall y varied site sur­ gravettian; unit 3 is of Mesolithic age. In unit I 7, a roundings, and sporadically hunted elk and au­ burial with painted stones was found. The other rochs. Some carnivores of different species are shelters, B and C, yielded Epigravettian, Late also represented, as well as birds and fish. Sum­ Mesolithic, and Neoli thic finds. mer or early autumn seasonal slaughters on the Partially destroyed by recent quarry activi­ mountain overhanging the shelter are suggested ties, the anthropogenic deposits provided no de­ by the very young age as was estimated for certain tailed data on the settlement organization. Tapho­ ungulates. nomic and archaeozoological analyses proved Except for some detailed analyses on the that the bone accumulation was connected with laminar technological procedures (Montoya, pers. the hunting of ungulates: ibex is the most repre­ comm.) and truncations (Peresani eta!., in press), sented species in the lower part (units 17-10) at present only the lithic assemblage has received while, inversely, red-deer increases in the upper preliminarily analysis (Battaglia et a/., 1992). part (units 9-4). Few remains of wi ld boar, roe­ Cores and flaking products derive from methods buck, and small carnivores were found (Aimar et aimed at obtaining blades and bladelets. With re­ a/., 1992; Aimar and Giacobini, 1995). gard to the retouched implements, apart from the Recently, the lithic industry has been the ob­ backed knives that have been recovered only in ject of a preliminary investigation, both from the Trapezoids and double truncations in the Epigravettian 93 technological (Montoya et a/., in press) and the in the context of mountain environments, and a typological standpoints (Aimar et a!., 1992). Re­ large area has been under investigation since 199 1 garding flint supply, petroarchaeological studies by the Trento Natural Science Museum (Bassetti carried out on geological and archaeological sam­ eta/., 1998). ples revealed some affinities with the gray flint The units forming the sedimentary sequence sources in the Biancone limestones 4 km from the are normally made of clast-supported breccia with site (Benedetti eta/., 1992). Flint was exploited in a silty loam, quartzitic-micaceous matrix, and order to produce blades and bladelets. Among the show evidence of different climatic phases, i.e. retouched tools, frontal endscrapers are the most from relatively cold and humid, to a temperate pe­ frequent with very short, thumbnail and semicir­ riod during the human occupation which dates cular outlines; other forms occasionally occur. On back to the Allemd chronozone and, then to an­ the other hand, some blades were shaped by other cold context. A geoarchaeological study abrupt bifacial retouch in backed knives and other (Angelucci and Peresani, 1998) conducted by retouched implements. Microliths are represented means of routine analyses and micromorphologi­ by microgravettes, commonly, backed truncated cal investigations provided data for reconstructing bladelets and very few geometric pieces (backed the natural and cultural formation processes of the lunates and triangles). The unique four trapezoids lower Epigravettian un it (unit 26). The results in­ were found in layer 5. dicate that humans occupied a soil surface that was already stable and affected by soil formation. La Cogo!a Rockshe/ter The human induced microfabrics derive from Recently discovered at the head of the Astico various activities: the preparation of structured Valley by the Trento Natural Science Museum occupation surfaces through the deposition of (G. Dalmeri), La Cogola Rockshelter was investi ­ aeolian silt sediment that was collected outside gated in 1998, 1999 and 2002. Settled at I ,070 m the shelter; the disposal of organic refuse more altitude at the foot of a modest limestone cliff fac­ than its accumulation; trampling, and manufactur­ ing south, the site shows a well-preserved thin ing. Depositional and postdepositional modifica­ stratigraphic sequence formed by a clast-suppor­ tion processes have also been observed (organic ted cryogenic deposit, ranging from the Final Up­ and clay alluviation, organic soil bioturbation, per Paleolithic (units 19 and 18) to the Early degradation by cryoturbation of the structured oc­ Mesolithic (unit 16; Dalmeri et aL., 1995) in cupation surface). which abundant faunal remains, bone and 'flint ar­ Faunal remains (Baitolomei, 1998; Fiore et tifacts have been recovered. The present-day re­ a/., 1998; Girod, 1998), flint and bone artifacts search involves studies on the natural and cultural (Bassetti and Dalmeri, 1993; Bassetti et a/., 1995 formation processes, wood charcoals, lithic as­ 1998; Cusinato, 1999), ornamental objects (Dal­ semblages, micro and macro-mammal remains, meri and Fiocchi, 1998), and painted stones (Dal­ and birds (Dalmeri et a/., in press). The Epigra­ meri, 1998; Dalmeri eta/., in press) were found. vettian assemblage includes bladelet cores, trun­ As regards the above mentioned macromammal cation and dihedral burins, freq uent short and studies (Curci and Tag! iacozzo, 2000; Fiore eta/. , very short endscrapers, truncations, and several 1998), they suggest that the shelter was inhabited backed bladelets, microgravettes, and 37 trape­ during the summer or early autumn and that it was zoids of which 13 were recovered in unit 19, 21 a specialized ibex hunting site. were recovered in unit 18, and 3 pieces were re­ A sample of the lithic assemblage was ana­ covered in unit 16. lyzed typologically and a first assessment has been proposed about the organization of lithic Dalmeri Rockshelter production (Cusinato, 1999; Bassetti eta/., 1995; Dalmeri Rockshelter is situated at I ,240 m on Peresani et a/., in press). Detailed lithic techno­ the northern slope of the Asiago Plateau at the logical, typological and functional analyses are in head of a stream valley tributary of the Brenta progress (Dalmeri, Cusinato, Montoya, pers. river. This site is notable both for its geographic comm.). The local gray flint provisioned from the setting and its well preserved dwelling structures Biancone marly-limestones, and various exoge- 94 S. Ferrari & M. Pcrcsani

nous types have been exploited in order to obtain quence: below units I and 2 which contain re­ blades, bladelets, and some flakes (Cusinato, worked Upper Paleolithic and Middle Paleolithic 1999). flint artifacts, there is a silty loam soil (unit 3, with Regarding the retouched tool set, a certain va­ arbitrary levels from 3A to 3F) with Epigravettian riety is represented by the most frequent imple­ remains, mostly found on a living floor (levels ment endscrapers and by burins, retouched blades 3D- 3F); unit 4 is a coarser steri le deposit. Field and truncations. Backed knives, points, and borers data and micromorphological analyses performed appear in low frequencies. Pieces esqui/lees, den­ in a geoarchaeological study (Angelucci and ticulates, scrapers, and retouched flakes are rare. Peresani 1995) indicate that the soil profile devel­ Endscraper typology is dominated by the short oped from a loess cover resting on slope deposits, frontal, while the remaining types are sub­ and that two late glacial pedogenctic phases, one circular, oval, thick or long frontal. Truncation of which related to the 3D-3F levels, occurred. burins in majority, and dihedral, multiple or on Furthermore, the authors confim1 (Angelucci and fracture. Truncations occur more on blades than Peresani, in press) that various postdepositional on bladelets, and arc equall y divided into oblique processes acted in reducing the bulk of archaeo­ and nonnally oriented types; most are stra ight and logical remains (dissolution of bones and other their retouch seems to show a proportional in va­ non-flint material) and distributing both vertica lly sivity with reference to the linear extension (Pere­ and horizontally the li thic artifacts. sani eta/., in press). Marginal or invasive retouch The archaeological remains were found scat­ occurs on blades. tered inside the soil profile except on the lower Truncated backed bladelets and backed floor, where they were grouped near the wall points largely prevail on backed bladelets and rare amid some boulders, suggesting the existence of triangles, lunates, and trapezoids. Some microbu­ flint workshops; moving beyond these, flints de­ rins occur too. Truncated backed bladelets form a crease abruptly. Unworked or tested blocks, vari­ large typological variety depending on the trunca­ ous by-products, prepared cobbles, cores and tion occurrence and arrangement: single or double large cortical flakes, flakes, blades, bladelets and truncation, normal or inclined, with an oblique or a few stone hammers were recovered usually in obtuse angle, symmetrically or asymmetrically heaps, the contemporaneity of which was proven arranged. The backed points are mainly micro­ by several refittings. At the base of the wall, some gravettes and other different types. There are four waste pits with an abundance of flaking by­ trapezes total. products, unexploited cobbles, cores and pre­ cores were connected with the workshop. In the Val Lastari second sector a few meters away from the foot­ Val Lastari is an open air site that lies on the slope, a cache was found that contained unworked Asiago Plateau, in the Lastari Valley at the foot of or tested flint cobbles that had been selected from the southem slope, ncar a shallow limestone wall the clay-with-flint deposits along the valley­ at I ,060 m. Excavations managed from 1990 up to bottom (Peresani, 1992; Peresani, in press). 1996 by A. Broglio and M. Peresani (Broglio et The aims and modalities of toolmaking were a/., 1992) surveyed a 70 m2 area, including two identified by technological analyses (Montoya main sectors. Currently, research is focused on the and Peresani, in press) that revealed the occur­ organization of lithic production, by means of rence of three independent sequences largely per­ technological (Montoya and Peresani, in press) formed on local flint: production of straight and functional analyses. Detailed studies on the blades to be shaped into backed knives or to be retouched tools and microliths as well as spatial used as unretouched pieces; production of blade­ analyses will be the next step. Disregarding the lets for obtaining two morphometric categories large chronological interval, the set of radiocar­ useful in shaping backed pieces; lastly, a few bon dates indicate that site occupation mainly oc­ barely retouched blanks were obtained from a curred during Allcn;,d interstadial (more details subordinate production of laminar flakes. are given in Broglio e1 a/., 1992). Regarding retouched tools, this assemblage Four main units form the stratigraphic se- was not polluted by Mesolithic and Neolithic Trapezoids and double truncations in the Ep igravcttian 95 tools - both technologically and typologically di­ Mesolithic or more recent occupations was de­ agnostic elements. Tools (mainly endscrapers, bu­ tected. Blades and bladelets were chipped from rins, truncations, backed kn ives, and other re­ blocks, pebbles and thick flake-cores: from the touched blades) and microliths (backed points, core technological features, it is not surprising bladelets, tnmcated bladclcts) are recorded with that reduction involved more than one single se­ almost the same frequency (Broglio eta/., 1992). quence and that it was influenced by the proper­ Usually obtained from fl akes and various by­ ties and shapes of exotic and local raw material. products, the endscrapers are frontal, very short, Backed knives are absent. The other tools depict a with thumbnail and semicircular outlines; other large typological variety dominated by trunca­ types occur with very low frequencies. Dihedral, tions and retouched blades, with a lesser presence truncation, on fracture btu·ins were obtained from of retouched flakes, short frontal endscrapers blades and thick Oakes. On the other hand, regular (with sub-parallel sides, thumbnail, or fan blades were selected to be shaped into numerous shapes), burins and scrapers. These were shaped truncations (Pcresani et a/., in press), backed on various flaking end- and by-products like corti­ knives and retouched implements. Among the mi­ cal blades and flakes, trimming-core blades, strik­ eroliths, many different types of backed poin ts ing platform rejuvenation flakes, and broken were recognized: microgravettes with direct or hi­ pieces. Even if highl y fragmentary, the microliths facial back and accessory retouches, types with include truncation-points, backed points, backed proximal point and natural base, short and large or and truncated backed bladelets, retouched blade­ truncated pieces. Backed truncated bladclets dis­ lets, rare scalene triangles, several (26) trapezoids play a large typological vari ety derived from the and multiple truncations. Backed points make up truncation number, arrangement and orientation. the single largest component with different forms: We point out the occasional evidence of the mi­ Microgravenes, very finely retouched, with a thin croburin method, as well as the presence of geo­ or abrupt thick base, with additional truncation. metri c pieces (triangles and three trapezoids). with double back.

Bus de La Lum THE ANALYSIS OF EPIGRAVETTIAN Excavated fro m 1997 to 2000 on a 80 m2 TRAPEZOIDS AND COMPARISONS large area, Bus de La Lum lies at 995 m on the WITH LATE MESOLITHIC Cansiglio karst plateau in an extremely uneven IMPLEMENTS zone with sinkholes, ponor and limestone pave­ ments, where the preservation of extensive level Materials and methods ground, like that which hosted the camp, is excep­ Overall, 82 implements have been recovered tional (Peresani et a/., I 999- 2000). The archaeo­ from the selected sites, (Table I, Fig. 4); 27 of logical record was strongly reduced by karstic and which are partially fragmentary or incomplete. pcdogenetic processes, both of which were re­ Since the lithic sets from Bus de La Lum and La sponsible for the dissolution of bones and non­ Cogola shelters make up the most consistent bulk flint artifacts and of a characteristic dispersion of of the collection (76.8%), they are likely to influ­ lithics in the underlying Upper Pleistocene loess ence the main features of the overall techno­ deposit (Angelucci and Pcresani. in press). Pres­ typological aspect of the sample. Since the field­ ently, research is in progress and concerns study work at La Cogola is sti ll in progress, trapezoids of natural depositional and postdepositional pro­ from the Younger Dryas sites will be the subject cesses as well as spatial, technological, and func­ of a specific investigation in the future. tional analyses designed to reconstruct the mo­ For the purposes of the present study, and tak­ dalities of human occupation(s) dated back to the ing into account the particular variability ex­ Younger Dryas. pressed by the Epigravettian specimens, we usc Apart from a few sandstone pebbles used as the term trapezoid in a broad sense, that is to in­ hammers or retouchers and two refitted hematite clude all blanks that bear double truncation or. at scales, more than 6,000 flint tools constitute the least, one truncation opposite to one unretouched Epigravettian lithic assemblage. No evidence of fracture which represen ts the characteristic geo- 96 S. Ferrari & M. Peresani

metric shape to the artifact. Concepts and criteria Table 2 of the analytical procedure are inferred by Boh­ Main morphometric features (in mm) of the mers and Wouters ( 1956) and, in keeping with the trapezoids (total=82) Mesolithic contexts of northeastern Italy, by Bro­ .--- glie and Kozlowski ( 1983). This latter analytical max 111111 111 ode average procedure was usefu l in making comparisons be­ len >th main base 35 8 15 15.7 tween the Epigravettian and Castelnovian imple­ ~ng th shoner base 3 1 2 7 8.7 ments in order to point out their main technologi­ cal and typological differences. ~adth 20 6 10 11.6 Each structural element of the geometric mi­ css 5 2 t 2 2.6 crolith was recorded and analyzed: truncations (retouched or unretouched fractures), blank fea­ of the parameters that has been retained for distin­ tures (sides, sagittal profile, transverse section), guishing different types of blanks is breadth, even retouched edges (truncation and accessory re­ if width is affected by accessory rctouch(es) that touches), and size (length, breadth, thickness). left the original size unmodified. By synthesizing Each specimen was oriented by placing its base the body of data that we accumulated from studies on the right. In order to recognize the original devoted to recognizing methods and variability in blank manufactured for toolmaking, and to distin­ the Epigravettian laminar production of northeast­ guish between flakes, blades, and bladelets, the ern Italy (Montoya et at. , in press; Montoya and general outline and the arrangement of scars on Peresani, in press), we fixed tbe boundary be­ the dorsal face were documented. Other features tween blades and bladelets around a breadth of 12 like the sagittal profile, transverse section, thick­ mm. Trapezoids have breadth values ranging ness and breadth were retained as useful for deter­ from 6 to 20 mm and the majority of cases (52) mining which cri teria blanks were selected. Trun­ are in the 6 to 12 mm interval (Table 2). Thick­ cations were evaluated on their degree of incline ness varies from 2 up to 5 mm, the highest fre­ (in degrees, clockwise) and their outline. Finally, quency being recorded in the lowest class. As data about orientation, inclination and retouch in­ concerns the original blank outl ine, pieces with vasivity, both on truncations and the sides, were straight sides are the most frequent (3 1% ), but for documented. The lengthening index (l.i.) is calcu­ the remaining 69%, we note a high variability due lated as the rate between the longest side (or the to almost all the combinations possible between main base of the trapezoid) and its breadth. In ad­ straight, convex, concave, sinuous, being the dition to the truncation pattern, the lengthening straight/convex, straight/sinuous and straight/ index contributes to the identification and the concave the most freq uent. The sagittal outline is definition ofthe specimen, in accordance with the straight in the majority of cases too (80%), but typological protocol. different combinations (concave, straight, con­ vex, sinuous) equally occur. Transverse section is The technological and typological features of vari able: trapezoidal and triangular, the both sym­ the Epigravettian geometries metri c and asymmetric, with regular or irregularly All of the specimens exhibit similar condi­ crossed dorsal ridges. In some cases no arris is tions of preservation: patina is very light, edges present on the dorsal face. are unabraded; only a few of them are affected by The above data put into evidence several fea­ breakage and chromatic modification due to ther­ tures: in addition to the blades (29.2%) and blade­ moclastism. In perfect accordance with the litho­ lets (58.8%) showing regular outlines, a few logical composition of the corresponding indus­ flakes with mainly convex sides and convex or tries, the trapezoids were manufactured on the sinuous sagittal outlines were used (6%). Finally, same flint types. 2.2% of the sample still bears cortical portions.

Size and blank outline Truncations Various types of blanks were shaped: blades, Both the upper and lower truncations define bladelets, and flakes. The geometric layout is one two value ranges: 10°-70° (14.8%) and 310°- Trapezoids and double truncations in the Epigravettian 97

0 ~ ~ G(} m-o2 3 4 5 6

8 9 10 11 12 13

0 0 [] t1] (f] 14 15 16 17 18 19 20 21

0

I ~ ... ) . ,_ 22 23 24 26 28 29

0

·~ 30 31 32 33 34 35 36

0

0 3 em -.~==:....

37 38 39 40

Fig. 4. Trapezoids recovered at the Recent Epigravcttian sites of northeastern Italy: La Cogola shelter (n. 1~, 6, 8, 12, 15, 24, 29), Soman shelter (n. 5, 13), Bus de La Lum (n. 7, 9- 11, 16, 19- 23, 25- 28, 30-34, 36), Dalmcri shel­ ter (n. 14, 35), Yillabruna A shelter (n. 17), Val Lastari (n. 18, 39, 40), Tagliente shelter (n. 37, 38) (drawings by G. Almerigogna, $. Ferrari, A. Paoli llo) 98 S. Ferrari & M. Pcresani

Table 3 Frequency distribution of the inclination values (in degrees, clockwise) for proximal (left) and distal (above) truncations

I r-- 0 Incl. ( ) Inclination distal truncation 1--- e> -.- -- ~ro'rune. 10 20 30 50 70 310 320 330 340 350 360 total 10 I 3 4 8 _10 l I 2 5 2 l - 12 30 I 3 I 4 3 7 18 ___iO I 2 3 2 3 I 12 I - 1- _ 50 I I 3 I 6 60 I I I _1.!0 I I - 320 I I _140 I l l 3 I - 350 2 3 ___1§0 I - - I I 6 r------1-- total 3 3 2 2 I 7 10 15 15 7 74 '-- -'----- 9 I

360° (85.2%) for the upper; I 0°-60° (77%) and Sides 310°-360° (23%) the lower (Table 3). Thus, it fol­ The breadth and length scatter (Fig. 5) puts lows that the upper truncations are usually turned into evidence the high dispersion of values and to the left and that the lower ones usually to the shows a contraction in the length range at 12 mm, right. Additionally, data shows the occurrence of that is not related to a particular type of object or trapezoids with normal truncations to be 8.2%. As with a clear concentration of pieces from one site. concerns the zenital outline, the frequency of Finds from new excavations probably will clarify pieces with both straight truncations is significant the reasons for this. The length of the sides may (33.3%) with respect to those with concave, con­ be an useful parameter for the typological distinc­ vex or irregular edges. A few pieces have both tion of the trapezoids. Concerning the shorter concave truncations. side, the higher frequency falls in the 5- 8 mm in­ terval; the longer in the 11 - 17 mm interval (Table Retouch 2). An equal length was measured on 6 (7.2%) im­ Most of the trapezoids were manufactured by plements showing parallel, oblique or nonnal means of retouched truncations (80%), while in truncations. On 6 pieces, the length of one side the remaining pieces, a simple transverse fracture was double that of the other side. is visible. Five pieces from La Cogola show one Trapezoids from La Cogola and Bus de La unretouched fracture. Any fracture produced Lum show interesting accessory retouches: 15 through the microburin method has been ob­ finds bear retouch only on one of the two sides (9 served. The majority (65.0%) of the cases bear to­ on the shorter side); this retouch does not seem to tally retouched truncations, wh il e partial retou­ be homogeneous in distribution, incl ination or in­ ches usually affect more than half of the length. vasivity. Distribution may be total (7 pieces) or Retouching was direct (88.8%), alternate (9.7%) more limited in extension (the last fourth of the and inverse (1.5%). lnvasivity has been observed edge length). on upper (87.9%) and lower (84.3%) truncations. Trapezoids and double truncations in the Epigravettian 99

40 Lengthening index and morphological differentiation The lengthening index defines a continuous 35 range of values, from 0.5 up to 3.5; the highest frequencies have been recorded in 0.5-1.0 and 1.5- 2.0 intervals (76%), while 36.1 % falls in the 30 1.0- 1.5 class. This distribution gives ri se to the following typological differentiation: 54.2% arc very short trapezoids, 2 I. 7 %arc short, and 24. 1% 25 arc long (Fig. 6). Among the types, we record a .s::. predominance of symmetric forms (68.0%), g. 20 Q) 45.1% of which have straight both truncations), _.J by comparison to the asymmetric (32.0%), the I 0.0% of which are very short with decalee base. 15 Among the long forms, asymmetric prevail (60.0%) on remaining symmetric, while on the contrary, among the short forms asymmetric are 10 lightly predominant (52.9%) on symmetric. Very short pieces have asymmetric outlines. 5

Most differences between the Epigravettian and Castelnovian trapezoids by a sample 0 comparison 0 5 10 15 20 25 Breadth From layers AB I and AB2 of the Mesolithic series of Romagnano III Rockshelter in the Adige Fig. S. Distribution of length and breadth values (in Valley (Alessio et a/., 1978; Broglio and Ko­ mm) for 79 trapezoids zlowski, 1983), we examined 94 artifacts with the 40 intent to cover the widest typological variety. The 35 sample illustrates the almost exclusive utilizati on 30 of bladelets, and a very hi gh degree of standardi­ 25 zation (straight sagittal outline, parallel sides, and

including Italy, is likely to be interpreted as there­ devoted to maintaining the same caliber and mor­ sult of large-scale cultural renewals, as a conse­ phology of bladelets during the flaking proce­ quence of the climatic and environmental amelio­ dures, to the disavantage of retouching and shap­ ration that followed the Last Glacial Maximum. ing both the truncations. Tt is questionable whe­ The widening of colonizable territories, favored ther the low frequency of standardized features by the ample reforestation of the eastern Italian observed in the Epigravettian bladelets represent Alps, was accompanied by the opening of the the style of flaking or are the result of the use of mountain passes making the crossing of this re­ flaking end- and by-products. For the Epigravet­ gion easier. In accordance with the Broglio's tian bladelets, new information is available from statement ( 1997), the occurrence of trapezoids fits the technologial studies of the Epigravettian in­ in with the appearance of different retouched tools dustries (Montoya et a/., in press) which share and various hunting implements as well as of with other late glacial cultural systems (Azilian, some technological procedures used in toolmak­ Federrnesser) in northern and southern France ing in relation to the settlement dynamics ob­ (Bracco et a/., 1997) the exclusive adoption of served in Friuli, Trento, and Veneto. soft hammer stones in order to obtain bladelets. In assessing the significance of such a mas­ The consequence of this dramatic technological sive renewal both in toolmaking and in lithic as­ shift - the abandonment of the organic hammer semblage compositions during the late glacial in­ and use of soft hammer stones - was a complex terstadials, improvements arc required from the innovation in the design of hunting implements techno-functional studies that, at present do not and, obviously, in the system of hunting modali­ support tracing the main activities connected with ties (Pelegrin, 2000), that might have concerned the manufacture and use of the trapezoids. Paral­ also the trapezoids. lels with similar Late Mesolithic implements seem to be inadequate due to the cultural, tempo­ Acknowledgments ral and environmental distance between the two We are grateful for the research funds provided by periods. As explained above, Castelnovian tra­ M.U.R.S.T. (40%) for our research project entitled, pezes in northern Italy reveal a high degree of "Origini ed evoluzione del Popolamcnto umano in lta­ standardization both in blank procedures and tool lia: paleobiologia, comportamcnto e strategic di sussis­ tcnza". The authors are very grateful to Mr. Giampaolo manufacture, mak ing the difference with the Epi­ Dalmeri (Trcnto Natural Science Museum), and to gravettian items evident. On the other hand, these Prof. Alberto Broglio and Prof. Antonio Guerreschi latter implements show a large variability in all (University of Ferrara) for pennission to study the im­ the morphological and dimensional features con­ plements from La Cogola, Dalmeri, Soman and Tagli­ sidered. The blanks shaped are usually bladelets, ente shelters. but we cannot neglect the use of blades, flakes and cortical blanks, whose sides are not parallel and Notes straight in their sagittal outline. Truncations are I. The typological reference adopted for the ana­ also variable in their inclination and plane outline, lysis of Kebaran industries with geometries, defines which are mostly straight and convergent. Con­ trapezoid implements as hi-truncated pieces or trapezes cerning their manufacture, it fits well what is without a back, or as a trapezoidal outline lacking a known about the shrewd methods used for the for­ back. The frequency of this latter type is part of the mal tools of some Epigravettian sites (Peresani et criteria for distinguishing the development phases of a/., in press): no application of microburin this culture (Bar-Yosef, 1970). 2. The chronology of the Taglientc succession, as method; abrupt retouch, mostly invasive and com­ well as the criteria adopted for the recognition of the plete, usually direct, and sometimes inverse. above mentioned phases at this site, have been debated The use of various types of blanks to manu­ recently by Montoya et al. (in press) who points out the facture microliths might have occurred in the am­ organi zation of lithic technology. bit of economical behavior that involved a higher investment in manufacture by retouch. Especially REFERENCES after examining the Castelnovian specimens, it is AI MAR A., GIACOBlNI G. 1995. Analisi dei resti fau­ evident that most techno-economical efforts were nistici del deposito epigravettiano dci Ripari Vi l- Trapezoids and double truncations in the Epigraveltian 101

labruna (Val Rosna, Belluno). In: Alii del I Con­ AVIGLIANO R. , Dl ANASTASIO G., PERESANI vegno Nazionale di Archeozoologia. Padusa, Quad­ M., RA VAZZI C. 2000. A New Late glacial-Early erni I, 125- 134. Holocene Palaeobotanical and Archaeological Re­ AIMAR A., ALCIATI G .. BROGLIO A., CASTEL­ cord in the Eastern Pre-Alps: the Palughctto basin LETT! L., CATTANI L., D'AMICO C., Gl­ (Consiglio Plateau, Italy). Journal of Quatem01y ACOBINI G., MASPERO A., PERESANI M. 1992. Science, 15/8, 789-803. Les Abris Vi llabruna dans Ia Vall ee du Cismon. AZZI C. M., BIGLIOCCA L., PIOVAN E. 1973. Flor­ Preistoria Alpina, 2811, 227- 254. ence Radiocarbon Dates II. Radiocarbon, 15/3, ALESSIO M., BELLA F., CALDERON! G., 479-487. CORTESI C., IMPROTA S. 197fr77. Carbon-14 AZZI C. M., BIGLIOCCA L., PIOVAN E. 1974. Flor­ Dating of bone co llagen from Upper Palaeolithic ence Radiocarbon Dates II. Radiocarbon, 161 l, I 0- Palidoro deposits. Quaternaria, 19, 181 186. 14. ALESSIO M., BELLA F., IMPROTA S., BEL­ BAGOLINI A., DALMERl G. 1983. Site paleolithique LUOMINI G., CORTESI C., TURI B. 1970. Un i­ ta rdi f- mesolithique du Lac du Terlago (Trento). versity of Rome Carbon-14 Dates VIII. Radiocar­ Preistoria A lpina, 19, 189- 196. bon, 12/2, 591-616. BAGOLINI B., GUERRESCHI A. 1978. Notizie pre­ ALESSIO M., BELLA F., IMPROTA S., BEL­ liminari sulle ricerche 1977 1978 nell' insediamento LUOM INI G., CALDERON! G., CORTESI C., paleolitico delle Viotte di Bondone (Trcnto). Preis­ TURI B. 1976-77. University of Rome Carbon-14 loria Alpina, 14, 7-31. Dates XIV. Radiocarbon, 18/3,321-349. BAGOLINI B., FERRARl A., LANZINGER M., ALESSIO M., ALLEGRI L., BELLA F., IMPROTA PASQUALl T. 1984. Pian dci Laghctti-S. Martino S., BELLUOM INI G., CALDERON! G., CORTESI di Castrozza. Preistoria Alpina, 20, 39 52. C., MANFRA L., TURl V. 1978. University of BARRIERE C. 1956. Les civilisations Tardenoisiennes Rome Carbon-14 Dates XVI. Radiocarbon, 20. en Europe Occidemale. Ed. Biere, Bordeaux. 79- 104. BARTOLOME! G. 1998. Riparo Dalmeri: a high­ ANGELUCCI D. E., PERESANI M. 1995. Geoarchae­ altitude late glacial site in the venetian pre-alps. Pa­ ology of the epigravcttian site of Val Lastari (Ve­ laeoecological and palaeoclimatological data from neto Prealps). Preistoria Alpina, 31, 13- 21. the micromammal remains. In: Atti XXXIII Riunione ANGELUCCI D. E., PERESANI M. 1998. The sedi­ Scient!fica lstituto Italiano Preistoria e Protostoria. mentary sequence of Riparo Dalmeri (Grigno Val­ Preistoria Alpina, 34, 167- 172. sugana, TN): pedostratigrafic and micromorpho­ BARTOLOMEI G., BROGLIO A., PALMA Dl CES­ logical features. In: Atti XXXIII Riunione Scientiflca NOLA A. 1977. Chronostratigraphic et ecologic de lstituto Italiano Preistoria e Protostoria. Preistoria I'Epigravettien en Italic. In: Sonneville-Bordes D., Alpina, 34, 155- 162. de (ed.) Lajin des temps glaciaires en Europe. Actes ANGELUCCI D. E., PERESANI M. in press. The open du Colloque International, CNRS, 297- 324. sites of Val Lastari and Bus de Ia Lum (Venetian BARTOLOMEI G., BROGLIO A., CATTANI L., Pre-Alps, Northern Italy): preliminary results on CREMASCHI M., GUERRESCHl A., MAN­ spatial analysis and the study of postdepositional TOY ANI E., PERETTO C., SALA B. 1982. f depo­ processes. In: Bie M.,de, Caspar, J-P. (eds.), Drifted siti wtlrmiani del Riparo Tagliente. Annali Univer­ Apart. Stone age sites in bioturbated sediments and sita Ferrara. sez. XV, Ill, 4, 61-105. their value for intra-site spatial analysis. Proceed- BARTOLOMEI G., BROGLIO A., CASTELLETTI ! ings XIV UISPP Congress. B.A.R. L. , CATTANI L., CREMASCHI M., GUERRE­ 1 ANGELUCCI D. E., BROGLIO A., PERESANT M. SCHI A., MANTOVANI E., MASPERO A., PER­ 1995. A Geoarchacological Approach to Living ETTO C., SALA B. 1992. Abri Taglicnte. Preisto­ Floors: Some Remarks on Palaeo-Mesolithic Sites ria Alpina, 28/2, 299-3 13. of the South-Eastern Alps. In: Proocedings of the l BARTON R.N.E. 1992. Hengistbwy head dorset. The International Congress on: "Science and Technol­ late Upper Palaeolithic and early Mesolithic sites. ogy jor the Safeguard of Culwral heritage in the Oxford University Committee for archaeology, 34, Mediterranean Basin", CNR, Catania, 1447- 1457 299 pp. AVANZINI M., BROGLJO A., DE STEFANI M., BAR-YOSEF 0. 1970. The Epipalaeolithic cultures of LANZINGER M., LEMORINI C., ROSSETTI P. Palestine. Ph. D. Thesys, llebrew University, Jeru­ 1998. The Tschonstoan Rock shelter at Alpe di Siusi salem. - Seiser Aim. In: Alii XXXJ/1 Riunione Scientiflca BASSETTI M., CUSINATO A., DALMERI G., KOM­ lstituro Italiano Preistoria e Protostoria. Preistoria PATSCHER K. , KOMPATSCHER M. 1995. Ri­ Alpina, 34, 81-98. paro Dalmeri (Trento). L'industria litica negli spazi 102 S. Ferrari & M. Percsani

d'abitato epigravettiani. Preistoria Alpina, 31. 23- BOHMERS A. 1960. Statistiques et graphiques dans 36. !'etude des industries lithiques prehistoriques. Con­ BASSETTI M., DALMERI G., KOMPATSCHER K., sideration general au sujet du Hambourgien, du KO MPATSCHER M., LANZINGER M. 1998. Re­ Tjongericn, du Magdalenicn et de I' Azi li en. Palaeo­ search on the Epigravettian site of Riparo Dalmeri historia, 8, I 37. on the Sette Comuni Plateau (Trento): some prelimi­ BOHMERS A., WOUTERS A. 1956. Statistic and gra­ nary comments. In: Alii XXXI!! Riunione Scientifica phy in the study of flint assemblages. PalaeohistO­ lstituto Italiano Preistoria e Protostoria. Preistoria ria, 5, 1- 38. 14 Alpina, 34, 139- 154. BON DESAN A. 1999. Una nuova data C nell 'anfitca­ BATTAGLIA L., BROGLIO A., CASTELLETTI L., tro morenico di Vittorio Veneto e Ia costruzione del LANZINGER M., MASPERO A. 1992. Abri So­ suo sandur (Pianura del F. Meschio). 8th Italian man. Preisroria Alpina, 28/2, 291-298. Glaciological Meering 'Alpine glaciers and climate BELLUOMINI G. 1980. Datazioni direttc di ossa fos­ change', Abs tract volume, 87- 89. sili umane provcnienti da siti archeologici dcll' ltalia BORDES F., SONNEVILLE-BORDES D. 1979. centro-meridionale effettuatc con il mctodo della L'azilianisation dans Ia vallee de Ia Dordogne: lcs racemizzazione dell'acido aspanico. Geograjia donnees de Ia Gare de Cou.te (Dordogne) et de l'abri Fisica e Dincunica Quatemaria, 3, 25- 34. Morin (Gironde). ln: Colloque 1ntcmational CNRS, BENEDETTI R., CAM PANA R., D'AMICO C., 27 I, 449-459. NANNEITI M. C. 1992. Petroarchaeomctry of epi­ BOSCHIAN G., MALLEGNI F., TOZZI C. 1995. The gravcttian and mcsolithic flints in the Val Cismon­ cpigravettian and mesolithic site of Frcdian shelter Lagorai area (NE Italy). The flint supply question, (Toscany). Quaternaria NOI'CI , 5, 45- 80. Preistoria Alpina, 28/ 1 33-49. BRACCO J. P., GAGNEPA IN J., STOUVENOT C., BIANCHINI G., GAMBASSINI P. 1973. La Grotta BIDART P., VlGIER S., 1997. L'industrie lithique dcii'Acqua Fitusa (Agrigento). I. Gli scavi e Ia li­ cpigravetticnnc de Saint-Antoine - locus 2 (Vi­ tica. Rivisra di Scienze Preistoriche, 28, 3- 55. trolles, Hautes-Aipes): premiere analyse. Paleo. 9. BIETTI A. 1976-77. Analysis and illustration of the 221-243. Epigravettian industry collected during the 1955 ex­ 13REU1L H. 1921. Note sur Ia communication de cavations at Palidoro (Rome, Italy). Qualernaria, E.Cartailhac: Observations sur !'Hiatus ct le Neo­ XIX, 197-387. litbiquc. L 'Anthropologie, XXXI, 349- 354. BIETTI A. I984. Primi risultati sui giacimento BROGLIO A. 1980. Culn•rc e ambienti della fme del dell'cpigravettiano finale di Riparo salvini (Ter­ Paleolitico e del Mesolitico nell'ltalia nord-orien­ racina, Latina). In: Alfi XXIV Riunione Scientifica tale. Preistoria Alpina, 16, 7-29. lstillllo Italiano Preistoria Pro/os/oria, 195- 205. BROGLIO A. 1992. Mountain sites in the context of BIETTl A. (cd.) 1994. The Upper Pleistocene deposits the North-Eat italian upper Palaeolithic and Meso­ of the Arene Candide cave (savona. Italy): new stud­ lithic. Preistoria Alpina, 28/ 1. 293-310. ies on the 1940-42 excavations. Quatemaria Nova, BROG LI O A. 1997. Considerations sur I' Epigravetticn IV, 370 pp. italique. In J.M. Fullola, N. Soler (cds.), £1 mon BIETTI A., MARTINI F., TOZZI C. 1985. L'Epi­ mediterranei despres del Plenig/acial ( 18.000- gravetticn evoluc ct final de Ia zone moye1me et 12.000 BP), Serie Monografica, 17, Museu d'Ar­ bassc tyrrehnicnne. In: La posirion ICL'Asia go (Prealpes de Ia Venetie). Preistoria paro Dalmeri (TN). In: Alii del Jl Convegno Nazion­ Alpina, 28/1, 207- 225. ale di Archeozoologia, Abaco Edizioni, Forli, 23- CAPUZZI P., SALA B., 1980. II Riparo Tagliente. 30. Analisi delle faune, biostratigrafia e cronologia dei CUSINATO A. 1999. L'industria litica epigravettiana livelli tardiglaciali. In Aspes A. {ed.) II Territorio di Riparo Dalmeri. Rivista di Archeologia, XXIH, Veronese dalle origini al/'eta Romano. Fiorini, Ve­ 5- 26. rona, 130-136. CUSINATO A., DALM ERI G., FONTANA F., CARDIN! L. 1971. Rinvenimento paleolitico nella GUERRESCH1 A. in press. The Epigravettian of Grotta Giovanna (Siracusa). Alii Xlll Riunione sci­ North-Eastern Italy. In: Bracco J.P., Montoya C. entijica l.I.P.P., 28- 34. (eds.) Les systhnes techniques lirhiques pendant le CASADORO G., CASTIGLIONI G. B., CORONA E., Tardiglaciaire autour de Ia Medirerranee nord­ MASSARI F., MORETTO M.G., PAGANELLI A., occidentale. Actes de Ia Table Ronde. TERENZIANI F., TONlELLO V. 1976. Un de­ DALMERI G. 1998. The geometrical figures of Riparo posito tardowurmiano con tronchi subfossili alle Dalmeri (Eastern Trentino) of the Recent Epigravet­ Fornaci di Revine (Treviso). Bollettino del Comitato tian period. In: Atti XXXIII Riunionc Scientifica ls­ Glaciologico Italiano, 24, 22- 63. tituto Italiano Preistoria e Protostoria. Preistoria CA VALLTN A., BARONI C., BINI A., CARTON A., Alpina, 34, 193- 196. MARCHETTI M., OROMBELLI G., PELFINI M., DALMERI G., FIOCCHI C. 1998. The mal acological ZANCH! A. 1997. Geomorphology of the Central collection from Riparo Dalmeri. In: Ani XXXIII Ri­ and Southern Alps. Guide for the excursion of the unione Scientifica lstituto Italiano Prcistoria e Pro­ Fourth International Conference of Geomorphol­ tostoria. Preistoria Alpina, 34, 185-192. ogy. Geogra.fia Fisico e Dincunica Quaternaria, 3 DALMERJ G., BASSETTI M., CUS INATO A., DE­ (supplement 2), I 3-47. GASPER! N., KOMPATSCHER K., KOMPAT­ CAUVIN M.C., COQUEUGNIOT E. 1988. L'Oasis SCHER M. 1995. "La Cogola" nuevo sito in riparo d'EI Kowm et le Kebarien Gcometrique. Paleorient, sottoroccia presso Carbonare di Folgaria (Trento). 14/2,270-282. Preistoria Alpina, 31, 53- 59. CHI ERIC! G. 1875. Le selci romboidali. Bailettinodi DALMERI G., BASSETTI M., CUS1NATO A., KOM­ Paletnologia Italiano, I, 2. PATSCHER, HROZNY-KOMPATSCHER M., CILLI C., GUERRESCHI A. 2000. Studio archcozoo­ LANZINGER M. in press. Le pietre dipinte del site logico e tafonomico di un'area di concentrazione di cpigravelliano di Riparo Dalmeri. Campagna di reperti faunistici di Eta epigravettiana (Riparo Ta­ ricerche 2001. Preistoria Alpina, 37. glientc, VR). In: Atti dell/ Convegno Nazionale di DANIEL R., VIGNARD E. I 953. Tableaux synop­ Archeozoologia, Abaco Edizioni, Forli, 141- 150. tiques des principaux microlithes gcometriques du CILLI C., MALERBA G., ROCCI RlS A., GLA­ Tardenoisien franr;:ais. Bulletin de Ia Sociere Prehis­ COBINI G., GUERRESCH I A. in press. Analyse ar­ torique Fran9aise, 50, 314-322. cheozoologique ct taphonomique d'un niveau epi­ DECHELETTE J. 1924. Manuel d'Archeologieprehis­ gravcttien de I'Abri Tagliente (Verene, Italic). Pro­ torique, cellique et galla-romaine. 1: Archeologie ceedings ofXJV U.!.S.P.P. Congress. Prehistorique. Paris, 748 pp. CREMASCHI M. 1990. The loess in Northern and DEWEZ M. 1979. Problematique de !'etude des Central Italy: a loess basin between the Alps and the groupes culturels du Paleolithique final en Belgique. Mediterranean region. Quaderni di Geodinamica In: Sonneville-Bordes de D. (ed.) La fin des temps 104 S. Ferrari & M. Peresani

glaciaires en Europe. Chronostratigraphie et ecolo­ gl iente (Grezzana - Verona), BuJiettino di Palet­ gie des cultures du Paleolithique final. Actes du nologia fta/iana, 86, 19- 58. Colloques Jmernational, C.N.R.S., 271/l , 791 - 793. HEDGES R. E. M., HOUSLEY R. A., BRONK RAM­ ESCALON DE FONTON M. 1953. Bulletin de Ia Soci­ SEY C., van KLINGEN G. I. 1994. Radiocarbon ete Prehistorique Franr;aise, 50, 218-221. dates from the Oxford AMS system: Archaeometry FERRARIS. 2000-2001.1 trapezi del/'£pigraveuiano Datelist 18. Archaeomelly, 36/2, 337- 374. recente dell 1talia nordorientale: cronologia, dis­ HEINZELIN de BRAUCOURT J. 1962. Manuel de Ty­ rribuzione, aspelli mOtfotecnici e afjinita in ambito pologie des industries lithiques. Bruxelles. extralpino. Tesi di laurea inedita, Universita di HTNO UT J. 1973. Classification des rnicrolithes tarde­ Venezia, 201 pp. noisiens du Bassin parisien. Technologic, typo­ FIOCCHI C. 1998. Contributo alia conoscenza del metric ct statistiques. Bulletin de Ia Societe Prehis­ comportamen to simbolico di Homo sapiens. Le con­ torique Fran~aise, 70, 229- 234. chiglie marine nei siti del Paleolitico superiore eu­ KAUFMAN D. 1988. New radiocarbon dates for the ropeo: strategie di approvvigionamento, reti di Geometric Kcbarian. Pafeorient, 14/ 1, 107- 109. scambio, utilizzo. Tesi di Donorato inedita, Consor­ KOFLER W. 1994. Die vegetationscntwicklung im zio Universitario Bologna, Ferrara, Pam1a. Spatpalaolithikum und Mesolithikum im Raume FIORE 1., TAGLIACOZZO A., CASSOLJ P. F. 1998. Trient. Preistoria Alpina, 28/ I, 83- 103. Ibex exploitation at dalmcri rockshelter (TN) and KOBUSIEWI CZ M., KABACINSKI J. 1992. Late Pa­ ''specialized hunting" in the sites of the Eastern laeolithic site at Wojnowo, Zielona G6ra vo ivode­ Alps. In: Alii XXXIII Riunione Scientiflca lstitlllo ship. Fontes Archaeologici Posnanienses, 37, 23- Italiano Preistoria e Protostoria. Preistoria Alpina, 46. 34, 173-184. KOZLOWSKI S. K. 1977. Le Palcolithique final entre FRli-JOHANSEN K. 1919. Une station du plus ancien les Carpathes et 1'0ural. In: Sonneville-Bordes de age de Ia pierre dans Ia tourbiere de Svaerdborg. D. (ed.) La fin des temps glaciaires en Europe. (Trad. da Aarboger 1919) Memoires de Ia Societe Chronostratigraphie et ecologie des cultures du Pa­ Des Antiquaires du Nord, 241- 359. leolithique final. Actes du Colloques International, GAUBERT! A. 1997. II Paleolitico e if Mesolitico C.N.R.S., 27 1/ 1, 510-528. della Toscano. Catalogo della mostra. Lalli ed., LAPLACE G. 1964a. Essai de typologie systematique. Siena, 205 pp. Annali deii'Universita di Ferrara, Sez. XV, Jill, GEHRIG R. 1997. Pollenanalytische Untersuchungen 1-85. zur Vegetationsund Klimageschichte des Val Ca­ LAPLACE G. 1964b. Les subdivisions du Lepto­ monica (Norditalien). Dissertationes Botanicae, lithique ltalien. Essai de typologie analytique. Bul­ 276, 148 pp. lettino di Paletnologia ltaliana, XV, 73, 155-240. GIROD A. 1998. Initial considerations on the WUrmian LEMORINI C. 1997. A functional approach through Tardiglacial malacological records at Dalmeri rock­ trace wear analysis. In C. Baroni, P. Biagi (a cura shelter (Grigno-Trcnto). In: Atti XXXIII Riunione di), Excavations at the high altitude mesolitbic site Scientiflca !still/to Italiano Preistoria e Protostoria. of Laghetti del Crestoso (Bovcgno, Brescia­ Preistoria Alpina, 34, 163- 166. Northem Italy), Commentari deii'Ateneo di Brescia. GROUPE D'ETUDES DE L'EP1PALEOLITHIQUE­ Suppl., 48- 57. MESOLITIQUE (G.E.E.M.) 1969. Epipalcolithi­ LlAGRE J. 2001 2002. Riparo Tagliente (Verona. lta­ que-Mcsolithique. Lcs microlithes geometriques. lia). Analyse techno-economique de I 'industrie lithi­ Bulletin de Ia Societe Prehistorique Fram;aise, 66, que de I'Epigravettien final: exemple du silex «or­ 355-366. ganogena». Mcmoire DEA Prehistoire, Archeolo­ GUERRESCHI A. 1983. Tendenze evolutive in senso gie, Histoire et Civilisations de I'Antiquite et du mesolitico dell'Epigraveniano italico fmale dell' lta­ Moyen- Age, Universite de Provence. lia nord-orientale. Preistoria alpina, 19, 209- 212. LO VETRO D., MARTINI F. 1999- 2000. L'Epi­ GUERRESCH1 A. 1984. II sito epigravettiano diAn­ gravettiano finale siciliano: dinamiche strutturali e dalo (Trento) ed alcunc considerazioni suii 'Epi­ trasformazioni tccno-tipologiche degli strumenti a gravettiano finale ncl Nord ltalia. Preistoria Alpina, dorso e dei geometrici. Rivista di scienze Preis­ 20, 15- 38. toriche, L, 39- 102. GUERRESCHI A. (ed.) 1996. II sito preistorico del Ri ­ MARTINI F. 199 1. L'Epigravettiano dei Pianali paro di Biarzo. Museo Friulano di Storia Nawralc, (Firenze) nel quadro dei complessi tardiglaciali 39, 144 pp. dell'alto versante tirrenico. Swdi e Materiali, VI. 3- GUERRESCHI A., SQUINTAN1 M. B. 1995. La prima 15. officina 1itica epigravettiana (T. lOA) di Riparo Ta- MEZZENA F., PALMA DI CESNOLA A. 1967. Trapezoids and double truncations in the Epigravettian l OS

L'Epigravctliano della Grotta Paglicci nel Gargano tionale, Memoires du Musee de Prehistoire d 'Ile­ (scavi F. Zorzi 1961- 1963). Rivista di Scien:e Pre­ de-France, 7, 73- 86. istoriche, 22, 23- 156. PELLEGRINI G. B., SURJAN N. 1994. Late Pleisto­ MONTOYA C., PERESANI M. in press. Premiers cene geomorphological evolution in the Vallone elements d'analyse technologique de l'industrie Bcllunese. Geografla Fisico e Dinamica Quaterna­ lithiquc du gisement epigravetticn de plein air de ria, 17,67-72. Val Lastari (Haut plateaux dei Sette Comuni, Preal­ PELLEGRINI G. 8., SURlAN N. 1996. Geomorpho­ pes de Ia Venetie). In: Bracco J.P., Montoya C. logical study of the Fadalto landslide, Venetian Pre­ (eds.) Les systhnes techniques lithiques pendantle alps, Italy. Geomo1phology, 15, 337- 350. Tardiglaciaire autour de Ia Mediterranee nord­ PERESANJ M. 1992. Flint exploitation at epigravettian occidentale. Actes de Ia Table Ronde. sites in the Asiago Plateau (Venetian Prealpes). Pre­ MONTOYA C., BROGLIO A., PERESANI M. in istoria Alpina, 28/1, 193- 205. press. Premieres observations technologiqucs sur les PERESANI M. in press. Flint caches and raw material industries lithiques de I'Epigravettieo recent-final economy in the Late Upper Palaeolithic of the East­ D'Italie Nord Orientale. In: Bracco J.P., Montoya C. em Italian Alps. Proceedings VIII International (eds.) Les systemes techniques lithiques pendant le Flint Symposium, Bochum, Germany, 13-17 settem­ Tardiglaciaire autour de Ia Mediterram1e nord­ brc 1999. occidentale. Actes de Ia Table Ronde. PERESANI M., ZIGGlOTTl S., DALMERI G. in MORTILLET A. de 1896. Les petit silex A contours press. Truncations and pseudo-truncations in theRe­ geometriqucs trouves en Europe, Asie et Afrique. cent Epigravettian industries of Nonh-Eastem Italy. Revue d'Etude d'Anthropofogie de Paris, 6, 377- Preistoria Alpina, 37. 405. PERESANI M., BERTOLA S., DE STEFANI M., Dl NANCE J. D. 1971. Functional interpretations from ANASTASIO G. 1999-2000. Bus de La Lum and microscopic analysis. American Amiquity, 36, 361- the Epigravettian occupation of the Venetian Pre­ 366. Alps during the Younger Dryas. Rivista di Scienze NUZHNYJ D. 1989. L'utilisation des microlithcs Preistoriche, 50, 103- 132. geometriques et non geometriques comme arma­ ROZOY J. G. 1967. Essai d'adaptation des methodes tures de projectiles. Bu{{etin de Ia Societe Prehisto­ statistiques a l'Epipah!olithiquc (Mesolithique). rique Fran<;aise, 86/3, 88 96. Liste-type provisoirc et premiers rcsultats. Bulletin NUZHNYJ D. 1993. Projecti le weapons and teclmical de lo Societe Prehistorique Fran9aisc, l , 209- 226. progress in the Stone Age. In: Anderson P., Beyries ROZOY J. G. 1978. Les demier chasscurs. L'Epipaleo­ S., One M., Plisson H. (cds.) Trace et fonction: les lithique en France ct en Belgique. Bullett in de Ia So­ gesres retrouves. Colloquc International de Liege, ciete Archeologique de Champenoise, Tome I, 605 Ed. ERAUL, 50, 41-53. pp. NUZHNY J D. 2000. Development of microlithic pro­ SCHNEIDER R. 1985. Palynologic research in the jectile weapons in the Stone Age. Anthropofogie et Southern and Southeastern Alps between Torino Prehistoire, Ill, 95- 10 I. and Trieste. Dissertationes Botanicae, 87, 83- 103. ODELL G. H. 1978. Preliminarcs d'une analyse fonc­ SCHNEIDER R., TOBOLSKI K. 1983. Palynolo­ tionelle des pointes microlithiques de Bergumer­ gische und stratigraphische Untersuchungen im meer (Pays-Bas). Bulletin de fa Societe Prehistori­ Lago di Ganna (Varese, ltalien). Botanicae Helveti­ que Fram;aise, 75, 37-49. cae, 93: 115- 122. OTTE M. 1983. Le Paleolithique de Belgique; essai de SIRET L. 1893. L 'Espagne prehistorique. Revue des synthese. L 'Anthropologie, 87/3, 291 - 321. questions scientiflques, 78 pp. PALMA Dl CESNOLA A. 1993. 11 Paleolitico superi­ TAGLlACOZZO A., CASSOLI P. F. 1992. La macro­ ore in ltafia. lntroduzione a{{o swdio. Fircnze, 575 faune de I'Abri Soman (Val d'Adige-Italie). Preis­ pp. loria Alpina, 28/ 1, 181- 192. PALMA Dl CESNOLA A., BlETTI A., GAUBERT! TENSORE R Le, J. M. I 981. Le Paleolithique de A. 1983. L'Epigravetticn cvolue et final dans les I' Agenais. C.N.R.S., Cahiers du Quaternaire, Paris, Pouilles. Rivista di scienze Preistoriche, 38, 267- 3. 300. TIXIER J. 1963. Typologie de I'Epipah!olithique du PELEGRIN J. 2000. Les techniques de debitage lami­ Maghreb. Memoires du Centre de Recherches An­ naire au Tardiglaciairc: criteres de diagnose et thropologiques, Prehistoriques et Ethnographiques. quelques rcncxions. In: Valentin B., Bodu P., Chris­ Arts et Metiercs, Paris, I, 209 pp. tensen M. ( cds.) L 'Europe centrale et septentrionale TRlNGHAM R., COOPER G., ODELL G., VOYTEK au Tardiglaciare. Actes de Ia Table-Ronde lntema- B., WHITMAN A. 1974. Experimentation in the 106 S. Ferrari & M. Peresani

formation of edge damage: a new approach to lithic Wojnowo, Zielona G6ra voivodeship. Fontes Ar­ analysis. Journal ofField Archaeology, I, I7 1- 196. chaeologici Posncmienses, 37, 47-63. VOGELJ. C., WATERBLOCK H. T. 1963. Groningcn ZALIZNY AK L. 1999. Terminal Palaeolithic of Radiocarbon Dates IV. Radiocarbon, 5, 163- 202. Ukraine, Belarus and Lithuania. In: Kobusiewicz WICK L. 1996. Late-Glacial and early-Holocene pa­ M., Kozlowski J. K. (cds.) Post-Pleniglacial re­ laeoenvironments in Brianza, Northern Italy. Italian colonization of the Great European Lowland Pro­ Journal of Quaternary Sciences, 9 (2), 653- 659. ceedings of the International Conference. VIII WTNIARSKA-KABACrNSKA M. 1992. Functional Comm. UISPP, Folia Quaternaria. Krakow, 333- analysis of flint tools from Late Palaeolithic site at 361.