lable at ScienceDirect

Quaternary International xxx (2014) 1e9

Contents lists avai

Quaternary International

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

The Pirro Nord site (Apricena, Fg, Southern Italy) in the context of thefirst European peopling: Convergences and divergences

Marta Arzarello a, *, Carlo Peretto a, Marie-H�el�ene Moncel b

a Dipartimento di Studi Umanistici, Universit�a degli Studi di Ferrara, C.so Ercole I d'Este 32, 44100 Ferrara, Italyb UMR 7194, CNRS, D�epartement de Pr�ehistoire, Mus�eum national d'Histoire Naturelle, 1 Rue Ren�e Panhard, 75013, Paris, France

a r t i c l e i n f o

Article history:Available online xxx

Keywords:First European peoplingMode 1 technologyPirro Nord site

* Corresponding author.E-mail address: marta.arzarello@unife.it (M. Arzar

http://dx.doi.org/10.1016/j.quaint.2014.08.0511040-6182/© 2014 Elsevier Ltd and INQUA. All rights

Please cite this article in press as: Arzarellopeopling: Convergences and divergences, Q

a b s t r a c t

Pirro Nord, located in Apricena, Southern Italy, represents one of the earliest records of hominindispersal into Europe. The site is dated on a biochronological basis between 1.6 and 1.3 Ma and a sig-nificant vertebrate assemblage (Pirro Nord Faunal Unit) is associated with lithic artefacts which showmany similarities with more or less contemporaneous lithic assemblages in Europe and with the AfricanMode 1.

Since 2006, 340 artefacts have been found in the Pirro 13 fissure, and the lithic series is considered as ahomogenous assemblage in terms of taphonomical, technological and raw material characteristics. Flintis the main raw material and the available and exploited flint cobbles and pebbles have different mor-phologies and sizes. The raw material has been collected in a secondary position near the actual locationof the deposit.

Two main reduction sequences have been performed by hominins in Pirro Nord: an “opportunistic”debitage, with multiple striking platforms in order to produce flakes of different morphologies includingsome “debordant” flakes and at least one cutting edge, and a “centripetal” or convergent debitage for theproduction of flakes with convergent cutting edges. The “opportunistic” approach has been adopted forthe reduction of the largest flint cobbles and the polyhedral pebbles/cobbles. This method is consideredas an adaptation to the raw material shape. The debitage is always made by direct percussion and hardhammer. Most of the cores are abandoned non-exhausted. In some rare cases, the use of flint pebbles ofpoor quality is also documented and seems to be the result of a very opportunistic and expeditiousbehaviour. Four flakes are retouched and are from an advanced phase of the debitage.

The patterns from Pirro Nord fit well with the technological features provided by most of the EuropeanMode 1 sites. Some peculiar features may be observed, possibly due to the main use of flint. The site withthe highest similarities with Pirro Nord is C�a Belvedere di, possibly because the raw materials are ovoidflint pebbles. The lithic strategies at Pirro Nord are characterized, (1) by “opportunistic” and expedientbehaviour adapted to the raw material constraints, suggested by the very short reduction sequences andby the non-exhaustion of the cores and, (2) by a series of products (pointed flakes) that appears toindicate a presumable intentionality.

© 2014 Elsevier Ltd and INQUA. All rights reserved.

1. The Pirro Nord site

The site of Pirro Nord (sometimes also known as Cava Pirro orCava Dell'Erba) is located at the northwestern margin of the Gar-gano promontory, close to the village of Apricena (Foggia District,Apulia, Southern Italy). It represents one of the earliest records ofEuropean peopling as it is dated, on a biochronological basis, to1.6e1.3 Ma (Arzarello et al., 2012, 2009, 2007).

ello).

reserved.

, M., et al., The Pirro Nord siuaternary International (2014

The site is located inside an active limestone quarry and thelithic industry was found within a karst fissure. The P13 shaft issituated at the top of theMesozoic limestone formation and, duringthe Pleistocene, was part of a very complex interconnected karstsystem. It is the result of a dissolution that was effective along thefractured corezone of the Pliocene fault that bordered the “Apri-cena horst” to the south (Pavia et al., 2012).

Each fissure containing Villafranchian paleontological remainshas been named “P” followed by a progressive number. Inside theP13 fissure, the only fissure with paleontological remains associ-ated with lithic industries, the lithic assemblage was found in as-sociation with vertebrate fossils of the Pirro Nord Faunal Unit

te (Apricena, Fg, Southern Italy) in the context of the first European), http://dx.doi.org/10.1016/j.quaint.2014.08.051

Fig. 1. Localization of the Pirro Nord site inside the “Cave dell'Erba” quarry and synthetic stratigraphy of the excavated area.

M. Arzarello et al. / Quaternary International xxx (2014) 1e92

(Abbazzi et al., 1996; Gliozzi et al., 1997). The faunal assemblage ischaracterized by the earliest occurrence of Bison degiulii, Capreolussp., Equus altidens, andMeles meles, by the presence of 20 species ofamphibians and reptiles (Delfino and Bailon, 2000), 47 species ofbirds (Bedetti, 2003) and over 40 species of mammals such asStephanorhinus sp., Pachycrocuta brevirostris, Homotherium latidens,Axis sp., Praemegaceros obscurus and Mammuthus meridionalis.Among rodents, the arvicolid vole species Allophaiomys ruffoi ispresent (Arzarello et al., 2009).

Paleoenvironmental reconstructionwas done on the basis of thefaunal remains found at the site. The presence of birds Otis tarda,Tetrax tetrax and Pterocles oreintalis, together with other species of

Please cite this article in press as: Arzarello, M., et al., The Pirro Nord sipeopling: Convergences and divergences, Quaternary International (2014

Anatidae and Charadriiformes, indicates that there was an openenvironment, tending to dry, but with a seasonal wetland.Furthermore, the significant presence of Alaudidae indicates thatopen areas were characterized by low-type vegetation (Bedetti,2003; Arzarello et al., 2009).

The stratigraphy of the P13 fissure is characterized by aclayesandymatrix containingmany blocks of limestone with sharpedges and lacking preferential orientation. The sedimentary fillingis derived from the top, by gravity, and follows the position of thelarge limestone blocks that made up the skeleton of the fissure. Thesediments were deposited in a chaotic way (Fig. 1). The lithic in-dustry and the faunal remains were found arbitrarily dispersed

te (Apricena, Fg, Southern Italy) in the context of the first European), http://dx.doi.org/10.1016/j.quaint.2014.08.051

Table 1Composition of the lithic assemblage of Pirro Nord. Thematerial comes from the stratigraphic units A (9%), B (12%), C(19%) and D (60%); the different amounts of material foundin the stratigraphic units are function of the differentthickness of each SU. In the table we have named as “debris”the fragments (coming indisputably from a debitage pro-cess) that cannot be attributed to a clear technologicalcategory. The flakes-cores were divided on the basis of theperformed core technology.

Type N�

Cores 371 streaking platform 62 streaking platforms 83 > streaking platforms 3Centripetal exploitation 12Indet./fragment 8Flakes 231Unipolar removals 81Bipolar removals 10Orthogonal removals 14Crossed removals 49Centripetal removals 39Kombewa l.s. 9Indet. 29Debris 72Tot. 340

M. Arzarello et al. / Quaternary International xxx (2014) 1e9 3

within the whole sequence, and no particular concentration wasrecorded. All materials are characterized by the same taphonomicalterations, mostly FeeMn concretions. The GIS study applied tobones and lithics found in the fissure clearly showed that uniformorientation patterns and random distribution of materials areprobably correlated with a rapid, chaotic process such as debris-flow (Giusti, 2013).

2. Chronology

The chronological attribution of P13 is at present based onbiochronology. The fossils are attributed to the late Early Pleisto-cene and the association represents the last Faunal Unit of theVillafranchian Mammal Age in the Italian biochronological scale(Gliozzi et al., 1997). The Pirro Nord Faunal unit can therefore beattributed to an interval between 1.6 and 1.3 Ma (Arzarello et al.,2007) or to a slightly lower age (around 1.3 Ma) according tosome authors (Masini and Sala, 2011, 2007).

Some paleomagnetic measurements were made in the fissureP10 (near to the P13 fissure and characterized by the same FaunalUnit). The preliminary results, together with the palaeontologicaland geological data, suggest that the Pirro Nord sediments can bereferred to the Matuyama, post-Olduvai reverse polarity Chron(Pavia et al., 2012).

3. The lithic assemblage

Since 2006, 340 artefacts related to an anthropic activity havebeen found in the Pirro 13 fissure. The lithic industry is consideredas a homogenous assemblage showing a high degree of similarity interms of taphonomical and technological characteristics, and rawmaterial (Arzarello et al., 2012).

The whole series is characterized by a very good state of pres-ervation. The artefacts have sharp edges and do not show macro-scopic evidence of transportation: 35% carry traces of manganeseon their surface, as observed on more than 70% of the faunal re-mains. Post-depositional fractures have affected 20% of the lithicmaterial and they are probably due to the falling into the fissure.

Inside the P13 fissure, unknapped flint was also found, but theaverage size of the small pebbles is considerably smaller than the

Fig. 2. Dimensional range of non-knapped flint and knapped flint found in the PirroNord site (whisker length: one sigma, whisker type: standard error, y-axis: maximumsize). In the knapped elements we consider only the artifacts that could give in-dications of the initial size of raw material.

Please cite this article in press as: Arzarello, M., et al., The Pirro Nord sipeopling: Convergences and divergences, Quaternary International (2014

anthropic artefacts. The unknapped flint is 5e20 mm long (Fig. 2)and forms small rounded cortical pebbles. These small pebbles havenot been considered as manuports due to their limited size, andcould be brought to the fissure by natural processes. Even if theartefacts were found in a secondary deposition, the dimensionalanalysis shows the consistency of the assemblage where all phasesof the reduction sequences are represented: from decortication,passing through the production of small waste, to core abandon-ment (Table 1).

3.1. Raw material

The available and exploited flint cobbles and pebbles havedifferent morphologies and sizes. The raw material has beencollected in a secondary position (river beds or slope deposits) nearthe actual location of the deposit, no more than 7 Km from P13.Four main different types of flint were exploited (all coming fromthe Gargano Cretaceous succession): (1) brown oolitich flint, (2)grey homogeneous flint, (3) grey bedded flint and a black flint. Thefirst three types of flint are of excellent quality, in terms of knappingsuitability, while the black flint has several inner fractures of tec-tonic origin.

There is no clear relationship between the kind of flint and themethod of debitage adopted for the reduction. However, we mayobserve a relationship between the pebble/cobble morphology andsize, and the flaking method. The collected pebble/cobble di-mensions range between 3 and 10 cm in diameter, with a group ofcobbles with a maximum of 5 cm length.

3.2. Reduction sequences

The decortication phase is attested by 133 flakes, among which31 of them represent the first initial phase of pebble knapping. Theyare fully cortical flakes. The pebbles' opening was made in mostcases by direct hard hammer percussion, but a bipolar percussion isalso attested when the spherical shape of the raw material did notoffer suitable angles for starting the flaking by direct percussion.

The decortication cannot be seen as an independent segment ofthe reduction sequence, simply aimed at producing cortical re-movals. During this phase, cortical flakes (38%) with a sharp edge

te (Apricena, Fg, Southern Italy) in the context of the first European), http://dx.doi.org/10.1016/j.quaint.2014.08.051

M. Arzarello et al. / Quaternary International xxx (2014) 1e94

opposite to the cortical edge are obtained. Two main reductionsequences have been adopted by hominins at Pirro Nord: (1) an“opportunistic” debitage used to exploit multiple striking platforms(from 1 to 5) and flaking surfaces aiming producing flakes ofdifferent morphologies but always with at least one cutting edge. Itcould be compared in more simple to some S.S.D.A. cores describedby Forestier (1993), (2) a “centripetal” debitage for, in most cases,the production of flakes with convergent cutting edges.

3.2.1. Debitage by opportunistic strategiesThe “opportunistic” reduction sequence has been adopted for

the reduction of the largest flint cobbles and the polyhedral peb-bles/cobbles (Figs. 3 and 4). This method is considered as anadaptation to the rawmaterial shape in order to obtain the greatestquantity of blanks with a sharp cutting edge and with the leastnumber of gestures. The debitage is always made by direct per-cussion and hard hammer and all flakes show a clear impact point.In most cases, a pronounced bulb and a cortical or plain buttcharacterize the technical aspect of the flakes, and, occasionally, thebulb is flat and the platform is linear when the stroke is given nearthe cornice. This flat feature does not seem to be due to wedgingfractures with strong percussion or to post-depositional processes.The inferior face of the flake show typical features of detachment.

The flakes' morphologies derive from the initial coremorphology. Most are quadrangular or trapezoidal with a sharpangle close to 80e85�. The flaking method also led to the produc-tion of several debordant flakes. The knapping is often conducted touse the natural convexities. Debordant flakes total 30% of the“opportunistic production”. We can speculate that the back on oneedge was useful for grasping.

Each debitage surface produces few removals and was just usedonce. Flakes with unipolar scars are the most frequent with some

Fig. 3. Core with 3 striking platforms managed by unipolar debitage. The core is not exhaustthe UNIFE TekneHub laboratory by Julie Arnaud with a Scanner Breuckmann smartSCAN3D e

with OPTOCAD 2011.

Please cite this article in press as: Arzarello, M., et al., The Pirro Nord sipeopling: Convergences and divergences, Quaternary International (2014

cortical patches (53% of the flakes from what we have named“opportunistic debitage”) (Fig. 5).

The rawmaterial is almost never fully exhausted andmost of thecores were abandoned while still usable. The reduction sequenceson the largest cobbles are usually longer, but they always stopbefore the full exploitation of the core. These features cannot bedue to the size of the collected blanks, as small-sized blocks havealso been used and their dimensions are smaller than the final flakesizes produced from larger blocks.

In some rare cases at Pirro Nord, the use of flint pebbles of poorquality (rich in inner fractures of tectonic origin) is also docu-mented. This data cannot be related to a lack of rawmaterial, whichmay be ruled out by the evidence that most of the cores were notfully exploited. It seems to be the result of a very opportunistic andexpeditious behaviour.

3.2.2. A more complex method: centripetal debitageWhat is termed “centripetal debitage” has been almost

exclusively performed on spherical/ovoid small pebbles andcobbles (maximum length between 20 and 70 mm). All thosecores (n ¼ 12) led to the production of medium-small-sizedflakes, mostly debordant and with two convergent cuttingedges. This method may be considered as the best/easiest wayfor exploiting small ovoid pebbles, but the recurrence and“standardization” in the blanks production may also be evidenceof a voluntary technical choice made by the Pirro Nord knappers(Figs. 6 and 7).

The reduction sequence starts with the opening of the pebblesurface by direct or bipolar percussion. A small cap is removed first.The second removal is often managed perpendicularly to the firstone and is later used as a striking platform in order to begin thecentripetal flaking. When this cortical removal is absent, it meansthat the debitage has been started by a natural striking platform.

ed and has produced a minimum of 12 flakes of middle/big sizes. The scan was made inDuo, with a field of View of 225 mm and a resolution of ±9 m. The data were processed

te (Apricena, Fg, Southern Italy) in the context of the first European), http://dx.doi.org/10.1016/j.quaint.2014.08.051

Fig. 4. Flakes issued from an opportunistic debitage. The black colour is due to the deposit of Fe/Mn oxides. All flakes are coming from the SU D.

Fig. 5. Dimensional analysis of complete flakes issued from an opportunistic debitage.The graphic shows, in relation to flakes with unipolar removals, that the final phase ofthe reduction sequence is often made by the opening of a new debitage surface.

M. Arzarello et al. / Quaternary International xxx (2014) 1e9 5

Please cite this article in press as: Arzarello, M., et al., The Pirro Nord sipeopling: Convergences and divergences, Quaternary International (2014

The flaking surface is manufactured by 3e4 flakes, detachedclose to each other's and usually removing four-fifths of the initialcore surface. Then the core is abandoned not exhausted. The rawmaterial would allow a second phase of production. The paralleldirections of some scars on cores suggest that this centripetaldebitage could also be described as a “convergent unipolar debit-age” with independent series of removals.

The Kombewa l.s. (Owen, 1938) debitage is also attested both forpebbles and cobbles. The five largest flakes, in most casescompletely cortical, are produced by centripetal debitage on theventral face of the cores on flakes or by unipolar debitage using theventral face of the core on flake as a striking platform.

3.3. The retouch

The modification of the cutting edges by retouch is very rare atPirro Nord (Fig. 8). It is attested by four flakes, all broken and twoshowing a bending fracture. The retouched pieces are one notch,one denticulate and two side scrapers, both with an inverseretouch, one on the distal edge of the flake and the other one on thelateral edge. All four retouched blanks are issue from an opportu-nistic debitage with multiple flaking surfaces. We cannot make anycomments on the dimensions of the retouched flakes, as they are allbroken. As they have no cortex, however, we can admit that theseblanks are issue from an advanced phase of the debitage, andpossibly from the knapping of medium to large-sized cobbles.Pseudo-retouches due possibly to utilization are also attested on afew flakes.

te (Apricena, Fg, Southern Italy) in the context of the first European), http://dx.doi.org/10.1016/j.quaint.2014.08.051

Fig. 6. Flakes (on the right) and core attesting a centripetal debitage. The morphology of the flakes is a d�ejet�e point with an opposite back (drawings D. Aureli).

M. Arzarello et al. / Quaternary International xxx (2014) 1e96

4. The Pirro Nord site and the first European peopling

Discussions about the first European peopling are nowadayssupported by several well-dated sites providing detailed informa-tion about the technical behaviour adopted by the hominins thatfirst populated the continent. Apart from Pirro Nord, the oldestrecord is Damnisi in Georgia, dated to 1.8 Ma, older than Pirro Nord(Baena et al., 2010; Mgeladze et al., 2011; Lordkipanidze et al.,2013). It is followed by the Spanish sites of Sima del Elefante atAtapuerca, Barranco Le�on and Fuente Nueva 3 at Orce, dated toaround 1.2e1.0 Ma (Bernal and Moncel, 2004; Par�es et al., 2006;Carbonell et al., 2008; Barsky et al., 2010; Toro-Moyano et al.,2011, 2009; Mosquera et al., 2013), and the French sites of Pont deLavaud dated to 1.1 Ma (Despri�ee et al., 2010, 2009, 2006) and LeVallonnet dated to 1.1 Ma (Cauche et al., 2001; Cauche, 2009; DeLumley et al., 2009). In the Italian peninsula, C�a Belvedere diMontepoggiolo attesting an old occupation is dated to about 1.0 Ma(Peretto et al., 1998; Peretto, 2005; Arzarello and Peretto, 2010).

The patterns obtained at Pirro Nord fit well with the techno-logical features provided by these sites, but some peculiar featuresmay be observed. In general we can say that the lithic strategies ofPirro Nord are characterized, (1) on the one hand, by “opportu-nistic” and expedient behaviour adapted to the raw material con-straints suggested by very short reduction sequences and by thenon-exhaustion of the cores and (2) on the other hand, by a se-ries of products (pointed flakes) that appears to indicate a pre-sumable intentionality in terms of the flaking purpose.

The older Dmanisi site, and especially Dmanisi layer IV (Baenaet al., 2010) shares technological similarities with Pirro Nord. Inboth sites, strategies of debitage are simple, short and influenced bythe initial stone morphology. Whereas at Pirro Nord the quality ofthe raw material did not seem to have affected the technologicalchoices, in Dmanisi the raw materials of the best quality were used

Please cite this article in press as: Arzarello, M., et al., The Pirro Nord sipeopling: Convergences and divergences, Quaternary International (2014

for complex flaking methods. The core technology is similar in bothsites but it shows a greater variability at Dmanisi, possibly due tothe greatest number of pieces or to various behaviours into theMode 1 hominin groups.

If we compare the lithic series of Pirro Nord with the series ofBarranco Le�on and Fuente Nueva (Toro-Moyano et al., 2011), in bothcases the same lithic strategies are practiced (unipolar, bipolar,multidirectional and centripetal). We observe a strong derivationfrom the initial morphology of the raw material, the use of bothdirect and on-anvil percussion (although rare at Pirro Nord), theprocurement of local raw material and the production of shortcutting edges. The main difference between the two sites exists inthe management of limestone blocks associated to flint in theSpanish sites. In Pirro Nord limestone is not exploited, also if pre-sent in the environment, probably precluding the presence ofchopper-like tools at this site. Regarding the flake edge modifica-tion, it is still doubtful at the Spanish sites but seems to be attestedin a very few cases at Pirro Nord.

Compared to Sima del Elefante, 1.2 Ma (Carbonell et al., 2008;Mosquera et al., 2013), the core technology used at Pirro Nordseems to be more complex. At Sima del Elefante, only one knappingmethod is described (unipolar longitudinal with a single strikingplatform). The stone tools published from Sima del Elefante, how-ever, are few at this moment and more similarities may emerge inthe future.

The lithic assemblage of Pont de Lavaud (Despri�ee et al., 2010,2009) is the most abundant series among the oldest Europeansites. In this site, the production is also strongly conditioned by theraw material (morphology and petrography) which is quartz, andthe reduction sequences are rather short. The techniques used aredirect percussion by hard hammer and bipolar percussion on anvil.The bipolar percussion is employed more often than at Pirro Nord,possibly due to the use of quartz and the knapping methods are

te (Apricena, Fg, Southern Italy) in the context of the first European), http://dx.doi.org/10.1016/j.quaint.2014.08.051

Fig. 7. Centripetal debitage; 1,9: cores; 2e8: fakes; A: outline of centripetal production, the dashed arrows indicate not mandatory steps.

M. Arzarello et al. / Quaternary International xxx (2014) 1e9 7

mainly unipolar, bipolar, multidirectional and, in some cases, cen-tripetal. As in the Spanish sites of Barranco Le�on and Fuente Nueva3, a rare shaping is also attested. The largest similarities with PirroNord are for the retouched flakes, modified on their longest cuttingedge and by the presence of “pointed-end” products. The produc-tion of “pointed” products obeys to different reduction processes atPirro Nord (“centripetal” or convergent debitage) and at Pont deLavaud (shaping) but it can attest to a specific and common optionchosen for a precise aim. Unfortunately, use wear analysis could notbe performed for any of the sites, and therefore it is impossible todiscuss more hypotheses.

In the Vallonnet cave, the lithic pieces (about 100) are made onlocal raw materials, mostly limestone and a lesser extent flint and

Please cite this article in press as: Arzarello, M., et al., The Pirro Nord sipeopling: Convergences and divergences, Quaternary International (2014

quartzite, and show again short reduction sequences (De Lumleyet al., 2009). At this site as well, shaping in non-siliceous raw ma-terials is associated with a core technology and the main reductionmethods are based on unipolar, bipolar and multidirectional stra-tegies. As for Pirro Nord and the other older European sites, thereduction sequences are strongly adapted to raw materialconstraints.

The site which shows the highest similarities with Pirro Nord isC�a Belvedere di Montepoggiolo (Peretto, 2005; Arzarello andPeretto, 2010), possibly because in both sites the raw materialsare spherical/ovoid flint pebbles/cobbles. The core technology andthe techniques are the same and in both sites shaping is notattested.

te (Apricena, Fg, Southern Italy) in the context of the first European), http://dx.doi.org/10.1016/j.quaint.2014.08.051

Fig. 8. Retouched flake coming from the SU D, square C4. The retouch is located on thedistal edge of the flake and is denticulated. The retouched blank is a flake from theS.S.D.A. core technology. It is a distal fragment and the breakage is probably due topost-depositional processes.

M. Arzarello et al. / Quaternary International xxx (2014) 1e98

The technological features of the earliest European lithic in-dustries indicate a fairly homogeneous behaviour, even if somedifferences may be noted. The reduction sequences are short andlargely influenced by the nature and the morphology of the rawmaterials which is local.

The major differences are the presence/absence of shaping ev-idence and retouched tools and may be related exclusively to thetype and shape of the rawmaterials. Shaping is absent when flint isthe only stone. From a general point of view, we can consider theselithic productions as rather “opportunistic”, understood as the ca-pacity to exploit opportunities offered by the rawmaterials, readilyadapted to the circumstances in order to reach an efficient result.

Acknowledgements

Fieldwork was financially supported by Universit�a degli Studi diFerrara and Apricena Municipality. The excavations were possiblethanks to the concession by the “Soprintendenza Archeologica dellaPuglia” and the “Ministero dei Beni e delle Attivit�a Culturali” andthanks to the collaboration of Dott.Ssa Anna Maria Tunzi. Theexcavation was also possible thanks to the kind disposability ofGaetano and Franco dell'Erba. We thank Juelie Aranud, DomenicoGiusti, Gabriele Berruti, Sara Daffara, Claudio Berto, Elisa Luzi,Sandro Caracausi, Fabio Buccheri, Cecilia Buonsanto, Giorgio Ven-tricelli, Katia Russi, Angela Leggeri, Brunella Muttillo and GiuseppeLembo for their indispensable cooperation during the fieldwork atPirro Nord. We also thank Prof. Benedetto Sala for advice about thechronology. Last, but not least, a particular thanks goes to the morethan 200 people from various countries who attended the exca-vation campaigns, without whom this work would not have beenpossible. Finally, we would like to thank the two reviewers whohelp us to enhance this paper.

References

Abbazzi, l., Benvenuti, M., Boschian, G., Dominaci, S., Masini, F., Mezzabotta, C.,Rook, L., Valleri, G., Torre, D., 1996. The Neogene and Pleistocene succession andthe mammal faunal assemblages of an area between Apricena and PoggioImperiale (Foggia). Memorie della Societ�a Geologica Italiana 51, 383e402.

Arzarello, M., Peretto, C., 2010. Out of Africa: the first evidence of Italian peninsulaoccupation. Quaternary International 223e224, 65e70.

Please cite this article in press as: Arzarello, M., et al., The Pirro Nord sipeopling: Convergences and divergences, Quaternary International (2014

Arzarello, M., Marcolini, F., Pavia, G., Pavia, M., Petronio, C., Petrucci, M., Rook, L.,Sardella, R., 2007. Evidence of earliest human occurrence in Europe: the site ofPirro Nord (Southern Italy). Naturwissenschaften 94, 107e112.

Arzarello, M., Marcolini, F., Pavia, G., Pavia, M., Petronio, C., Petrucci, M., Rook, L.,Sardella, R., 2009. L'industrie lithique du site Pl�eistoc�ene inferieur de Pirro Nord(Apricena, Italie du sud): une occupation humaine entre 1,3 et 1,7 Ma.Anthropologie 113, 47e58.

Arzarello, M., Pavia, G., Peretto, C., Petronio, C., Sardella, R., 2012. Evidence of anEarly Pleistocene hominin presence at Pirro Nord (Apricena, Foggia, southernItaly): P13 site. Quaternary International 267, 56e61.

Baena, J., Lordkipanidze, D., Cuartero, F., Ferring, R., Zhvania, D., Martín, D., Shelia, T.,Bidzinashuili, G., Roca, M., Rubio, D., 2010. Technical and technologicalcomplexity in the beginning: the study of Dmanisi lithic assemblage. Quater-nary International 223e224, 45e53.

Barsky, D., Celiberti, V., Cauche, D., Gr�egoire, S., Leb�egue, F., de Lumley, H., Toro-Moyano, I., 2010. Raw material discernment and technological aspects of theBarranco Le�on and Fuente Nueva 3 stone assemblages (Orce southern Spain).Quaternary International 223e224, 201e219.

Bedetti, C., 2003. Le avifaune fossili del Plio-Pleistocene italiano: sistematica,paleoecologia ed elementi di biocronologia. Universit�a la Sapienza, Roma.

Bernal, M.T., Moncel, M.-H., 2004. Contribution �a l'�etude de la technologie duPal�eolithique « archaïque » du sud de l'Europe selon le Syst�eme Logique Ana-lytique (SLA). Application aux sites du Vallonnet (Roquebrune-Cap-Martin,France), de Gran Dolina TD6 (Burgos, Espagne), de Ca'Belvedere. Anthropologie108, 307e329.

Carbonell, E., Bermudez de Castro, J., Pares, J., Perez-Gonzalez, A., Cuenca-Bescos, G.,Olle, A., Mosquera, M., Huguet, R., van der Made, J., Rosas, A., Sala, R.,Vallverdu, J., Garcýa, N., Granger, D.E., Martinon-Torres, M., Rodrýguez, X.,Stock, G.M., Verges, J., Allue, E., Burjachs, F., Caceres, I., Canals, A., Benito, A.,Dýez, C., Lozano, C., Mateos, A., Navazo, M., Rodrýguez, J., Rosell, J., Arsuaga, J.L.,Bermúdez de Castro, J.M., Par�es, J.M., P�erez-Gonz�alez, A., Cuenca-Besc�os, G.,Oll�e, A., Vallverdú, J., García, N., Martin�on-Torres, M., Rodríguez, X.P.,Verg�es, J.M., Allu�e, E., C�aceres, I., Díez, C., Lozano, M., Rodríguez, J., 2008. Thefirst hominin of Europe. Nature 452, 456e470.

Cauche, D., 2009. Les strat�egies de d�ebitage dans les industries lithiques archaïquesdes premiers habitants de l’Europe. Anthropologie 113, 178e190.

Cauche, D., Celiberti, V., Barsky, D., Notter, O., Bidittu, I., Lumley, H. de, 2001. Les plusanciennes industries lithiques du Latium, Italie. In: du, L.S. (Ed.), Congr�es, XIV.Congr�es UISPP, Li�ege, pp. 49e57.

De Lumley, H., Barsky, D., Cauche, D., 2009. Les premi�eres �etapes de la colonisationde l’Europe et l'arriv�ee de l'Homme sur les rives de la M�editerran�ee. Anthro-pologie 113, 1e46.

Delfino, M., Bailon, S., 2000. Early Pleistocene herpetofauna from Cava Dell'Erba andCava Pirro (Abulia, southern Italy). Herpetology Journal 10, 95e110.

Despri�ee, J., Gageonnet, R., Voinchet, P., Bahain, J.J., Falgu�eres, C., Varache, F.,Courcimault, G., Dolo, J.-M., 2006. Une occupation humaine au Pl�eistoc�eneinf�erieur sur la bordure nord du Massif Central. Comptes Rendu PALEVOL 5,821e828.

Despri�ee, J., Voinchet, P., Gageonnet, R., D�epont, J., Bahain, J.-J.J.-J., Falgu�eres, C.,Tissoux, H., Dolo, J.-M.J.-M., Courcimault, G., 2009. Les vagues de peuplementshumains au Pl�eistoc�ene inf�erieur et moyen dans le bassin de la Loire moyenne,r�egion Centre, France. Apports de l'�etude des formations fluviatiles. Anthro-pologie 125e167 doi:10.101.

Despri�ee, J., Voinchet, P., Tissoux, H., Moncel, M.-H., Arzarello, M., Robin, S.,Bahain, J.-J., Falgu�eres, C., Courcimault, G., D�epont, J., Gageonnet, R., Marquer, L.,Messager, E., Abdessadok, S., Puaud, S., 2010. Lower and middle Pleistocenehuman settlements in the Middle Loire River Basin, Centre Region, France.Quaternary International 223e224, 345e359.

Forestier, H., 1993. Le Clactonien: Mise en application d'une nouvelle m�ethode ded�ebitage s'inscrivant dans la variabilit�e des syst�emes de production lithique duPal�eolithique ancien. Paleo 5, 53e82.

Giusti, D., 2013. A GIS approach to the Paleolithic site of Pirro Nord, Apricena, Italy:from data management to spatial analysis through a Wiki documentation. In:ESHE. European Society for the Study of Human Evolution, Vienna, p. 96.

Gliozzi, E., Abbazzi, L., Azzaroli, A., Caloi, L., Capasso Barbaro, L., Di Stefano, G.,Esu, D., Ficcarelli, G., Girotti, O., Kotsakis, T., Masini, F., Mazza, P., Mezzabotta, C.,Palombo, M.R., Petronio, C., Rook, L., Sala, B., Sardella, R., Zanalda, E., Torre, D.,1997. Biochronology of selected mammals, molluscs and ostracods from theMiddle Pleistocene to the Late Pleistocene in Italy. Rivista Italiana di Paleon-tologia e Stratigrafia 90, 369e388.

Lordkipanidze, David, Ponce de Le�on, Marcia S., Margvelashvili, Ann, Rak, Yoel,Rightmire, G. Philip, Vekua, Abesalom, Zollikofer, P. E. Christoph, 2013.A complete skull from Dmanisi, Georgia, and the evolutionary biology of EarlyHomo. Science 342, 326e331.

Masini, F., Sala, B., 2007. Late Pliocene and Pleistocene small mammal chronology inthe Italian peninsula. Quaternary International 160, 43e56.

Masini, F., Sala, B., 2011. Considerations on an integrated biochronological scale ofItalian Quaternary continental mammals. Il Quaternario 24, 193e198.

Mgeladze, A., Lordkipanidze, D., Moncel, M.-H., Despriee, J., Chagelishvili, R.,Nioradze, M., Nioradze, G., 2011. The hominin occupations at Dmanisi site,Georgia, Southern Caucasus: lithic technical abilities in the first settlements ofEurasia. Journal of Human Evolution 60 (5), 571e596.

Mosquera, M., Oll�e, A., Rodríguez, X.P., 2013. From Atapuerca to Europe: tracing theearliest peopling of Europe. Quaternary International 295, 130e137.

Owen, W., 1938. The Kombewa Culture. Kenya Colony, pp. 203e205.

te (Apricena, Fg, Southern Italy) in the context of the first European), http://dx.doi.org/10.1016/j.quaint.2014.08.051

M. Arzarello et al. / Quaternary International xxx (2014) 1e9 9

Par�es, J.M., P�erez-Gonz�alez, A., Rosas, A., Benito, A., Bermùdez de Castro, J.M.,Carbonell, E., Huguet, R., 2006. Matuyama-age lithich tools from the Sima delElefantesite, Atapuerca (northernSpain). JournalofHumanEvolution50,163e169.

Pavia, M., Zunino, M., Coltorti, M., Angelone, C., Arzarello, M., Bagnus, C., Bellucci, L.,Colombero, S., Marcolini, F., Peretto, C., Petronio, C., Petrucci, M., Pieruccini, P.,Sardella, R., Tema, E., Villier, B., Pavia, G., 2012. Stratigraphical and palae-ontological data from the Early Pleistocene Pirro 10 site of Pirro Nord (Puglia,south eastern Italy). Quaternary International 267, 40e55.

Peretto, C., 2005. The first peopling of Southern Europe: the Italian case. Palevol 5,283e290.

Peretto, C., Amore, O., Antoniazzi, A., Bahain, J.J., Cattani, L., Cavallini, E., Esposito, P.,Falgu�eres, C., Hedley, C., Laurent, I., Le Breton, V., Longo, L., Milliken, S.,

Please cite this article in press as: Arzarello, M., et al., The Pirro Nord sipeopling: Convergences and divergences, Quaternary International (2014

Monegatti, P., Oll�e, A., Pugliese, A., Renault-Miskosky, J., Sozzi, M., Ungaro, S.,Vannucci, S., Verg�es, J.M., Wagner, J.J., Yokoyama, Y., 1998. L'industrie lithique deC�a Belvdere i Monte Poggiolo: stratigraphie, mati�ere premi�ere, typologie,remontage et traces d'utilisation. Anthropologie 102, 343e466.

Toro-Moyano, I., Barsky, D., Cauche, D., Celiberti, V., Gr�egoire, S., Lebegue, F.,Moncel, M.H., de Lumley, H., 2011. The archaic stone tool industry from Bar-ranco Le�on and Fuente Nueva 3, (Orce, Spain): evidence of the earliest homininpresence in southern Europe. Quaternary International 243, 80e91.

Toro-Moyano, I., de Lumley, H., Fajardo, B., Barsky, D., Cauche, D., Celiberti, V.,Gr�egoire, S., Martinez-Navarro, B., Patrocinio Espigares, M., Ros-Montoya, S.,2009. L'industrie lithique des gisements du Pl�eistoc�ene inf�erieur de BarrancoLe�on et Fuente Nueva 3 �a Orce, Grenade, Espagne. Anthropologie 113, 111e124.

te (Apricena, Fg, Southern Italy) in the context of the first European), http://dx.doi.org/10.1016/j.quaint.2014.08.051