OJS380 Peresani et al DEF corretto Potì.pub


   

 Available online http://amq.aiqua.it 
ISSN (print): 2279-7327, ISSN (online): 2279-7335 
 
Alpine and Mediterranean Quaternary, 35 (1), 2022 , 5-29 

THE MIDDLE PALAEOLITHIC SITE OF VAJO SALSONE IN THE MONTI LESSINI, ITALIAN ALPS.  
FIRST REPORT ON THE ARCHAEOFAUNA AND LITHIC  

ASSEMBLAGE WITH FOLIATE TOOLS. 
 

Marco Peresani 1,2, Davide Delpiano 1, Ursula Thun Hohenstein 1, Gianluca Arnetta 1, 
Hugues-Alexandre Blain 3,4, Vittoriana Espedito 1, Juan Manuel López-García 3,4, 

Alessandro Potì 1, Ugo Sauro 5, Davide Visentin 1,6, Giorgio Chelidonio 1, 
Davide Margaritora 1, Yvonne Tafelmaier 7 

with the collaboration of: Lorena Beltrame 1, Cristina Failla 1, Federico Pellegrini 8, Paolo Pellegrini 8 
 

1 Università di Ferrara, Dipartimento di Studi Umanistici, Sezione di Scienze Preistoriche e Antropologiche, Ferrara, Italy. 
2 IGAG - CNR, Milano, Italy. 
3 IPHES, Tarragona, Spain. 

4 Universitat Rovira i Virgili, Departament d'Història i Història de l'Art, Tarragona, Spain. 
5 Ethnografic Museum, La Lessinia, l’uomo e l’ambiente, Boscochiesanuova (VR), Italy.  

6 Archaeology of Social Dynamics Group, IMF-CSIC, Barcelona, Spain. 
7 Regierungspräsidium Stuttgart, Landesamt für Denkmalpflege, Referentin für Paläolithikum & Mesolithikum, Esslingen, Germany. 

8 Via Vagiol 44, 37020, Fane di Negrar, Verona, Italy. 

 
Corresponding author: Marco Peresani <marco.peresani@unife.it>  

 
ABSTRACT: This first report aims to illustrate the Middle Palaeolithic site of Vajo Salsone in the Monti Lessini plateau in the 
eastern Italian Alps, its geological and geomorphological context, the conditions leading to its discovery occurred after a road 
cutting, and the archaeological excavation of a karst structure where the sediments, faunal and cultural remains were still preser-
ved. The karst infill is a massive clast-supported breccia with abundant animal bones and lithic artifacts coated of carbonate 
encrustations. The first analysis of the samples of small and large mammal assemblages recovered in the karst deposits has 
revealed the abundance of Microtus arvalis, a rodent currently reported to live in open environments and in relatively drier re-
gions of northern Italy. Still, mosaic habitats with the presence of stony areas and scattered low scrubs are suggested by the 
presence of H. viridiflavus among the herpetofauna. Amongst the abundant faunal assemblage, the most common represented 
species are Cervus elaphus, Capreolus capreolus and Rupicapra rupicapra along with few carnivores. 
At the present state, only red deer bones showed taphonomic evidence ascribable to hunting and exploitation, similarly to other 
Middle Palaeolihic sites of the region. The lithic industry consists of a huge number of chert artifacts produced using Levallois as 
the main knapping method together with an ephemeral use of discoid and volumetric blade exploitation. No appreciable differen-
ce in the taphonomic layout has been observed between the by- and end-products of these technologies. Predetermined blanks, 
cores and numberless of flat, elongated, cortical flakes attest the integrity of the Levallois reduction sequences. Retouched tools 
are mostly simple, double, often converging and, occasionally, transverse scrapers, and retouched points. In addition to these 
typical Mousterian tools, the distinguishing feature of Vajo Salsone is the presence of foliate tools made through invasive bifacial 
shaping, an already known model in Central Europe. Based on what resulted from our preliminary data, we have provisionally 
reconstructed the morphogenesis of the site in accordance with the following hypothesis: 1) in the basal part of a wall, inside a 
large natural niche evolved by weathering and karst processes, groups of Neanderthals settled, leading to the deposition of an-
thropogenic sediments; 2) following phenomena of rockfall and toppling of large blocks, retreat of the wall and dismantling of the 
large niche occurred, together with the erosion of the sediments of the shelter. These sediments were transported down-valley 
and partly trapped in a small karst shaft. Even though most of the evidence from the Vajo Salsone lithic assemblage is compa-
rable with the one of other sites in the region and the northern Mediterranean area, it does however provide new details about 
the Late Pleistocene cultural scenario, thus contributing to enrich the Middle Palaeolithic picture. 
 
Keywords: Mousterian, Foliate tool, Fauna, Prealps, Italy.  
 

1. INTRODUCTION 
 

Monti Lessini, in the central Southern Italian Pre-
Alps (Fig. 1), is a mountainous area closely surveyed for 
the reconstruction of the Middle Palaeolithic settlement 
system and Neanderthal culture. Within the framework 
of multidisciplinary projects led by the Natural History 
Museum of Verona, the University of Ferrara, the Super-
intendence for Archaeological Heritage of Veneto and 
local societies, surface prospections and excavations 

have been carried on from the end of the 19th century, 
and they are still ongoing (see contributions in Aspes, 
2003). Monti Lessini count the highest number of Mous-
terian sites within the North of Italy. A considerable bulk 
of data was obtained from deposits in caves, rock shel-
ters and open air sites, becoming even more significant 
when compared to other Italian mountain ranges 
(Margaritora et al., 2020). However, despite the remark-
able information acquired through the study of these 
sites, our knowledge about the variability of Neanderthal 

https://doi.org/10.26382/AMQ.2022.01 



   

 

material culture is still unexpectedly far from being con-
sidered exhaustive. A factor of bias is the decreasing of 
the archaeological record as consequence of destruc-
tive post-depositional processes, which affect the de-
gree of preservation and therefore the information value 
of the finds coming from open-air and stratified sites. 

Stratigraphic deposits preserved in atrial cavities 
and under rock shelters, indeed act as geological ar-
chives, preserving key information about chronology, 
evolution of natural environments and expressions of 
human behavior. Among the contexts that had been 
systematically explored, it is worth to mention Riparo 
Zampieri, Riparo Mezzena, the “cave A” of Veja, Riparo 
Tagliente, Grotta della Ghiacciaia and Grotta di Fu-
mane. In particular, the latter is a crucial site for the 
understanding of the Middle and Upper Palaeolithic 
transition, bearing the oldest cultural attestations ascrib-
able to Anatomically Modern Humans (Peresani, 2012; 
Peresani et al., 2016; Broglio et al., 2006). Despite the 
dating of some stratigraphic sequences still need to be 
confirmed, according to the material culture found at 
these sites it appeared that human groups settled in the 
area roughly during the Late Pleistocene. Older lithic 
assemblages are represented by thick flakes sometimes 

retouched or by Levallois industries, both associated 
with some bifaces. Still, the Quinzano quarries remain 
the most important site, also because of the recovery of 
faunal remains of deer, elephant, fallow deer, and roe 
deer (Zorzi & Pasa, 1944-45). 

In north-eastern Italy, it is commonly assumed that 
the Levallois technology, with its recurrent unidirectional 
and centripetal modalities, dominates the lithic indus-
tries of the sites of this region and of Monti Lessini as 
well (Peresani, 2001, 2011). Flexibility in flake-making 
resulting from the application of these different recurrent 
modalities has been observed both at open-air and at 
multilayered sites. Technological similarities amongst 
the lithic industries have been recognized within the tool
-sets, mostly composed of side, transverse, converging 
scrapers, retouched points, and, to a much lesser ex-
tent, of denticulates. Conversely, none of these assem-
blages specifically aimed at a Levallois points produc-
tion. Yet, a core volumetric concept related to the Quina 
sphere was undoubtedly identified at two sites 
(Peresani, 2012; Jéquier et al., 2015). Discoid technolo-
gy was recorded as well. It becomes more frequent 
during the Marine Isotope Stage 3 and participates to a 
more complex cultural phenomena which highlights an 

Fig.1 - North-eastern Italy showing the geographical location of Vajo salsone and of the Middle Palaeolithic sites mentioned in the text: 1) 
Vajo Salsone; 2) Fumane cave; 3) Ghiacciaia cave; 4) Tagliente rockshelter; 5-6) Broion cave, Broion rockshelter; 7-8) San Bernardino 
cave, Paina cave; 9) De Nadale cave; 10) Podere due Pozzi; 11-13) Palazzone-Ca Roma, Peverella, Ca’ San Carlo; 14) Brusaida; 15) 
Ghiardo; 16) Monte Netto. Coordinate system ETRS89 / UTM zone 32N (EPSG 25832); Digital Elevation Model (base topography – 
Copernicus Land Monitoring Service (CLMS), 2019 and General Bathymetric Chart of the Oceans (GEBCO), 2019; Rivers and lakes 
(http://www.sinanet.isprambiente.it/it/sia-ispra/download-mais/reticolo-idrografico). 

 
 
6 

 
 

Peresani M. et al. 



   

 

increase in the variability at the end of the Middle Pal-
aeolithic (Peresani, 2011). Discoid industries are typi-
cally represented by a range of flakes and pseudo-
Levallois points, scrapers, points and denticulates, and 
purposedly modified backed tools (Delpiano et al., 
2019a). 

With the aim of shedding new light on the Nean-
derthal cultural variability of this area, the following pa-
per presents the recently discovered (2017) site of Vajo 
del Salsone (Vajo Salsone). Its archaeological content 
is represented by a huge amount of lithic artefacts and 
faunal bones contained in a karst pit developed in oolitic 
limestones and opened during new road cutting. The 
cultural value of this site is supported by the unex-
pected finding of several leaf points which add a touch 
of originality in the Middle Palaeolithic of this part of 
Mediterranean Europe. 

 
2. PRESENTATION OF THE VAJO SALSONE SITE 
 
2.1. The site in the morphostructural setting of the 
Lessini Mountains  

The Vajo Salsone site is located in the Monti Les-
sini, a mountain group belonging to the Venetian Pre-
Alps (Fig. 1). The Lessini group can be considered a 
monocline structure with a trapezoidal shape, where its 
shortest side faces north. It is radially dissected by sev-
eral valleys developed along tectonic lines, shaping 
canyons - locally called vaj - in some sections. The 
relief is mainly sculptured in the context of a series of 
carbonate rocks ranging in age between the upper Tri-
assic and the Eocene. This is the so-called “Serie Ve-
neta" which includes the following geological for-
mations: Calcari Grigi, Oolite di San Vigilio, Rosso Am-
monitico Veronese, Maiolica, Scaglia Rossa, Eocene 
limestones, formed in marine environment. Volcanic 
rocks are equally attested in the upper part of the suc-
cession (Bosellini et al., 1967; Carraro et al., 1969). In 
its medium-high portion, the Serie Veneta is rich in 
chert. This one can be found both in primary outcrops - 
as nodules and layered deposits in late Jurassic, Creta-
ceous and Eocene rocks - and in secondary position in 
superficial deposits, soils and coarse alluvial deposits 
(Bertola, 2016). 

If the main features of the relief can be explained 
by the lithological and tectonic characteristics of the 
different formations, the understanding of the morpho-
structural building and its evolution requires a summary 
of the geological history and an analysis of the sort of 
deposits resulting from the main recent seismic events 
(Sauro, 2003; Sauro & Zampieri, 2001; Sauro & Ferra-
rese 2016). 

The most significant tectonic events for the history 
of the morphostructural building began with a relaxing 
phase, developed during Jurassic and Cretaceous peri-
ods, when the sedimentary rock complex was still cov-
ered by the sea. The movements of continental masses 
generated fractures in the crust, with formation of a 
submarine rift consisting in a sort of plateau elongated 
in the N-S direction, corresponding to the current Vero-
na-Trentino area (called “Trento platform”). This plateau 
was bordered on both sides by a stairway of narrow and 
elongated blocks, which marked the passage towards 

the bottom of two basins located on the sides: on the 
west side the “Lombard basin”, and on the east side the 
“Belluno basin” (Castellarin, 1981). In this phase, the 
different depths of the seabed influenced the sedimen-
tation processes, and therefore determined the charac-
teristics and thickness of the formations themselves. 

The subsequent deformation events were con-
trolled by the fault systems identified in the context of 
the Mesozoic rift. The ancient faults were reactivated 
due to different geodynamic conditions, mainly of com-
pressive type. In addition, new fault systems were origi-
nated, such as the Schio-Vicenza system (Zampieri, 
1995, 2000). 

This evolution is explained by the collision of the 
African craton against the Eurasian one, which, in the 
Mediterranean, resulted in the formation of the Alps. 
The directions of maximum tectonic compression have 
changed over time, rotating counterclockwise, from N 
55° direction (between 50 and 35 million years ago), 
followed by N 25° directions (between 30 and 16 million 
years), N 340° (between 16 and 6 million years), and N 
310° (between 6 and 2 million years) (Castellarin et al., 
2006). 

Nowadays, compressive mechanisms are still 
active, as shown by recent earthquakes such as the 
Friulian one in 1976, or the historical 1117 Verona 
earthquake, which turns out to be the most famous his-
torical seismic event in the Alps (Guidoboni et al., 1985, 
2005; Galadini et al., 2001; Sauro & Zampieri 2001; 
Sauro, 2003; Poli et al., 2008; Viganò et al., 2008; Sau-
ro & Ferrarese, 2016; Mantovani et al., 2021). The pre-
vailing compression from south to north is due to the 
advancement of the Apennine front, buried under the 
alluvial sediments of the Po River. Yet, at the same 
time, the Brescia Pre-Alps, being hindered in their 
movement to the north by the Adamello batholith, divert 
their thrust to the east, investing the Monte Baldo group 
and the Lessini mountains. Therefore, the compressive 
forces are exerted both in N-S and in E-W direction due 
to the anisotropy of the structural building. The result is 
a progressive fragmentation of the N-S elongated 
blocks and their deformation, together with vertical shift-
ing and western tilting (Massironi et al., 2009; Sauro & 
Ferrarese, 2016).  

Tectonic activity and its related fault systems 
(Zampieri, 2000) exert strong influence on erosional 
phenomena, represented by weathering, karst, and 
different classes of slope processes which led to the 
development of the present-day landscape. Karst is one 
of the most impacting process and together with hydro-
logical activity gives rise to different styles of physical 
landscape also controlled by the lithological features of 
the geological formations. These features can be identi-
fied through the hydrographic network and from a series 
of different superficial and underground morphologies, 
resulted by the numerous faults and fractures that ac-
celerate and facilitate the water flowing through rocks. 
According to the tectonic evolution and to the main 
mopho-dynamic processes, the pattern of the present-
day landscape may be defined as both fluviokarstic and 
tectokarstic (Sauro, 1973; Mietto and Sauro, 2000). 

In the Valpantena area these movements can be 
easily detected: surface faulting walls with displace-

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 

7 



   

 

ments that locally exceed two meters, folds created at 
the edges of the blocks, covers of debris and breccias 
on the slope that can be interpreted as seismites 
(Zanferrari et al., 1982; Sauro & Zampieri, 2001; Sauro 
& Ferrarese, 2016). The Vajo Salsone is located right 
besides an E-W fault that separates two of the blocks 
constituting the N-S ridge between Valpolicella and 
Valpantena. This explains the intense fracturing activity 
in proximity of the fault. 

During the Quaternary, the Monti Lessini were 
partly covered by the Adige glacier or by local glaciers 
attested on the upper part of the plateau and in the 
nearby Carega Massif (Sauro, 1973, 2020). Periglacial 
conditions activated strong erosion of the paleosols, 
accumulations of colluvial deposits in valley bottoms 
and dolines, especially in the central part of the plateau. 
From the Middle Pleistocene onward, the Loess sedi-
mentation affected the entire surface of the plateau 
(Magaldi & Sauro, 1982; Castiglioni et al., 1990). De-
posits in caves and rock shelters along the slopes al-
lowed to estimate that the deepening of gorges and 
valleys occurred at least from the late Middle Pleisto-
cene (Castiglioni et al., 1990; Cremaschi, 1990). An-
thropogenic soil erosion during the Holocene strongly 
affected the main slopes, especially starting from the 
Neolithic. 

 
2.2. The Valpantena and the Vajo Salsone 

The prehistoric site (5046787.37 N, 655509.97 E, 
376m a.s.l) is located on the left (northern) slope of Vajo 
Salsone, a deep cut in the right slope of the upper 
Valpantena (Fig. 2). This latter, a large erosional feature 
located NE of the town of Verona, looks like an inlet of 
the high venetian plain within the Pre-Alpine relief, in-
cluding a large alluvial plain that mostly originated dur-
ing the Middle and Late Pleistocene from the deposi-
tional activity of the Adige river and its Monti Lessini 
tributaries (De Zanche et al., 1977). Valpantena begins 
at the confluence of three canyon like valleys: Vajo della 
Marciora, Vajo dei Falconi and Vajo dell’Anguilla. Re-
garding the tectonic system, the afore-mentioned three 
fluviokarstic canyons together with Valpantena as their 
continuation, are set along a bundle of fault lines orient-
ed NNW-SSE and NNE-SSW. The right (western) 
slopes of these valleys are mainly in-facing slopes, 
while the left (eastern) ones are predominantly dip-
slopes (Sauro, 1978, 2013). This explains the asym-
metry of the valley slopes and the presence of high 
walls only on the right slopes. 

The profile of the main valley western slope sur-
rounding the archaeological site is characterized, in its 
lower part, by a strip inclined 30-35°, modelled in the 
limestone of the Lower Jurassic (formations of Calcari 
Grigi and Oolite di San Vigilio). The band is delimited 
upstream by a frame of Oolite di San Vigilio and Rosso 
Ammonitico Veronese limestone rock walls. The Rosso 
Ammonitico, consisting of micritic limestones more re-
sistant to degradation and erosion than the underlying 
rocks, acts as a “roof” for the more erodible Oolite lime-
stones, thus favouring the conservation of the walls. 
Above these, the profile develops in a large substructur-
al terrace followed upwards by a sequence of slopes 
shaped in Cretaceous limestone formations interbedded 

with volcanites, up to a sub-structural summit plateau in 
Eocene limestone, which characterizes the ridge be-
tween Valpantena and Valpolicella (Monte Comun).  

Karst landforms are widely distributed in the whole 
area either as surface karst (dolines, limestone pave-
ments, natural bridges) and deep karst (caves, shafts, 
swallowing cavities). South of the confluence of the 
three valleys, the valley floor, formed by gravelly, sandy 
and pelitic alluvial and colluvial-detrital deposits at the 
foot of the steepest slopes, widens from North to South, 
joining the high Venetian plain after 15 km. The thick-
ness of the sedimentary filling increases towards South, 
to such an extent that the rocky bottom of the middle 
and lower Valpantena is situated below the sea level. 
This can be explained by the Messinian event of 
“desiccation” of the Mediterranean and, therefore, by 
the over-excavation of the river due to the lowering of 
the base level (Bini et al., 1978). 

Fig. 2 - Geomorphologic map of Valpantena. Key: 1 - Vajo 
Salsone site; 2 - alluvional, fluvioglacial, lacustric and palustric 
deposits; 3 - Oolite di San Vigilio, Calcari Grigi, Encrinite of 
Fanes; 4 - Rosso Ammonico, Fonzaso formation; 5 - Maiolica, 
Soccher limestone; 6 - Scaglia Rossa, Scaglia Variegata, 
Scaglia Cinerea; 7 - Basaltic vulcanite of Euganei; 8 - Calvene 
formation, Salcedo Formation, Castelgomerto limestone; 9 - 
streams; 10 - Scarp � 10 m; 11 - Scarp n.d.; 12 - Fault line; 13 - 
Flexure scarp; 14 - Fault line valley; 15 - Small valleys along 
fault line; 16 - Caves. Technical note. Coordinate system 
WGS84. Digital Elevation Model (base topography DTM 5m - 
https://idt2.regione.veneto.it/). Geological layers from https://
idt2.regione.veneto.it/. Scarp from https://idt2.regione.veneto.it/. Faults 
from Sauro, 1978.  

 
 
8 

 
 

Peresani M. et al. 



   

 

2.3. Discovery of the site, rescue excavation and 
processing of materials  

The site was found during the construction of a 
truckable road between Stallavena (zona Revolto) and 
Coda (Fig. 1). On May 1st 2016, two of us (F.P. and 
P.P.) were walking on the fresh road cut at mid eleva-
tion along the eastern slope of Comun Mount, collecting 
bones and flint artifacts from the mined brecciated bed-
rock between the fourth and the fifth harpin turn of the 
road, at 376 m a.s.l. (Fig. 3a and 3b). Thanks to this 
discovery a rescue excavation campaign (Excavation 
code: GRZ17RFL45) was conducted by two of us (D.V. 
and A.P.) from May 23rd to June 3rd 2017, led by the 
Archaeological Superintendence and the University of 
Ferrara. 

At first the excavations for the construction of the 
truckable road revealed covers of diamicton with a 

thickness of between a few decimeters and about 2 m, 
poor in matrix, formed by angular fragments of lime-
stone, mostly small in size but including some big 
blocks. There was little or no cementation. Inside the 
covers there were also soil sediments, both in the form 
of bodies trapped inside cavities with irregular shapes 
and mixed with rock clasts. The artificial sections of the 
deposit allowed to recognize some cavities inside it, in 
form of small conduits, probably caused by piping pro-
cesses. 

The site is a small and narrow karst cavity filled by 
Pleistocene sediments yielding abundant Middle Paleo-
lithic artefacts and faunal remains. The finds were gath-
ered in a context located two meters below the original 
ground-level, the Vajo slope. During the excavation the 
reworked upper stratigraphic units (SUs 1-2-3) were 
removed as a whole, whereas the main layer (SU 5) 

Fig.3 - Slope and road between Stallavena and Coda where the archaeological deposit was found (3a); the study area before any ar-
chaeological interventions (3b); detail of the terraced S-E side of the cavity (3c); eastern view of SU 5 once recent layers were removed 
(3d); south-eastern view of US 5. Red nails delimitate the square-grid used for the excavation of US 5 (3e); South-western view of a 
portion of SU 5 preserved beside the karst chimney (3f).  

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 

9 



   

 

Fig. 4 - General plan of the excavated area with extention of SU 5 (drawing by D.V. and A.P.). 

with its palaeontological and archaeological content was 
manually excavated through a 50 cm grid system and 
artificial spits of different depths (Fig. 3c, 3d and 3e; Fig. 
4). Due to mines activity, the rock wall was highly unsta-
ble. For this reason, it was not possible to entirely inves-
tigate the archaeological deposit since part of the cavity 
extended beside the road cut (Fig. 3f). In addition to 
loose sediment samples, three columns of soil and 
abundant speleothems samples were collected for mi-
cromorphological analyses. Sediments were wet sieved 
through superimposed meshes ranging from 5 to 0.5 
mm. Lithic artifacts and faunal remains were systemati-
cally recovered in situ during the excavation process 
and after wet sieving the sediments. 
 
2.4. Sedimentary context  

Morphology of the karst structure. The site con-
sists in a narrow vertical shaft -a feature locally called 

“arso”- formed by the chemical dissolution and water 
erosion of a discontinuity in the geological bedrock, 
where two faults (respectively oriented WNW-ESE and 
NE-SW) meet (Fig. 5a). There were no traces of active 
water erosion. The eastern and southern sides of the 
cavity are terraced and consistent with the stratification 
of the bedrock, forming a sort of underground room. 
Due to the explosions and the removal of the reworked 
material the top of the cavity was taken off, thus show-
ing the central and lowest parts of the underground 
room. Despite this partial loss, it was possible to infer 
the presence of a second karst system. This latter is 
sub-horizontal, developing N-S, and it might have con-
tributed to the formation of the aforementioned under-
ground room. On the southern side of the excavation 
the filling deposits directly expose on the slope. The 
remaining of the vertical chimney was identified at the 
NW of the deposit on the side of the road cut (Fig. 5b). 

 
 
10 

 
 

Peresani M. et al. 



   

 

Nevertheless, this was not enough to allow to infer the 
original size and shape of the cavity. Once the archaeo-
logical layers were removed, it was clear that these last 
ones occupied only a small portion of the chimney, 
which deepens far below the excavated area without 
any filling. 

Sediments, distribution, and composition. Below 
the reworked layers (RIM) resulted by the construction 
works it lays the archaeological deposit. This filling is 
bounded by the cliff and the slope on its northern and 
southern sides, respectively. Overall, the volume of the 
deposit spread over a 4x4 m surface, and it was thick 
more than 2 m (without considering the chimney). Dur-
ing the archaeological excavation six Stratigraphic Units 
(SU) were identified, including a negative one: 

SU1: reworked layer located in the northern part of 
the excavation besides the rock wall. It contained faunal 
remains, lithics and modern glazed pottery. 

SU2: resulted by the digging of an artificial pit 
(SU6) in the southern part of the excavation. The filling 
was located E or NE of the pit and it was characterized 
by abundant lithic artefacts, faunal remains and lime-

stone clasts ranging in size from cm to dm. The artificial 
pit was realized in the framework of the first surveys 
carried on once the deposit was discovered but before 
any archaeological intervention. It deepens until SU5.  

SU3: present day soil preserved only in the SE 
area of the excavation and laying on slope deposits, 
lays on SU4 and partially on SU5. It was cut by con-
struction works - which did not affect B and C horizons - 
and its thickness measures about 15-40 cm. It is a 
loamy sand soil characterized by abundant and hetero-
metric clasts (<30 cm) randomly oriented. The layer is 
dark brown, homogeneous, friable, with roots, organic 
matter, lithic artifacts and faunal remains scattered in 
proximity of the boundary with SU 5. 

SU5: loamy sand and clast supported layer, slight-
ly hard, yellowish brown (10YR 6/6 Munsell Soil Color 
Charts®, 2001) with increasing content of clay in the 
lowest part. It contains monolithologic blocks of the size 
of decimeters, rare, smoothed blocks, fragmented spe-
leothems, abundant lithic artefacts, and faunal remains 
showing fresh and not abraded edges. Two are the 
most frequent carbonate deposits: A) fragments of spe-
leothems dispersed in the sediment grading > 30-40 cm; 
B) withish, polylobed and millimetric spots/stains. The 
layer covers a 4x4 m surface and its thickness spans 
from few centimetres to more than two meters in rela-
tion to the topography of the karst structure (Fig. 6).  

SU4: Carbonate bedrock exposed over the whole 
excavated area. 

Flowstone and fragments. In the deepest part of 
the karst chimney which contained the archaeological 
layers it was identified a calcite deposit, 3-4 cm thick. It 
develops besides the bedrock and it preceded the 
blockage of the cavity. 

 
3. THE FAUNAL BONE ASSEMBLAGE 

 
3.1. Small vertebrates  
3.1.1. Materials and methods  

Small vertebrates were identified following the 
general criteria given by Berto (2013) and López-García 
(2011) for insectivores and rodents, and Szyndlar 
(1984, 1991) for snakes. The specific attribution of this 
material is mainly based on the best diagnostic ele-
ments, mandible, maxilla and isolate teeth for small 
mammals, and trunk vertebrae for snakes. Fossils from 
SU5 were grouped using the minimum number of indi-
viduals (MNI) method, by means of which we deter-
mined the sample (i.e. from each level) through the 
counting of the most represented diagnostic elements. 
In the case of snakes, MNI has been approached taking 
into account size categories. 

 
3.1.2. Results 

Although the number of identified specimens 
(NISP) and the MNI of small vertebrates are low in Vajo 
Salsone SU5, the number of the species represented is 
relatively high (Tab. 1). As a matter of fact, small mam-
mals are represented at least by nine different taxa, one 
insectivore (Sorex sp.) and eight rodent species 
(Arvicola amphibius, Dinaromys bogdanovi, Clerthriono-
mys glareolus, Microtus arvalis, M. agrestis, M. 

Fig. 5 - The karst fissure exposed at the end of the excavation 
(5a) with portion of SU 5 deepening below the rockwall left 
unexcavated because of the instability of the rock. As shown by 
the picture 5b, the deposit extends inside the rockwall on the 
northern side of the excavation. The sediment within the karst 
chimney only partially fills the latter.  

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 
11 



   

 

(Terricola) multiplex-subterraneus, Apodemus 
(Sylvaemus) sp. and Glis glis, whereas the herpetofau-
na is only represented by snakes. Green whip snake 
(Hierophis viridiflavus) is probably represented by two 
individuals, an adult (one cervical vertebra, and 20 trunk 
vertebrae) and a juvenile (two cervical vertebrae, and 
nine trunk vertebrae), while a viper (Vipera aspis) is 
documented by two fragmentary trunk vertebrae. The 
relative abundance of M. arvalis, being the most abun-
dant species and the presence of the Balkan vole (D. 
bogdanovi) indicate a Late Pleistocene age for this as-
semblage. On the contrary, the Holocene in this area 
(according to Berto et al., 2019) from the Bølling-Allerød 
interstadial is characterized by a strong decrease of M. 
arvalis and the disappearance of D. bogdanovi. 

From a quantitative point of view, among the small 
mammals the common vole (M. arvalis) is the most 
abundant species, representing 39.3% of the total MNI. 
Such a high rate of M. arvalis is relatively common in 
late Pleistocene sites of north-eastern Italy, as attested 
during MIS 5 of Grotta Maggiore di San Bernardino 
(López-García et al., 2017), MIS 3 of Grotta del Broion 

(Colamussi, 2002), and Grotta Minore di San Bernardi-
no (López-García et al., 2019) in the Berici Hills, togeth-
er with MIS 3 of Fumane cave (López-García et al., 
2015) and MIS 2 of Riparo Tagliente (Berto et al., 2018) 
in Lessini Mounts. The abundance of M. arvalis is cur-
rently reported to be higher in open areas and in rela-
tively drier regions of northern Italy (Amori et al., 2008). 
Still H. viridiflavus prefers open, mosaic habitats with 
the presence of stony areas and scattered low scrubs. 
These conditions allowed both species to find open 
areas for thermoregulation, optimal shelters and hunting 
zones. On the other hand, they tend to avoid uniform 
environments such as extensive forest or large grass-
lands (Scali et al., 2008). Nowadays, in the region of 
Vajo Salsone, only two vipers are present: Vipera aspis 
and Vipera berus (Sindaco et al., 2006; Sillero et al., 
2014). V. aspis occupies a wide variety of environ-
ments, however, its thermal requirements oblige it to 
exploit habitats provided with stony substrate, relatively 
sunny and south-facing dry soils, and with some vege-
tation cover. In a similar way, V. berus requires a cer-
tain habitat complexity. In southern France and northern 

Fig. 6 - Stratigraphic sections of SU 5 (see figure 4 for position; drawing by D.V. and A.P.).  

 
 
12 

 
 

Peresani M. et al. 



   

 

Italy, it is found in either low lying wetlands (if dry soils 
are available) or at high altitudes (Mallow et al., 2003). 

Among the aforementioned sites - which are geo-
graphically close to each other, and cautiously taking in 
account data from Vajo Salsone SU 5 due to the small 
sample of remains, it resulted that this latter mammal 
assemblage is similar to Fumane cave’s one (Lopez-
García et al., 2015), spanning between 50-35 ka BP 
(Peresani et al., 2008; Higham et al., 2009). 

 
3.2. Large vertebrates 
3.2.1. Materials and methods  

Our analysis concerned a sample of bone assem-
blage coming from the upper part of SU 5 (spits I, II and 
III). Overall, 26,360 faunal remains were examined for 
this study. The analyzed sample consists of unidentified 
bone elements and shows a high degree of fragmenta-
tion on 85.9% of the finds (22,631) smaller than 2 cm, 
while only 3,729 remains are longer than 2 cm. Consid-
ering a total sample of 3.729 TNR selected both for 
archaeozoological and taphonomic analyses, only 
89.8% was scrutinized because 10.2% of the remains 
(380) is covered by concretions (Tab. 2). Taxonomic 
identifications were carried on with the help of reference 
collections of the Laboratory of Archaeozoology and 
Taphonomy (LAT) and of the Large Vertebrates of the 
Department of Humanities, University of Ferrara. Anato-
my atlases (Schmidt, 1972; Barone, 1980; Pales & 
Lambert, 1971; Pales & Garcia, 1981) were employed 
for skeletal and anatomical nomenclature. Unidentified 
remains have been classified according to bone mor-
phology (long, flat, and articular/compact bones), as 
well as animal groups by weight. 

We quantified the identified elements using the 
following indices: NR (Number of Remains); NISP 
(Grayson, 1984); MNE (minimum number of elements) - 
it has been counted considering side, body size and 
ontogeny (Klein & Cruz-Uribe, 1984); MNI (Klein & Cruz
-Uribe, 1984; Grayson, 1984; Lyman, 1994) and MAU 
(minimum animal units) (Binford, 1984). Fragmentation 

indexes (MNE/NISP) and skeletal survival rate, ob-
tained by comparing the minimum number of elements 
(MNE) with the number of elements expected (eNE) on 
the basis of the MNI, were calculated to estimate the 
skeletal representation of the most represented taxa 
(Binford, 1981; Brain, 1981; Lyman, 1994). We as-
sessed age at death mainly through the state of ossifi-
cation of the long bones and the study of deciduos and 
permanent isolated teeth, although in some instances 
teeth eruption degree, dental wear, and the stage of 
epiphysis fusion on long bones were used as well 
(Habermehl, 1961; Mariezkurrena, 1983). 

In order to distinguish the edaphic and anthropic 
modifications, we scrutinized all the selected specimens 
at both macroscopic and microscopic level. When nec-
essary, microscopic analyses of bone surfaces were 
carried out using a Leica S6D Greenough stereomicro-
scope with 0.75-70X magnification range, also em-
ployed for capturing images. We distinguished tapho-
nomic modifications according to the criteria established 
by Behrensmeyer (1978), Binford (1981), Lyman 
(1994), Fernández-Jalvo & Andrews (2016). We classi-
fied carnivore marks as follows: pits, scores, punctures, 
furrowing, gnawing and corrosion by gastric acid 
(Fisher, 1995; Domínguez-Rodrigo & Piqueras, 2003). 
Burnt modifications were recorded too (Bennett, 1999). 
Trampling marks were distinguished from butchering 

Tab. 1 - Representation of the Number of Identified Specimens 
(NISP), minimum number of individuals (MNI) and percentage 
of MNI (%) for small vertebrates of Vajo Salsone US 5 archae-
ofauna assemblage. 

Tab. 2 - Composition of a sample of Vajo Salsone US 5 archae-
ofauna assemblage with number of identified specimens 
(NISP), minimum number of individuals (MNI) and minimum 
animal units (MAU). Minimal Number of Elements (MNE) is 
referred to the most represented element useful to estimate 
both the MNI and MAU, except for red deer for which MNI has 
been estimated on teeth and MAU on astragalus (see Table 3). 

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 
13 



   

 

marks using the works of Blasco et al. (2008) and 
Domínguez-Rodrigo et al. (2009). Evidence of anthropic 
modification observed on faunal remains include butch-
ery marks (Binford, 1981; Potts & Shipman, 1981; Ship-
man, 1981; Galán et al., 2009) and bone retouchers 
(Mallye et al., 2012). 

 
3.2.2. Results: composition and taphonomy  

Among the 3,729 selected remains longer than 2 

cm, a taxonomic determination was only possible for 
1,097 remains representing 29.4% of the sample (Tab. 
2). Isolated teeth and phalanges predominate among 
the identified elements, followed by tarsal bones, meta-
tarsals and metapodials. However, other skeletal ele-
ments belonging to both the axial and appendicular skel-
etons are present although very fragmented. 

The faunal composition (Tab. 2) consists almost 
entirely of mammals and for a small percentage of birds, 

Tab. 3 - Skeletal elements of Cervus elaphus with estimation of minimum animal unit (MAU), fragmentation index (MNE/NISP) and skele-
tal survival rate (MNE/eNE-expected number of elements) of a sample of Vajo Salsone US 5 archaeofauna assemblage. The highlighted 
values are referred to the elements utilised for estimating MNI and MNE-MAU of red deer for Table 2. (Grey and light grey respectively).  

 
 
14 

 
 

Peresani M. et al. 



   

 

at the moment not yet identified at the species level. On 
the basis of NISP and MNI, the most represented spe-
cies among ungulates is red deer (Cervus elaphus) with 
62.0%. Roe deer (Capreolus capreolus) and chamois 
(Rupicapra rupicapra) are rare: 0.5% and 0.4% respec-
tively. The presence of elk (Alces alces) is attested by 
only two remains (0.2%). Among carnivores, wolf (Canis 
lupus) prevails with 6.9%, while red fox (Vulpes vulpes) 
reaches only 1.3%. Several remains were classified as 
Caprinae, Cervidae, large Cervidae, Canidae in cases 
when it was not possible to identify them at species or 
genus level because of the lack of diagnostic morpho-
logical elements and the high degree of fragmentation. 

Considering the age at death, estimated on the 
basis of the degree of dental eruption and wear, and on 
the state of ossification of long bones, most taxa belong 
to adults, while some young and sub-adults individuals 
have been recognized only for red deer. 

The estimation of minimum animal unit (MAU) vs 
the MNI suggests that more than a half of red deer car-
casses were introduced complete into the site (Tab. 2) 
and at least half carcasses for the other ungulates. 
However, only for Cervus elaphus it is possible to pro-
pose some preliminary considerations about their ex-
ploitation and fragmentation of anatomical elements. 
Red deer is represented by 18 individuals: distributed 
over 4 juvenile/ sub-adults and 14 adults, based on the 
presence of lower deciduous (dp4) and permanent iso-
lated lower teeth (M3). With regard to the distribution of 
anatomical elements there are both parts of the axial 
skeleton and skull with high percentage of teeth 
(30.6%), and parts of the appendicular skeleton with 
prevalence of phalanges (14.7%) and tarsals (12%), 
followed by carpals (7.8%), tibia (5.3%) and metatarsals 
(4.6%) (Tab. 3). Estimation of the skeletal survival index 
(MNE/eNE) shows that the more represented anatomi-
cal elements are astragals (0.78). The other elements 
belonging to appendicular skeleton are instead poorly 
represented, showing greater fragmentation (MNE/
NISP) of hind limb than forelimb bones. This could indi-
cate a possible selection in the hunting area of ele-
ments introduced into the site. The marked representa-
tion of teeth, upper (0.58) and lower (0.53), could sug-
gest a transport to the site not only of appendicular 

parts, but also of complete carcasses in agreement with 
the results of the comparison between MAU and MNI 
(Tab. 3). 

 
3.2.3. Results: taphonomy  

All bones and teeth have the same preservation 
characteristics and uniform surface features pointing for 
a common sedimentary context and exclude any type of 
evidence ascribable to mixing with material of Holocene 
or very recent origin. The faunal remains are in a good 
state of preservation, their color is very homogenous 
and ranges from whitish to light yellow (10YR8/6 Mun-
sell Soil Color Charts®, 2001), with presence of scat-
tered manganese oxides. 27% of the remains (NR 329) 
is completely covered by concretions that are present 
on the lithic assemblage too and did not allow to ob-
serve the bone surfaces. Weathering is light although 
the remains present post-depositional fractures. 

Of the whole ensemble of taphonomic traces, those 
ascribable to carnivore actions are rare, identified on 
only 0.08% of the sample and classified primarily as 
scores and scarce presence of digested bones. Rodent 
marks were found on a small percentage as well 
(0.29%), mainly on lateral phalanges and tarsals. 

Traces of anthropic modifications have been ob-
served in higher number than those of carnivores and 
rodents (Tab. 4). The remains bearing butchering marks 
are 1.7% of the sample and were recognized on bones 
of medium and large sized cervids: 17 remains belong 
to red deer, and one to elk. Considering red deer’s ana-
tomical elements, traces of butchery are mostly present 
on segments of the appendicular skeleton, exploited not 
only for the removal of the meat but also for the con-
sumption of marrow. Still, the attested cut-marks sug-
gest skinning and disarticulation (Fig. 7) operations, 
identified respectively on the tarsals and at the level of 
the articular heads of the long bones, and those relative 
to defleshing activites localized on the diaphyseal sur-
face of the metapodials. Impacts on long bones frag-
ments aimed to collect marrow were also recognized. 
Some thick fragments of long bones like the metapodi-
als bear notches testifying the use of bones as retouch-
er of stone artifacts. Burning traces were found primarily 
on remains smaller than 2 cm, while only 1.05% of the 
selected sample showed color changes due to expo-
sure to heating; these included a tooth fragment, a phal-
anx of roe deer, two diaphyseal fragments of metapodi-
al and a tarsus of red deer. 

 
4. LITHIC INDUSTRY 
 
4.1. Materials and methods  

The lithic industry consists of a huge number of 
chert artifacts picked up as single items during the ex-
cavation and after screening of wet sieved sediment. All 
artifacts are affected by extensive and deep whitish 
opaque patinae regardless the texture of the chert. Nev-
ertheless, the sandy textured chert artifacts of the Ju-
rassic San Vigilio Group were so intensively weathered 
that ended up disaggregating. However, we did not 
observe the presence of either the yellowish to dark 
reddish or bright alteration and patinae which usually 
affect the paleolithic artifacts contained in clayey pal-

Tab. 4 - Percentage distribution of the different typologies of 
bone surface modifications in a sample of Vajo Salsone US 5 
archaeofauna assemblage. 

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 
15 



   

 

aeosoils surveyed in this region and the whitish surface 
patinae frequently developed on the Neolithic - Bronze 
Age lithic artifacts (Magaldi et al., 1981; Chelidonio & 
Stocchiero, 1991; Chelidonio & Sommaruga, 2011). All 
artifacts are variably coated of carbonatic concretions.  

Wet sieved sediments were carefully scrutinized in 
order to exclude any recent contamination of the Vajo 
Salsone lithic assemblage. We examined each artefact 
of module (length + width) greater than 4 cm in order to 
determine the main technological and typological fea-
tures for an approximate total number of >10,000 items. 
The technological outlines of the major lithic production, 
together with the main typological and typometric fea-
tures of retouched blanks and leaf points are described 
below in the results section. The conceptual and analyti-
cal approach to this preliminary inspection is inspired by 
several authors according to the main volumetric con-
cepts in use during the Middle Palaeolithic (Boëda, 
1994; Delagnes, 1992). Moreover, it has been enriched 
with broader criteria for the definition of Levallois prede-
termined products (Grimaldi, 1995; Guette, 2002) and 
with insights on the variability of the Discoid core tech-
nology (Peresani ed., 2003). Volumetric blade techno-
logical features are consistent to what was assessed by 
Inizan et al. (1999). We referred typological taxonomy to 
F. Bordes, 1961. 

To provide a first reconstruction of the reduction 
sequences, we conducted morpho-technical and mor-

phometric observations on cores, the complete Leval-
lois and other predetermined blanks, as well as certain 
by-products deemed to have had a significant role in the 
production, including the phase of the reduction se-
quence from which they were obtained. We did not con-
duct any systematic refitting of artifacts. Technological 
outlines of the major lithic productions, together with the 
key typological features of the retouched blanks, are 
shown in the paragraphs below. 

 
4.2. Results: main technological outline 

The lithic industry does not contain any diagnostic 
artefact belonging to any of the Upper Palaleolithic 
blade-bladelet techno-typologically featured industries 
known in the area, like the Uluzzian with the backed 
knives and splintered pieces (Peresani et al., 2016; 
2019), the Aurignacian and Proto-Auignacian with the 
end-scrapers, retouched blades and bladelets, and 
bone industry (Broglio, 1993; Aleo et al., 2021), the 
Early Epigravettian with shouldered points (Peresani et 
al., 2021) and the Late Epigravettian with the backed 
points, short endscrapers and microliths (Broglio, 1992; 
Montoya et al., 2018). The same holds for the Mesolith-
ic to Bronze age technologically and typologically fea-
tured industries (Broglio, 1992).  

The assemblage is mostly based on the use of the 
Levallois method, as attested by predetermined blanks, 
abundant flat, elongated, cortical flakes and 50 Levallois 

Fig. 7 - Metacarpal of Alces alces with linear butchering marks produced during the recovery of the meat mass (A) and scrapings, refera-
ble to the periosteum removing phase, associated to cupules produced by flint retouching (B). Astragalus of Cervus elaphus with a series 
of short and repeated cut-marks probably due to skinning (C). Fragment of coxal bone ascribed to Cervus elaphus showing a series of 
insistent cutmarks (D) and chopmarks (E) in proximity of the acetabulum associated to the disarticulation of the femur. 

 
 
16 

 
 

Peresani M. et al. 



   

 

cores (Tab.5). The set of finds ascribable to Levallois 
core maintenance includes a wide range of items sug-
gesting the use of various knapping strategies, such as 
the trimming of striking platforms, predetermination, the 
repair of flaking faces and the ablation of flaking acci-
dents. All chert types were exploited: Scaglia Rossa, 
Maiolica and San Vigilio Group. 

The Levallois production consists of different re-
duction sequences based on recurrent unidirectional 
and centripete exploitation (Fig. 8, nn. 1 and 2). These 
ones, at the same time, involve the application of tech-
nical variants related to the unidirectional modality. The 
initialization mostly concerned cortical nodules rather 
than plates and blocks, that were already naturally suit-
able for the predetermination of lateral and distal con-
vexities. The blank was then processed, once that strik-
ing platforms were prepared while lateral convexities 
were shaped by means of detaching blades along the 
longest edges of the solid. The same procedure was 
used for distal convexities, which were either naturally 
shaped or improved with oblique or orthogonal remov-
als to the core axis. 

The evidence related to unidirectional flakes re-
flects the intent of optimizing the exploitation of the vol-
umes consistently with the technical criteria adopted 
during the main production phase, right up to deactiva-
tion. Thanks to this optimisation strategy, the decortica-
tion and initial shaping of the core convexities were 
avoided. The striking platforms were faceted in an al-
most systematic manner, whereas the lateral convexi-
ties were created and maintained by means of ordinary 

and Levallois core-edge removal flakes and the distal 
convexity was modelled with the removal of one or a 
series of moderately invasive flakes or from the ridges 
left by previous Levallois removals. Cores were exploit-
ed also by multidirectional Levallois series from the very 
first stages of the reduction, and sometimes on both the 
faces of the core. Centripetal modality had different 
roles in flake-making: it became active since both the 
first steps of core reduction and in the middle and final 
steps of the unidirectional sequence. This observation 
requires further confirmation, although several reasons 
might be already envisaged: hinged sensu lato unidirec-
tional removals leading to the activation of new flaking 
axes for repair purposes; core reduction, with the con-
sequent contraction in size to the detriment of the mor-
pho-technical features of the predetermined blanks; 
routinary turning from unidirectional to orthogonal orien-
tation of flake detachments with the core exploitation 
approaching the centripetal pattern with minimal modifi-
cations of the extenction of the striking platform. Thus, 
flakes acquired variable edges showing either polygonal 
or fan-shaped outlines.  

Except hinged or plunged flakes, or when the fur-
ther re-shaping of the convexities is considered eco-
nomically deficitary, the production ends with the reduc-
tion of the core volume. Only few discards attributable 
to incipient fissurations, voids or other phenomena pin-
pointing to a poor selection of lithic raw materials have 
been observed. 

Several flakes and some cores attest the exploita-
tion of the lower faces of flake-cores obtained from 

Fig. 8 - Selection of cores and artefacts representative of Vajo Salsone lithic assemblage: bifacial Levallois cores exploited in the recur-
rent centripetal variant (1); unidirectional Levallois core (2); centripetal core (3); overshot blade (4); Mousterian point on a Levallois blade 
(5); Levallois centripetal flake (6); simple scraper on a laminar flake showing a burin-like detachment on the left edge (7).  

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 
17 



   

 

large cortical flakes. In these instances, knapping oper-
ations varied and they mostly involved the proximal 
zone, where the bulb could be directly removed once 
convexities were prepared. Flakes are thin and invasive 
and the edges relatively long and regular, sometimes 
with the partial removal of the core-edge. 

Discoid production is attested by very few cen-
tripete cores exploited on both surfaces for obtaining 
pseudo-Levallois points, core-edge removal flakes and 
some centripete flakes. All these blanks were not re-
touched. Cores display the typical biconvex outline with 
very limited remnants of cortical surface on one of the 
two faces and do not show any valuable evidence for 
reconstructing the origin of their reduction sequence. 

Fig. 9 - Selection of leaf-shaped tools from Vajo Salsone: fragmented leaf points (1-2); fragmented and partially refitted leaf-shaped tool/
preform (3); bifacially leaf-shaped scraper (4); asymmetrical leaf knife (5); fragmented leaf point/knife (6).  

Tab. 5 - Composition of Vajo Salsone retouched lithic assem-
blage according to the main tool types. 

 
 
18 

 
 

Peresani M. et al. 



   

 

Volumetric blade technology is mainly attested by 
laminar products and maintainance pieces, while blade 
cores are less abundant. The production of elongated 
artefacts is thus based both on Levallois and laminar 
technologies: the latter can be recognized by thick 
blades characterized by curved profile, unidirectional or 
convergent scars and unprepared butt. Blade produc-
tion began when a natural edge was exploited on a 
major axis of the raw block. Blade cores management 
was carried out through the detachments of lateral, 
partially-crested blades and overshot blades (Fig. 8, n. 
4). Together with thick and wide blades, a production of 
thin bladelets is also attested, although not yet framed 
in their technological trajectories. Bladelets could be 
obtained both from unipolar cores and from the burin-
like reduction of flake blanks (Fig. 8, n. 7).  

 
4.3. Results: retouched tools and bifacial tools 

Retouched tools are mostly scrapers, followed by 
few points, notches and denticulates (Fig. 8, nn. 5 and 
6). Scrapers are simple, double, often converging and 
occasionally transverse (Tab. 5). Marginally retouched 
flakes are rare. Cortical and partially semicortical flakes 
were employed as well. We did not observe any evi-
dence related to the utilization of flakes affected by acci-
dents, fragmented flakes, recycled tools and cores. 
Regarding scrapers, these tools were mostly shaped on 
unidirectional Levallois flakes rather than centripete 
flakes, discoid flakes and blades. On the other hand, 
flakes and denticulates tended to be manufactured on 
cortical and flakes affected by imperfections, whereas 
points were made on Levallois flakes. Several scrapers, 
cortical flakes and Levallois flakes were also thinned on 
the ventral face. 

Artefacts yielding flat and invasive bifacial shaping 
represent one of the most distinctive features of Vajo 
Salsone lithic assemblage (Fig. 9). At least 23 tools 

bear this type of retouch, characterized by a certain 
diversity in typology and raw material, chosen from 
those available locally: high quality, fine-grained cherts 
obtained from Cretaceous limestone formations 
(Maiolica, Scaglia Variegata Alpina, Scaglia Rossa) 
represent the majority, while a couple of artefacts are 
manufactured on the coarse-grained chert from Jurassic 
Oolitic limestones. 

The foliate tools assemblage is almost exclusively 
composed by fragmented pieces (Fig. 9), presumably 
abandoned on site as a result of the supervening frac-
tures, characterized by the same heavy patinas affect-
ing the knapping surfaces. A couple of rejoints of frag-
mented elements show the existence of very large-
sized, bifacially shaped tools with symmetrical lense-like 
sections, which presumably reached 15 cm in length 
when complete (Fig. 9, n.3). A second rejoint allowed to 
reconstruct a preform 16 cm long, with large bifacial 
detachments and irregular edges, including a thick lat-
eral portion used as striking platform. The manufacture 
of leaf-points from the further reduction of bifacial 
backed knives is a known model in Central Europe 
(Nerudova & Neruda, 2017), although in this instance 
the thick back might be hint of an unintentional residual 
due to the premature discharge of the tool.  

Among the “finished” tools, only one is bearing a 
regular, thick back opposed to the bifacially-shaped 
edge (a leaf-knife), while the others are generally char-
acterized by a bilateral techno-functional affinity as sug-
gested by the two opposed cutting edges. These lateral 
edges are often symmetrical, ending in a pointed tip 
shaped by bilateral and bifacial flat retouching (Fig. 9, 
n.1-2). When at least one end is present, the leaf-point 
tool-type can be often recognized. However, in some 
cases a light asymmetry between the two edges is at-
tested. Moreover, the tip could be curved or convex 
(Fig. 9, n.5), maybe corresponding to the base of the 

Fig. 10 - Dimensional scatter-plot of leaf-shaped tools from Vajo Salsone, grouped by tool sub-types.  

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 
19 



   

 

tool, or rounded as a consequence of the loss of func-
tionality of the point. 

The edges of finished tools are extremely regular, 
with slightly convex sagittal profile; the sinuous edges 
belong to large-sized preforms, not yet subjected to the 
last refinement. In fact, the retouching stages are mani-
fold. At the beginning, large and deep flakes are alter-
natively detached (presumably with the organic ham-
mer) on both surfaces from the periphery of the tool. 
The next step is to regularize the thickness of the blank 
with invasive and flat detachments, prior to the refine-
ment of the edges both in the longitudinal and sagittal 
views through scaled and more marginal detachments. 

Bifacial symmetry can be observed on about half 
artefacts, while others have one surface flatter than the 
other. This feature is directly related to the initial blanks 
of the tools, which in several cases were recognized to 
be flakes or blades (Fig. 9, n.4). In these cases, the 
retouching could range from partial to totally covering 
and therefore different than the retouching recorded on 
the leaf-points and leaf-knives. Very few pieces (mainly 
preforms) seem to be shaped directly from raw blocks 
or plaquettes, while in most of the tools the raw blank is 
unrecognizable. Moreover, some tools present a strong 
bifacial asymmetry and flat-convex transversal section, 
with scalar retouching on the upper face: these tools 
might be classified as bifacial or leaf-shaped scrapers 
(Fig. 9, nn. 4-6). 

Since these are almost exclusively fragmented 
tools, a dimensional plot showing width and thickness 
has been produced (Fig. 10). According to it, leaf-points 
and knives clearly overlap leaf-shaped flakes and 

blades, taking in account that the latter are less bifacial-
ly symmetrical. Leaf-shaped scrapers are significantly 
thicker, implying different tool-types both in functionality 
and manufacturing stages. Preforms record larger sizes, 
although already defined in a bifacial and bilateral sym-
metry. 

 
5. DISCUSSION  
 
5.1. A model of the morfo-dynamic evolution of the 
slope and the genesis of the karst infill  

The geomorphological environment in which the 
site is located is characterized by a rock wall and a belt 
of the slope immediately below it. The wall is affected by 
the formation of large niches rockshelter-like, and locally 
also by the development of karst caves and pits. Niche 
formation is explained by the slow circulation of water 
within the fractures and pores of the rock with conse-
quent processes of clastic weathering and chemical 
erosion. Niches originate where the rock is most porous 
and / or fractured. The wall is also affected by collapse 
and toppling phenomena of large blocks of rock, favored 
by chemical erosion and often triggered by seismotec-
tonic events, such as high intensity earthquakes. 

The slope immediately below the wall is affected 
by these rockfall events with consequent crushing of the 
rock into both minute and coarse fragments, to consti-
tute discontinuous coverings of diamicton, interpretable 
as seismites, which can contribute to the filling of karst 
cavities such as pits or relict portions of caves. 
Seismites are hardly or not at all cemented, due to the 
slope processes and the repetition of the collapse 

Fig. 11 - Reconstruction of the morfo-dynamic evolution of the Vajo Salsone slope with indication of the main formation process of the 
karst infill.  

 
 
20 

 
 

Peresani M. et al. 



   

 

events. Inside the diamicton covers, conduits attributa-
ble to piping-type processes were observed, resulting 
from the flow of water and the removal of the finest rock 
fragments. 

Therefore, given this geomorphological context, 
the following hypothesis concerning the genesis of the 
site may be suggested: 1) the basal part of the wall 
presents a large shelter under the rock in which groups 
of Neanderthals settled depositing anthropogenic sedi-
ments containing both flint tools and animal bones; 2) 
following phenomena of rockfall and toppling of large 
blocks occurred, with consequent retreat of the wall and 
dismantling of the large shelter-niche; 3) the sediments 
of the shelter, in addition to calcite flowstones frag-
ments, have been exposed to the process of washout 
and transported along a suffusion conduit inside the 
diamicton cover, up to their accumulation place: a small 
karst pit interested by the formation of calcite flowstones 
below the diamicton cover. This pit functioned as a sedi-
mentary trap. 

It follows that, if seismic events of strong intensity 
(see, for example, the Verona earthquake of 1117) have 
occurred, in the most fragile areas, such as those locat-
ed in proximity of the faults with E-W orientation, cross-
ing the N-S elongated tectonic blocks, repeated collaps-
es occurred, with the  formation of covers of seismites.  

The formation of the archaeological deposit inside 
the sedimentary trap is contextual to the sedimentation 
of clasts resulted by the overlying diamicton and of frag-
mented speleothemes, as the finds were found both in 
the interstitial spaces between the blocks and crushed 
between each others (Fig. 11). The arrangement of all 
these elements appears chaotic, indicating a massive 
sedimentary process, a fact that seems to confirm an 
important event of collapse of an overlying structure 
(cave or shelter?) which was the primary location of the 
archaeological record. The state of conservation of the 
artifacts, mostly damaged by very fresh mechanical 
fractures, linked to the action of mines but without trac-
es of floating and prolonged transport suggests that the 
primary deposition site should not have been very far 
from the one of the discovery. 

This sedimentary evolution hinders the possibility 
of discriminating any frequentations distributed over a 
chronological interval which, however, remains to be 
determined. The only possible approximation would 
lead to consider a limited range: a broad stratification 
would be expressed by a greater variety in the faunal 
spectrum, by possible taphonomic differences between 
the artifacts, as well as by a greater variability in the 
techno-typological characteristics of the lithic industry, 
as instead is detected by cultural sequences that in-
clude well distinct levels of Mousterian industries with 
Quina, Discoid and Levallois technology, the latter de-
clined in a series of techno-economic and typometric 
aspects (eg Fumane cave: Peresani, 2012; San Bernar-
dino cave: Peresani, 1996). 

 
5.2. Mammal fauna, ecology, and possible chrono-
logical range 

According to preliminary results obtained from the 
study, the faunal spectrum is mainly composed of medi-

um and large sized ungulates, with a clear prevalence 
of red deer and few remains of roe deer, chamois and 
elk. Regarding the ecological context, this faunal as-
semblage is representative of close and open forests 
with transitive and discontinuous Alpine grasslands or 
sparse vegetation on carbonate rocks (an environment 
vital to ibex), completed by humid environments and 
watercourses. The dominance of red deer and roe deer 
may suggests that forests were commonly the Neander-
thals’ most exploited hunting environment. However, 
caution is needed in declaring this assumption given the 
ecological plasticity of these kind of cervids, which can 
live in different ecosystems, although the roe deer pri-
mary habitat is forest clearings, hedges, and woodland 
edge (Danilkin, 1996; Geist, 1999). Nevertheless, the 
low number of remains attributed to elk suggests the 
presence of humid areas presumably in the valley bot-
tom, as attested by currently available evidence from 
Fumane cave (Cassoli & Tagliacozzo, 1994; Peresani 
et al., 2011; Romandini et al., 2014), Tagliente Rock-
shelter (Thun Hohenstein & Peretto, 2005) and San 
Bernardino cave (Terlato et al., 2019). The high inci-
dence of cervids is consistent with the faunal records 
from the late Middle Palaeolithic stratigraphic sequence 
of Fumane which, from units A11-A10 shows a progres-
sive increase of these species to the detriment of those 
favoring open environments (Fiore et al., 2004; Roman-
dini et al., 2020). Dominance of cervids is also recorded 
at Tagliente (Thun-Hohenstein & Peretto, 2005) and 
San Bernardino in layers of comparable age (Terlato et 
al., 2021). Moreover, the absence of large bovids at 
Vajo Salsone despite their representativeness of eco-
logical conditions diversity, spanning from dense forests 
with wetlands and small streams, to hilly grasslands and 
plains, disagrees with the presence, although ephemer-
al, of these ungulates at both Fumane and Tagliente 
(Terlato et al., 2019) and, above all, at De Nadale cave 
in the Berici Hills, where the higher frequency of bovids 
has also been related to the MIS 4 age of this site 
(Livraghi et al., 2021). 

Therefore, despite being so preliminar, these first 
data of the Vajo Salsone faunal assemblage agree with 
ecological contexts existing at the regional scale during 
MIS 4 and MIS 3, thus supporting this chronological 
position for the site. The main reference vegetation data 
in the alpine foreland are provided from the Fimon pol-
len core, where persistent afforestation (eventually lim-
ited during millennial scale cold events) is recorded 
throughout the entire early and middle part of the last 
glaciation (Early to Middle Würm; Pini et al., 2010). MIS 
3 records clear evidence of afforestation persisting at a 
long-term scale south of the Alps, although forest with-
drawals with expansion of grasslands and dry shrub-
lands occurred as a consequence of colder climatic 
conditions (Badino et al., 2020) which reinforced the 
presence of Alpine ibex, chamois, and marmot at low 
altitudes, as well as the presence of micromammals in 
steppic environments, and the diffusion of birds in open 
environments (Romandini et al., 2020). However, these 
conditions are not recorded at Vajo Salone. Concerning 
carnivores, the presence of wolf at Vajo Salsone and 
other sites in the region confirms the existence of open 

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 
21 



   

 

spaces and an abundant ungulate biomass, whereas, 
red fox is suited to a predominantly forested landscape.  

 
5.3. A first insight on the origin of the zooarchaeo-
logical assemblage 

According to the taphonomic data, anthropic modi-
fications such as cutmarks, scraping marks and impacts 
have been recognized on cervid bones (red deer, elk), 
showing that during this phase of occupation of the Vajo 
Salsone site, hunting activity was mainly focused on this 
ungulate.  

Hunting and related activities of this ungulate are 
largely recorded in the Pre-Alpine belt, where tapho-
nomic data indicate that the predominant association of 
game is consistent with the ecological conditions in the 
proximity of each specific site and shows shifts correlat-
ed with the most relevant climatic oscillations. De 
Nadale cave in the Berici hills was frequented by hunt-
ers of giant deer, red deer, bison and auroch during MIS 
4 (Livraghi et al., 2021). San Bernardino Cave, Fumane 
Cave and Tagliente Rockshelter during MIS 3 share 
similar exploitation models consisting of the selection of 
young adult and adult prey and of primary butchering. 
The main hunted ungulates during the final Mousterian 
were thus mostly red deer, roe deer and giant deer with 
lesser amounts of chamois and ibex (Fiore et al., 2004; 
Thun Hohenstein & Peretto, 2005; Peresani et al., 2011; 
Romandini et al., 2014; Terlato et al., 2021) and limited 
exploitation of elk, wild boar and Bos/Bison (Terlato et 
al., 2019). The exploitation of marmots, beavers 
(Romandini et al., 2018a) and of some carnivores like 
bear, wolf and fox (Romandini et al., 2018b) is also doc-
umented and suggests recovery for meat, marrow and 
fur.  

 
5.4. The cultural outline of Vajo Salsone in the North
-eastern Italy 

Even though most of the evidence from the Vajo 
Salsone assemblage is comparable with that of other 
sites in the region and the northern Mediterranean area, 
it does provide new details of the Late Pleistocene cul-
tural scenario and contributes in making the Middle 
Paleolithic picture more complex. Not surprisingly, as it 
occurred in almost all sites around the North-Adriatic 
region, Levallois technology is largely prevalent in the 
lithic industries. This flaking method either forms the 
substrate or contributes to cultural complexes tradition-
ally labelled as Typical Mousterian rich in scrapers, 
Eastern Charentian Mousterian, Pontinian, in a large 
area from the Rhône valley to the western Balkans 
along the Alpine arch and Apennine ridge, displaying 
variations in each different facies in relation to the fre-
quence of retouched tools and other aspects (Palma di 
Cesnola, 1996; Peretto, 1992). Nevertheless, none of 
these industries had the specific target of producing 
neither points nor converging scrapers at a so relevant 
number like at Vajo Salsone. 

Although it is still lacking of a precise chronological 
position, the Levallois technology of Vajo Salsone is as 
much variable as other sites even in the Alpine footbelt 
(Peresani, 2001) and subalpine area like Monte Versa 
on the Euganean hills (attributed to the last interglacial; 

Peresani, 2000-2001), Monticello (undated, Duches & 
Peresani, 2009) and Cavaso del Tomba (undated: De 
Stefani et al., 1999-2001), among others. At Vajo 
Salsone, this system mostly focused on longer-than-
wide blanks using the unidirectional recurrent modality 
and, to a lesser extent, by means of centripetal flake-
making as it has been observed in the alpine foreland at 
sites dated to MIS 3 (Peresani, 2011; Delpiano et al., 
2019b), and along the northern slope of the Apennines 
(Lenzi & Nenzioni, 1996; Fontana et al., 2018). Further-
more, at the present state we cannot confirm whether 
the different modalities (unidirectional and centripetal) 
expressed by the Levallois variability at Vajo Salsone 
relate to different core surface maintenance strategies 
aimed at the setting of specific predetermination param-
eters for each lithic material exploited (coarsely vs finely 
texture chert). As a matter of fact, locally available re-
sources were employed for flake-making, with a subtle 
prevalence of Maiolica and Scaglia Rossa over the 
coarser materials catched in the carbonatic sandstones 
along the Vajo Salsone slope. As the provisioning of 
nodules, blocks and plates appears to be unconstrained 
in the site surroundings - we did not observe the use of 
stream pebbles -, then preferences in relation to the 
main flaking modalities employed might be related to 
specific technical economical requirements. Further 
analyses are needed to verify whether mechanical apti-
tude of the chert types or economic reasons shall be 
considered forcing agents in the design of tool sets. 
Nonetheless, these latter were produced through the 
exploitation of all kinds of raw material. 

Moreover, throughout the observation of the exam-
ined assemblage we did not record the presence of the 
same Levallois reduction patterns already attested in 
certain stratified sequences, where the unidirectional 
exploitation systematically turned to the centripetal one 
when approaching the end of core reduction (Peresani, 
1996; 2012). On the contrary, in our instance unidirec-
tional exploitation was applied until almost the end of 
the reduction sequence, as the presence of elongated 
flakes with thin and regular edges selected for the side 
and converging scrapers suggests. These patterns 
show that the above goals were achieved constantly, 
without caring for the optimization of the available re-
sources, which may have not been subject to deficien-
cies in terms of either quality or quantity. 

On a regional scale, at open-air sites the Levallois 
reduction stops at an intermediate grade of the potential 
exploitation. Conversely, in shelters the exploitation was 
more exhaustive and affected structural and typometric 
arrangements, to such an extent that microlithism was 
observed in some assemblages (Peresani, 1996). Set-
tlement dynamics were thus responsible both for frac-
tionation in reduction sequences and for variations in 
the grade of retouching of the predetermined blanks and 
by-products, these latter being left unretouched at Vajo 
Salsone. 

Still, although preliminary, the data presented here 
express one further element of lithic variability which 
consists in the co-presence of the Discoid and Laminar 
volumetric concepts, whereas we do not exclude other 
more expeditious productions. Discoid industries have 

 
 
22 

 
 

Peresani M. et al. 



   

 

been described at Fumane and Rio Secco caves, where 
a high resolution in Neanderthal occupations within the 
first part of MIS 3 shows diachronic variability framed in 
cultural and behavioural contexts. At Fumane, the Dis-
coid concept substitutes the Levallois along a densely 
stratified sequence (Peresani, 2012; Delpiano et al., 
2018), while at Rio Secco it is found in the same context 
with the Levallois (Peresani et al., 2014). Concerning 
volumetric laminar production, it is ephemeral in the 
area and only represented by bladelet making at Fu-
mane unit A5-A6 (Peresani et al., 2013) or remains 
poorly known and undated as at Tagliente (Carmignani, 
2021).  

 
5.5. Foliate tools: comparing evidence in Italy and 
behind the Alps 

The foliate tools of Vajo Salsone, including leaf 
points, knives and leaf-shaped flakes, are a unique 
evidence for the Northern Adriatic region. Nonetheless, 
in the southern margin of the Po plain, where tens of 
open-air Lower and Middle Palaeolithic sites are pre-
served on the Pleistocene terraces of the lower Apen-
nines of Emilia-Romagna, five sites located between 
Bologna and Imola (Fig. 1) contain isolated and some-
times fragmented leaf-points, found in assemblages 
characterized by developed laminar tendency achieved 
with the Levallois method. These sites are generally 
related to the Bellaria/San Biagio stratigraphic unit that 
is supposed to be framed in MIS 7 and covered by MIS 
6 loess (Fontana et al., 2010; 2018). However, lack of 
absolute dating raised questions and issues: indeed, 
there would be a chronological divergence between 
these and other regional Levallois complexes that re-
cently yielded MIS 4 (Ghiardo) or early MIS 3 (Monte 
Netto) dates (Cremaschi et al., 2015; Delpiano et al., 
2019b). 

Outside the boundaries of the Great Adriatic Po 
Region, archaeological sites yielding Middle Palaeolithic 
leaf-points are mainly situated in the southern part of 
Germany, spreading along the Danube and its tributar-
ies in an E-W direction. Most of the sites are located in 
the Swabian and Franconian Jura, preserved in caves 
and rock-shelters, but also in open air sites resulting 
from surface collections of isolated leaf-points. The 
highest number of leaf-points has been found in Bavar-
ia, decreasing towards West and North. Assemblages 
of southern Germany containing leaf-points attracted 
scientific attention very early in the 20th century. Except 
for a few examples, these sites have been excavated 
within the first half of the last century and, due to this 
reason, they are affected by lacking documentation, 
unreliable chronology and poor contextual information 
(Uthmeier, 2004; Richter, 2009). Since the pioneering 
work of G. Freund, and later T. Uthmeier (2004) and M. 
Bolus (2004), no comprehensive research study fo-
cused exclusively on leaf-points and their role in the late 
Middle Palaeolithic of southern Central Europe. Both 
Uthmeier (2004) and Bolus (2004), as well as Richter 
(2009), hypothesized that leaf-point assemblages in 
southern Germany are part of the regional Middle Pal-
aeolithic (Richter, 1997) and should not be interpreted 
as an innovative behaviour of late Neanderthals. Ra-

ther, leaf-points might be related to specific task activi-
ties, probably hunting, within the context of a larger 
settlement system.  

Nonetheless, leaf points are often considered as 
cultural expressions of Late Pleistocene Neanderthals 
in Europe. Particularly relevant to this issue is the so-
called Szeletian (Allsworth-Jones, 1986), an archaeo-
logical leaf-points complex distributed in Poland, Hun-
gary and Moravia, although it is still far from being clear-
ly defined in time, space and technological characteris-
tics of stone tool production (Mester, 2018). To the 
West, in southern Germany, the outshoot of the Sze-
letian should be considered the so-called Altmühlian 
(Bohmers, 1939, 1951) - despite the significance of this 
complex as transitional industry has been recently dis-
cussed (Richter, 1997; Bolus, 2004; Uthmeier, 2004) - 
being lately considered as the last phase of Keilmesser-
gruppen/M.M.O. complex, which is the main cultural 
expression of central european Neanderthals (Richter, 
2016; Delpiano & Uthmeier, 2020). For Bavaria, only 
the so-called Lincombian-Ranisian-Jerzmanovician 
(LRJ) (Flas, 2011) is accepted as a potential transitional 
complex leading from the Middle to the Upper Palaeo-
lithic. Further east, the Bohunician complex is wide-
spread in Moravia, where it is characterized by Levallois 
alongside blade technology, with the presence of bifa-
cial leaf-points in some sites (Tostevin & Skrdla, 2006); 
however, doubts have been raised on the links between 
Bohunician and the neighboring transitional techno-
complexes (Neruda, 2009). Incertitude thus concerns 
the chronology and geographic distribution of complex-
es in Central and Eastern Europe containing bifacially 
worked leaf-points. Furthermore, foliates do not seem to 
be constrained to the phase immediately before the 
Upper Palaeolithic, but may appear discontinuously and 
selectively from possibly as early as MIS 7 to MIS 3. 
Sites bearing leaf-shaped tools in the Balkans and in 
northern Germany appear to belong to MIS 5 or MIS 7, 
even though bifacial reshaping and rejuvenation may 
have had heavily modified tools initially manufactured 
with a totally different conception (Kot, 2017). In Vajo 
Salsone, both leaf-points, leaf-knives and other leaf-
shaped flake tools are attested, abandoned at different 
manufacturing stages. These data suggest that they 
were possibly produced directly according to this mental 
template, rather than representing a consequence of 
the reduction of unifacial tools or other bifacial tools. 

 
6. CONCLUSIONS  

 
The discovery of the site of Vajo Salsone contrib-

utes to shed new lights on the complexity of the Middle 
Palaeolithic in the North of Italy, where Mousterian in-
dustries are mostly based on the application of the Le-
vallois technology, as recorded from cave and open air 
sites dated from MIS 7 to early MIS 3. This cultural sce-
nario is not static. In fact, Quina and Discoid technologi-
cally related industries are documented as well, except 
for few contexts. Furthermore, the discovery of leaf 
points, knives and leaf-shaped flakes in the alpine foot-
belt reinforces our view about the variability of the Ne-
anderthal world, thus providing a unique evidence not 

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 
23 



   

 

only for the North of Italy but also for the whole Adriatic 
region. It confirms the reliability of previous findings 
reported at open-air sites on the northern slope of the 
Apennines, although isolated from their originary pedo-
stratigraphic contexts. Dating the Apennine and Vajo 
Salsone evidence using biostratigraphic and geochron-
ological methods, whenever these can be applied, is 
needed for assessing the values of affinities or differ-
ences between the foliate tools featured industries at 
the South and the North of the Alpine range, in terms of 
biological and cultural population dynamics. 

 
ACKNOWLEDGEMENTS 

The authors are indebted to the Archaeological 
Superintendence (Dott. Gianni De Zuccato, So-
printendenza Archeologia, Belle Arti e Paesaggio per le 
Provincie di Verona, Vicenza e Rovigo) for permission 
to the study of the archaeological assemblage, to Refill 
s.r.l. the owner of the Vajo Salsone area for making 
possible excavation and for access the site area, to all 
students and collaborators of the University of Ferrara 
who sieved and carefully scrutinized archaeological 
sediments and two reviewers who contributed to 
strengthen our paper. G. Cattabriga revised the english 
text. Authors contributions: M.P. designed and directed 
fieldwork and studies at Vajo Salsone, D.V. and A.P. 
conducted the rescue excavation, U.S. analyzed the 
geomorphological context, D.V. and A.P. provided field 
description of the stratigraphy and sediments, G.A. and 
U.T.H. examined archaeofauna remains, J.M.L.G and 
H.A.B. examined the small vertebrates, D.D., V.E. and 
M.P. analyzed the lithic assemblage, D.M. made the 
geomorphological map, LB., C.F., F.P., P.P. treated and 
prepared artefacts and bones for analysis, M.P. wrote 
and composed the manuscript with contributions by 
G.A., D.D, H.A.B., J.M.L.G., A.P., U.S., D.V., Y.T., and 
U.T.H.  

 
REFERENCES 
 
Aleo A., Duches R., Falcucci A., Rots V., Peresani M., 

(2021) - Scraping hide in the early Upper Paleoli-
thic: insights into the life and function of the Pro-
toaurignacian endscrapers at Fumane cave. Ar-
chaeological and Anthropological Sciences, 13, 
137. 

Allsworth-Jones P. (1986) - The Szeletian and the tran-
sition from Middle to Upper Palaeolithic in Central 
Europe. Clarendon Press, Oxford. 

Amori G., Contoli L., Nappi A. (2008) - Fauna d’Italia. 
Mammalia II. Erinaceomorpha, Soricomorpha, 
Lagomorpha, Rodentia, Calderini, Milano. 

Aspes A. (2003) - Preistoria Veronese. Contributi e 
aggiornamenti. Memorie del Museo Civico di Sto-
ria Naturale di Verona, II serie, Sezione Scienze 
dell’Uomo, 5. 

Badino F., Pini R., Ravazzi C., Margaritora D., Arrighi 
S., Bortolini E., Figus C., Giaccio B., Lugli F., Mar-
ciani G., Monegato G., Moroni A., Negrino F., Oxi-
lia G., Peresani M., Romandini M., Ronchitelli A., 
Spinapolice E.E., Zerboni A., Benazzi S. (2020) - 
An overview of Alpine and Mediterranean palaeo-

geography, terrestrial ecosystems and climate 
history during MIS 3 with focus on the Middle to 
Upper Palaeolithic transition. Quaternary Interna-
tional, 551, 7-28. 

Barone R. (1980) - Anatomia comparata dei mammiferi 
domestici. Vol. 1: Osteologia. Bologna. 

Beherensmeyer A.K., (1978) - Taphonomic and ecologi-
cal information from bone weathering. Paleobiolo-
gy, 4, 150-162. 

Bennett J.L. (1999) - Thermal alteration of buried bone. 
Journal of Archaeological Science, 26, 1-8. 

Berto C., (2013) - Distribuzione ed evoluzione delle 
associazioni a piccoli mammiferi nella peninsola 
italiana durante il Pleistocene superiore (PhD The-
sis). Università degli Studi di Ferrara, Ferrara. 

Berto C., López-García J.M., Luzi E. (2019) - Changes 
in the Late Pleistocene small-mammal distribution 
in the Italian Peninsula. Quaternary Science Revi-
ews, 225, 106019. 

Berto C., Luzi E., Montanari G., Guerreschi A., Fontana 
F. (2018) - Climate and landscape in Italy during 
Late Epigravettian. The Late Glacial small mam-
mal sequence of Riparo Tagliente (Stallavena di 
Grezzana, Verona, Italy). Quat. Sci. Rev., 184, 
132-142. 

Bertola S. (2016) - Southern Alps (Trento plateau) and 
Northern Apennines cherts: ages and distribution. 
In: Ressources lithiques, productions et transferts 
entre Alpes et Méditerranée. Séances de la SPF, 
55-75. 

Binford L.R. (1981) - Bones: Ancient Men and Modern 
Myths. NY Academic, 320. 

Binford L.R. (1984) - Faunal Remains from Klasies Ri-
ver Mouth. Academic Press, Orlando, 305. 

Bini A., Cita M.B., Gaetani M. (1978) - Southern Alpine 
lakes; hypothesis of an erosional origin related to 
the Messinian entrenchment. In: Cita M.B., Ryan 
W. B. F. (eds). Messinian erosional surfaces in the 
Mediterranean. Marine Geology, 27 (3-4), 271 -
288. 

Blasco R., Rosell J., Fernández Peris J., Cáceres I., 
Vergés J.M. (2008) - A new element of trampling: 
an experimental application on the level XII faunal 
record of Bolomor cave (Valencia, Spain). Journal 
of Archaeological Science, 35, 1605-1618 

Boëda É. (1994) - Le Concept Levallois: variabilité des 
mèthodes. Monographies Centre Rercherches 
Archéologiques 9, CNRS, 280 

Bohmers A.(1939) - Die Ausgrabungen in den Höhlen 
von Mauern. Germanien 4, 151-156. 

Bohmers A. (1951) - Die Höhlen von Mauern. Palaeohi-
storia, 1, 1-107. J. B. Wolters, Groningen. 

Bolus M. (2004) - Settlement analysis of sites of the 
Blattspitzen complex in Central Europe In: Conard, 
N.J. (Ed.), Settlement dynamics of the Middle Pa-
laeolithic and Middle Stone Age, 201-226. Kerns 
Verlag, Tubingen. 

Bordes F. (1961) . Typologie du Paléolithique ancien et 
moyen. Institut Préhistoire Universitè Bordeaux 1, 
81. 

Bosellini A., Carraro F., Corsi M., De Vecchi G.P., Gatto 
G.O., Malaroda R., Sturani C., Ungaro S., Zanettin 

 
 
24 

 
 

Peresani M. et al. 



   

 

B. (1967) - Foglio 49 Verona. Note illustrative della 
Carta Geologica d’Italia. Servizio Geologico d’Ita-
lia, Roma. 

Brain C.K. (1981) - The Hunter or the Hunted? An Intro-
duction to African Cave Taphonomy. University of 
Chicago Press, Chicago, 376. 

Broglio A. (1992) - Mountain Sites in the context of the 
North-East Italian Upper Palaeolithic and Mesoli-
thic. Proc. Preistoria Alpina, 28/1, 293-310. 

Broglio A. (1993) - L'Aurignacien au Sud des Alpes. 
Actes XII Congrès Int. Sciences Préhist. Protohist. 
Bratislava 1991, 193-203. 

Broglio A., De Stefani M., Tagliacozzo A., Gurioli F., 
Facciolo A. (2006) - Aurignacian dwelling structu-
res, hunting strategies and seasonality in the Fu-
mane Cave (Lessini Mountains). In: Vasilev S.A., 
Popov V.V., Anikovich M.V., Praslov N.D., Sinitsyn 
A.A., Hoffecker J.F. (Eds.), Kostenki and the Early 
Upper Paleolithic of Eurasia: General Trends, Lo-
cal Developments. Nestor-Historia, Saint Peter-
sburg, 263-268. 

Carmignani L. (2021) - Between the flake and the blade: 
Associated systems of production at Riparo Ta-
gliente (Veneto, northern Italy). Journal of Lithic 
Studies, 4/1, 1-37. 

Carraro F., Malaroda R., Piccoli G., Sturani C., Venzo 
S. (1969) - Note illustrative della carta geologica 
d’Italia alla scala 1:100000. Foglio 48 Peschiera 
del Garda, ministero dell’industria del commercio e 
dell’artigianato. Direzione generale delle miniere. 
Servizio geologico d’Italia, Poligrafica e cartavalori, 
Ercolano. 

Cassoli P.F., Tagliacozzo A. (1994) - I resti ossei di 
macromammiferi, uccelli e pesci della Grotta Mag-
giore di San Bernardino sui Colli Berici (VI): consi-
derazioni paleoeconomiche, paleoecologiche e 
cronologiche. Bullettino di Paletnologia, 85, 1-71. 

Castellarin A. (1981) - Carta tettonica delle Alpi Meridio-
nali alla scala 1:200.000. C.N.R., P.F. Geodinami-
ca, 441, 220. 

Castellarin A., Vai G.B., Cantelli L. (2006) - The Alpine 
evolution of the Southern Alps around the Giudica-
rie faults: Late Cretaceous to Early Eocene trans-
fer zone. Tectonophysics, 414, 203-223. 

Castiglioni G.B., Cremaschi M., Guerreschi A., Mene-
ghel M., Sauro U., Van Vliet Lanoë B. (1990) - The 
loess deposits in the Lessini plateau, in Cremaschi 
M. (Ed.), The Loess in Northern and Central Italy: 
a Loess Basin between the Alps and the Mediter-
ranean Region. Quaderni di Geodinamica Alpina e 
Quaternaria, 1, 41-59. 

Chelidonio G., Sommaruga M. (2011) - Morfologie e 
tracce di frequentazioni paleolitiche a Monte Vegro 
(Illasi). La Lessinia ieri oggi domani, 79-86. 

Chelidonio G., Stocchiero L. (1991) - Monte Precastio 
(Tregnago/Verona): prime note sulla nuova segna-
lazione di un sito riferibile al Paleolitico Inferiore-
Medio nella Lessinia orientale. Quaderni di Ar-
cheologia del Veneto, VII, 123-135.  

Colamussi V. (2002) - Studi climatici sul Quaternario 
mediante l’uso dei Micromammiferi (PhD Thesis). 
Universita di Ferrara, Firenze, Parma. 

Cremaschi M. (1990) - Depositional and post-
depositional processes in rock shelters of Northern 
Italy during the late Pleistocene: their paleoclimatic 
and paleoenvironmental significance. Quaternaire, 
1, 51-64. 

Cremaschi M., Zerboni A., Nicosia C., Negrino F., Rod-
night H., Spötl C. (2015) - Age, soil-forming pro-
cesses, and archaeology of the loess deposits at 
the Apennine margin of the Po plain (northern 
Italy): New insights from the Ghiardo area. Quater-
nary International, 376, 173-188. 

Danilkin A. (1996) - Behavioural ecology of Siberian and 
European roe deer. Chapman & Hall, London, 
300. 

De Stefani M., Peresani M. (1999-2001) - Il Paleolitico 
della Valcavasia (Prealpi Trevigiane). Atti Società 
Preistoria e Protostoria Friuli-Venezia Giulia, XII, 
127-149. 

De Zanche V., Sorbini L., Spagna V. (1977) - Geologia 
del territorio del comune di Verona. Memorie Mu-
seo Civico di Storia Naturale di Verona (II s.), sez. 
Sc. Terra, 1, 51. 

Delagnes A. (1992) - L’organisation de la production 
lithique au Paléolithique moyen: approche techno-
logique à partir de l’étude des industries de La 
Chaise-de-Vouthon (Charente). PhD Dissertation, 
Université Paris X. 

Delpiano D., Heasley K., Peresani M. (2018) - Asses-
sing Neanderthal land use and lithic raw material 
management in Discoid technology. Journal of 
Anthropological Science, 96, 1-22. 

Delpiano D., Zupancich A., Peresani M. (2019a). Inno-
vative Neanderthals: Results from an integrated 
analytical approach applied to backed stone tools. 
Journal of Archaeological Science, 110, 105011. 

Delpiano D., Peresani M., Bertola S., Cremaschi M., 
Zerboni A. (2019b) - Lashed by the wind. Short-
term Middle Palaeolithic occupations within the 
loess-palaeosoil sequence at Monte Netto 
(Northern Italy). Quaternary International, 502, 137
-147. 

Delpiano D., Uthmeier T. (2020) - Techno-functional 
and 3D shape analysis applied for investigating 
the variability of backed tools in the Late Middle 
Paleolithic of Central Europe. PloS ONE 15, 1-55. 

Domínguez-Rodrigo M., De Juana S., Galán A., Rodrí-
guez M. (2009) - A new protocol to differentiate 
trampling marks from butchery cut marks. Journal 
of Archaeological Science 36, 2643-2654. 

Domínguez-Rodrigo M., Piqueras A. (2003) - The use of 
tooth pits to identify carnivore taxa in tooth-marked 
archeofaunas and their relevance to reconstruct 
hominid carcass processing behaviors. Journal of 
Archaeological Science, 30, 1385-1391. 

Duches R., Peresani M. (2009) - Tra Monti Berici e Colli 
Euganei nel Paleolitico. Interpretazione prelimina-
re del sito di Monticello di Barbarano (VI) sulla 
base dello studio tecno-economico dell’insieme 
litico. Preistoria Alpina, 44, 159-171. 

Fernández-Jalvo Y., Andrews P. (2016) - Atlas of tapho-
nomic identifications. Vertebrate Paleobiology and 
Paleoanthropology Series, Springer. 

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 
25 



   

 

Fiore I., Gala M., Tagliacozzo A. (2004) - Ecology and 
subsistence strategies in the Eastern Italian Alps 
during the Middle Palaeolithic. International Jour-
nal Osteoarchaeology, 14, 273-286. 

Fischer J.W. (1995) - Bone surface modifications in 
zooarchaeology. Journal of Archaeological Method 
and Theory, 2, 7-68. 

Flas D. (2011) - The Middle to Upper Paleolithic transi-
tion in Northern Europe: the Lincombian-Ranisian-
Jerzmanowician and the issue of acculturation of 
the last Neanderthals. World Archaeology, 43(4), 
605-627. 

Fontana F., Nenzioni G., Peretto C. (2010) - The sou-
thern Po plain area ( Italy ) in the mid-late Pleisto-
cene�: Human occupation and technical beha-
viours. Quaternary International, 223-224, 465-
471. 

Fontana F., Nenzioni G., Pagani C. (2018) - Depositi 
paleolitici e industrie nell’area bolognese orientale. 
Nuovi dati e dinamiche interpretative (a cura di F. 
Lenzi). Museo della Preistoria Luigi Donini e Istitu-
to beni Culturali della Regione Emilia Romagna. 

Galadini F., Galli P., Cittadini A., Giaccio B. (2001) - 
Late Quaternary fault movements in the Baldo-
Lessini Mts sector of the Southalpine area 
(northern Italy). Netherlands Journal of Geoscien-
ces/Geologie en Mijnbouw, 80, 187-208. 

Galán A.B., Rodríguez M., De Juana S., Domínguez-
Rodrigo M. (2009) - A new experimental study on 
percussion marks and notches and their bearing 
on the interpretation of hammerstone-broken fau-
nal assemblages. Journal of Archaeological Scien-
ce, 36, 776-784. 

Geist V. (1999) - Deer of the World. Swan Hill Press, 
Shrewsbury, England, 432. 

Grayson D. K. (1984) - Quantitative Zooarchaeology, 
Academic Press, New York. 

Grimaldi S. (1996) - Mousterian reduction sequences in 
central Italy. In: Bietti A., Grimaldi S. (Eds.), Re-
duction Processes (châines opératoires) in the 
European Mousterian. Quaternaria Nova, VI, 279-
310. 

Guette C. (2002) - Révision critique du concept de débi-
tage Levallois à travers l’étude du gisement mou-
stérien de Saint-Vaast-la-Hougue/le Fort 
(chantiers I-III et II, niveaux inférieurs) (Manche, 
Franche). Bulletin de la Société Préhistorique Fra-
nçaise, 99 (2), 237-248. 

Guidoboni E., Comastri A., Boschi E. (2005) - The 
‘‘exceptional’’ earthquake of 3 January 1117 in the 
Verona area (northern Italy): A critical time review 
and detection of two lost earthquakes (lower Ger-
many and Tuscany). Journal of Geophysical Re-
search, 110b, 20. 

Guidoboni E., G. Ferrari, Postpischl D. (1985) - Il terre-
moto del 3 gennaio 1117, p. 810., 47 foto, 12 map-
pe, Rep. RPT 37/85, Enel, Rome. 

Habemehl K. H. (1961) - Altersbestimmung bei Haustie-
ren, Peltzieren und beim Jagdbaren. Paul Parey, 
Berlin, Germany. 

Higham T.F.G., Brock F., Peresani M., Broglio A., Wood 
R., Douka K. (2009) - Problems with radiocarbon 

dating the middle and upper Palaeolithic transition 
in Italy. Quaternary Science Reviews, 28, 1257-
1267. 

Inizan M.L., Reduron-Ballinger M., Roche H., Tixier J. 
(1999) - Technology and terminology of knapped 
stone, Préhistorie de la pierre taillée, tome 5, 
Crep, Nanterre. 

Jéquier C., Peresani M., Romandini M., Delpiano D., 
Joannes-Boyau R., Lembo G., Livraghi A., López-
García J.M., Obradovi� M., Nicosia C. (2015) - The 
De Nadale Cave, a single layered Quina Mouste-
rian site in the North of Italy. Quartär, 62, 7-21. 

Klein, R. G., Cruz-Uribe, K. (1984) - The Analysis of 
Animal Bones from Archeological Sites, The Uni-
versity of Chicago Press, Chicago. 

Kot M. (2017) - Bifacial and unifacial technology: A real 
difference or a problem of typo-technological ap-
proach? The example of the Ehringsdorf assem-
blage. Quaternary International, 428, 66-78. 

Lenzi F., Nenzioni G. (1996) - Lettere di Pietra. I deposi-
ti pleistocenici: sedimenti, industrie e faune del 
margine appenninico bolognese. Editrice Compo-
sitori, Bologna. 

Livraghi A., Fanfarillo G., Dal Colle M., Romandini M., 
Peresani M. (2021) - Neanderthal ecology and 
exploitation of cervids at the onset of MIS4: A stu-
dy on De Nadale cave, Italy. Quaternary Interna-
tional, 586, 24-41. 

López-García J.M. (2011) - Los micromamíferos del 
Pleistoceno superior de la Península Ibérica. Evo-
lución de la diversidad taxonómica y cambios pa-
leo-ambientales y paleoclimáticos. Editorial Aca-
démica Española, Saarbrücken, 368. 

López-García J.M., Berto C., Peresani M. (2019) - Envi-
ronmental and climatic context of the hominin oc-
currence in northeastern Italy from the late Middle 
to Late Pleistocene inferred from small-mammal 
assemblages. Quateranry Science Reviews, 216, 
18-33. 

López-García J.M., Dalla Valle C., Cremaschi M., Pere-
sani M. (2015) - Reconstruction of the Neanderthal 
and Modern Human landscape and climate from 
the Fumane cave sequence (Verona, Italy) using 
small-mammal assemblages. Quaternary Science 
Reviews, 128, 1-13. 

López-García J.M., Luzi E., Peresani M. (2017) - Middle 
to Late Pleistocene environmental and climatic 
reconstruction of the human occurrence at Grotta 
Maggiore di San Bernardino (Vicenza, Italy) 
through the small-mammal assemblage. Quater-
nary Science Reviews, 168, 42-54. 

Lyman R.L. (1994) - Vertebrate Taphonomy. Cambridge 
University Press, Cambridge. 

Magaldi D., Peretto C., Sauro U. (1981) - Depositi di 
versnte contenenti reperti del Paleolitico inferiore 
nell’Alta Valpantena. Bollettino Museo Storia Natu-
rale di Verona, VII, 657-673. 

Magaldi D. & Sauro U. (1982) - Landforms and soil evo-
lution in some karstic areas of the Lessini Moun-
tains and Monte Baldo (Verona, Northern Italy). 
Geografia Fisica Dinamica Quaternaria, 5, 82-101. 

Mallow D., Ludwig D., Nilson G. (2003) - True Vipers: 

 
 
26 

 
 

Peresani M. et al. 



   

 

Natural History and Toxinology of Old World Vi-
pers. Malabar, Florida: Krieger Publishing Compa-
ny. 

Mallye J.B., Thiébaut C., Mourre V., Costamagno S., 
Claud E., Weisbeckerg P. (2012) - The Mousterian 
bone retouchers of Noisetier Cave: experimenta-
tion and identification of marks. Journal of Ar-
chaeological Science, 39, 1131-1142. 

Mantovani E., Brancolini G., Babbucci D., Tamburelli C., 
Viti M. (2021) - Possible Multiple Sources of the 
Strong 1117 Po Plain Earthquake, inferred from 
the Plio-Quaternary Evolution of the Northern 
Adriatic Area. International Journal of Geoscien-
ces, 12, 381-403. 

Margaritora D., Dozio A., Chelidonio G., Turrini M.C., 
Peresani M. (2020) - The Lower and Middle Pa-
laeolithic settlements in the Baldo-Lessini moun-
tains. Results from a GIS investigation. Alpine and 
Mediterranean Quaternary, 33/1, 115-132. 

Mariezkurrena K. (1983) - Contribution al conocimiento 
del desarrollo de la denticion y el esqueleto post-
craneal de Cervus elaphus, «Munibe», 35, 149-
202. 

Massironi M, Zampieri D., Bianchi M, Schiavo A, Fran-
ceschini A. (2009) - Use of PSInSAR data to infer 
active tectonics: Clues on the differential uplift 
across the Giudicarie belt. Tectonophysics, 476, 
363-369. 

Mester Z. (2018) - The problems of the Szeletian as 
seen from Hungary. Recherches Archéologiques, 
9, 19-48. 

Mietto P., Sauro U. (2000) - Le Grotte del Veneto: pae-
saggi carsici e grotte del Veneto, Regione Veneto. 
La Grafica Editrice, Verona, 480. 

Montoya C., Duches R., Fontana F., Peresani M., Vi-
sentin D. (2018) - Peuplement tardiglaciaire et 
holocène ancien des Préalpes de la Vénétie (Italie 
Nord Orientale): éléments de confrontation. In: 
Averbouh A., Bonnet-Jacquement P., Cleyet-Merle 
J.-J. (dirs.) L’Aquitaine à la fin des temps glaciai-
res - Aquitaine at the end of the Ice Age. Les so-
ciétés de la transition du Paléolithique final au 
début du Mésolithique dans l’espace Nord aqui-
tain. PALEO, numéro spécial, 193-202. 

Neruda P. (2009) - Moravia during OIS 3: cultural rela-
tions. Praehistoria, 9-10, 125-147. 

Nerudova Z., Neruda (2017) - Technology of Moravian 
Early Szeletian leaf point shaping: A case study of 
refittings from Moravský Krumlov IV open-air site 
(Czech Republic). Quaternary International, 428, 
91-108. 

Pales L., Garcia M.A. (1981) - Atlas osteologique pour 
servir a l’identification des mammiferes du Quater-
naire. Éditions du Centre national de la recherche 
scientifique, Paris. 

Pales L., Lambert C. (1971) - Atlas osteologique pour 
servir a l’identification des mammiferes du quater-
naire. Éditions du Centre national de la recherche 
scientifique, Paris. 

Palma di Cesnola A. (1996) - Le Paléolithique inférieur 
et moyen en Italie. Millon ed. 

Peresani M. (1996) - The Levallois reduction strategy at 

the Cave of San Bernardino (Northern Italy). In 
Bietti A., Grimaldi S. (eds.) Reduction Processes 
(châines opératoires) in the European Mousterian. 
Proc. International Round Table, Quaternaria No-
va, VI, 205-236. 

Peresani M. (2000-2001) - Il Paleolitico Medio dei Colli 
Euganei (Veneto): stratigrafia e industria litica del 
sito di Monte Versa. Rivista di Scienze Preistori-
che, LI, 73-137. 

Peresani M. (2001) - An overview of the Middle Palaeo-
lithic settlement system in North-Eastern Italy. In 
N.J.Conard (ed.), Settlement Dynamics of the 
Middle Palaeolithic and Middle Stone Age. Tübin-
gen Publications in Prehistory, Introductory Volu-
me, Verlag, 485-506. 

Peresani, M. (Ed.) (2003) - Discoid Lithic Technology. 
Advances and Implications. British Archaeological 
Reports, International Series 1120. 

Peresani M. (2011) - The end of the Middle Palaeolithic 
in the Italian Alps. An overview on Neandertal land
-use, subsistence and technology. In: Conards N., 
Richter J. (eds.), Neanderthal lifeways, subsisten-
ce and technology. One Hundred Fifty Years of 
Neanderthal Study. Vertebrate Paleobiology and 
Paleoanthropology Series, Springer ed., 249-259. 

Peresani M. (2012) - Fifty thousand years of flint knap-
ping and tool shaping across the Mousterian and 
Uluzzian sequence of Fumane cave. Quaternary 
International, 247, 125-150. 

Peresani M., Bertola S., Delpiano D., Benazzi S., Ro-
mandini M. (2019) - The Uluzzian in the north of 
Italy. Insights around the new evidence at Riparo 
Broion Rockshelter. Archaeological and Anthropo-
logical Sciences, 11(7), 3503-3536. 

Peresani M., Centi Di Taranto L.E. (2013) - Blades, 
bladelets and flakes: a case of variability in tool 
design at the onset of the Middle - Upper Palaeoli-
thic transition in Italy. Comptes Rendus Palevol, 
12/4, 211-221. 

Peresani M., Chrzavzez J., Danti A., De March M., Du-
ches R., Gurioli F., Muratori S., Romandini M., 
Tagliacozzo A., Trombino L. (2011) - Fire-places, 
frequentations and the environmental setting of the 
final Mousterian at Grotta di Fumane: a report from 
the 2006-2008 research. Quartär, 58, 131-151. 

Peresani M., Cremaschi M., Ferraro F., Falguères Ch, 
Bahain J.J., Gruppioni G., Sibilia E., Quarta G., 
Calcagnile L., Dolo J.M. (2008) - Age of the final 
middle Palaeolithic and Uluzzian levels at Fumane 
cave, northern Italy, using 14C, ESR,234U/230Th 
and thermoluminescence methods. Journal of 
Archaeological Science, 35, 2986-2996. 

Peresani M., Cristiani E., Romandini M. (2016) - The 
Uluzzian technology of Grotta di Fumane and its 
implication for reconstructing cultural dynamics in 
the Middle - Upper Palaeolithic transition of We-
stern Eurasia. Journal of Human Evolution 91, 36-
56. 

Peresani M., Monegato G., Ravazzi C., Bertola S., Mar-
garitora D., Breda M., Fontana A., Fontana F., 
Jankovi� I., Karavanic I., Komšo D., Mozzi P., Pini 
R., Furlanetto G., Giovanni Maria De Amicis M., 

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 
27 



   

 

Perho� Z., Posth C., Ronchi L., Rossato S., Vu-
kosavljevi� N., Zerboni A. (2021) - Hunter-
gatherers across the Great Adriatic-Po Region 
during the Last Glacial Maximum: environmental 
and cultural dynamics. Quaternary International, 
581-582, 128-163. 

Peresani M., Potì A., Visentin D., Thun Hohenstein U. 
(2018) - At the western end. A discussion on the 
significance of a Mousterian assemblage with leaf 
points discovered in Northern Italy. Session XVI-1, 
The variability of bifacial tools in the Middle Pa-
laeolithic and their importance in the transition to 
the Upper Palaeolithic in Western Eurasia: raw 
material, techno-typology, chronology and interac-
tions in their environmental and cultural context. 
XVIII Congresso UISPP, Parigi, 26. 

Peresani M., Romandini M., Duches R., Jéquier C., 
Nannini N., Pastoors A., Picin A., Schmidt I., Va-
quero M., Weniger G.C. (2014) - New evidence for 
the Mousterian and Gravettian at Rio Secco Cave, 
Italy. Journal of Field Archaeology, 39/4, 401-416. 

Peretto, C. (1992) - Il Paleolitico medio, In A.Guidi M. (a 
cura di) Piperno: Italia Preistorica, Laterza, 170 - 
197. 

Pini R., Ravazzi C., Reimer P.J. (2010) - The vegetation 
and climate history of the last glacial cycle in a 
new pollen record from Lake Fimon (southern 
Alpine foreland, N-Italy). Quaternary Science Revi-
ews, 29, 3115-3137. 

Poli M.E., Burrato P., Galdini F., Zanferrari A. (2008) - 
Seismogenic sources responsible for destructive 
earthquakes in north-eastern Italy. Bollettino di 
Geofisica Teorica ed Applicata, 49, 301-313. 

Potts R.B., Shipman P. (1981) - Cut-marks made by 
stone tools on bones from Olduvai Gorge, Tanza-
nia. Nature, 291, 577-80. 

Richter J. (1997) - Sesselfelsgrotte III: Der G-Schichten-
Komplex der Sesselfelsgrotte - Zum Verständnis 
des Micoquien. Quartär-Bibliothek, Saarbrücken. 

Richter J. (2009) - The role of leaf points in the Late 
Middle Palaeolithic of Germany. Praehistoria, 10. 

Richter J. (2016) - Leave at the height of the party: A 
critical review of the Middle Paleolithic in Western 
Central Europe from its beginnings to its rapid 
decline. Quaternary International, 411, 107-128. 

Romandini M., Nannini N., Tagliacozzo A., Peresani M. 
(2014) - The ungulate assemblage from layer A9 
at Grotta di Fumane, Italy: a zooarchaeological 
contribution to the reconstruction of neanderthal 
ecology. Quaternary International, 337, 11-27. 

Romandini M., Terlato G., Nannini N., Tagliacozzo A., 
Benazzi S., Peresani M. (2018b) - Humans and 
Bears a Neanderthal tale. Reconstructing uncom-
mon behaviors from zooarchaeological evidence in 
southern Europe. Journal of Archaeological Scien-
ce, 90, 71-91. 

Romandini M., Thun Hohenstein U., Fiore I., Tagliacoz-
zo A., Perez A., Lubrano V., Terlato G., Peresani 
M. (2018a) - Late Neandertals and the exploitation 
of small mammals in Northern Italy: fortuity, neces-
sity or hunting variability? Quaternaire, 29(1), 61-
67. 

Romandini M., Crezzini J., Bortolini E., Boscato P., Bo-
schin F., Carrera L., Tagliacozzo A., Terlato G., 
Arrighi S., Badino F., Figus C., Lugli F., Marciani 
G., Oxilia G., Moroni A., Negrino F., Peresani M., 
Riel-Salvatore J., Ronchitelli A., Spinapolice E.E., 
Benazzi S. (2020) - Macromammal and bird as-
semblages of Late Middle to Upper Palaeolithic 
transition in Italy: an extended zooarchaeological 
review. Quaternary International, 551, 188-223. 

Sauro U. (1973) - Il Paesaggio degli alti Lessini: Studio 
geomorfologico. Mem. Fuori, serie 6, Museo di 
Storia Naturale di Verona, 160. 

Sauro U. (1978) - Forme strutturali e neotettoniche nei 
Monti Lessini. Gruppo Studio Quaternario Padano, 
4, 31-60. 

Sauro U. (2003) - Evidenze morfologiche di terremoti 
recenti nei Monti Lessini e nel Monte Baldo. Atti e 
Memorie della Accademia di Agricoltura Scienze e 
Lettere di Verona, vol. 175, a.a. 1998-1999, 89-
110. 

Sauro U. (2013) - Un golfo di pianura insinuato nel rilie-
vo prealpino. In: B. Avesani (ed.) - Valpantena - 
dal Vinum Raeticum all’Amarone. Cantina Sociale 
Valpantena, Gianni Bussinelli Editore, 4-17. 

Sauro U. (2020) - Tracce di glaciazioni antiche nei Mon-
ti Lessini. La Lessinia ieri oggi domani, 33-44. 

Sauro U., Ferrarese F. (2016) - Montagne in evoluzione 
e terremoti: alcuni casi nelle Prealpi Venete, Tren-
tine e Gardesane. De Rerum Natura, numero mo-
nografico speciale, Museo di Storia Naturale e 
Archeologia Montebelluna, 100, ISBN 978-88-
9724-104-1. 

Sauro U., Zampieri D. (2001) - Evidences of surface 
faulting and surface rupture in the Fore-Alps of 
Veneto and Trentino (NE Italy). Geomorphology, 
40, 169-184. 

Scali S., Mangiacotti M., Bonardi A. (2008) - Living on 
the edge: habitat selection of Hierophis viridiflavus. 
Acta Herpetologica, 3, 85-97. 

Schmid E. (1972) - Atlas of Animal Bones for Prehisto-
rians, Archaeologists and Quaternary Geologists. 
London, Elsevier. 

Shipman P. (1981) - Life History of a Fossil: An Intro-
duction to Taphonomy and Paleoecology, Harvard 
University Press, Cambridge. 

Sillero N., Oliveira M.A., Sousa P., Sousa F., Gonçalves
-Seco L. (2014) - Distributed database system of 
the New Atlas of Amphibians and Reptiles in Euro-
pe: the NA2RE project: Amphibia-Reptilia 35, 33-
39. 

Sindaco R., Doria G., Razzetti E., Bernini F. ( 2006) - 
Atlante degli Anfibi e dei Rettili d’Italia. Atlas of 
Italian amphibians and reptiles. Edizioni Polistam-
pa. Firenze: Societas Herpetologica Italica. 

Szyndlar Z. (1984) - Fossil snakes from Poland. Acta 
Zoologica Cracoviensia, 28, 1-156. 

Szyndlar Z. (1991) - A review of Neogene and Quater-
nary snakes of central and eastern Europe. Part I: 
Scolecophidia, Boidae. Colubrinae. Estudios Geo-
logicos, 47, 103-126. 

Terlato G., Livraghi A., Romandini M., Peresani M. 
(2019) - Large bovids on Neanderthal menu: ex-

 
 
28 

 
 

Peresani M. et al. 



   

 

ploitation of Bison priscus and Bos primigenius in 
North-eastern Italy. Journal of Archaeological 
Science, 25, 129-143. 

Terlato G., Lubrano V., Romandini M., Marín-Arroyo 
A.B., Benazzi S., Peresani M. (2021) - Late Nean-
derthal subsistence strategies at San Bernardino 
Cave (Berici Hills - northeastern Italy) inferred from 
zooarchaeological data. Alpine and Mediterranean 
Quaternary, 34(2), 1-23. 

Thun-Hohenstein U., Peretto C. (2005) - The exploita-
tion of the faunal remains in the Mousterian levels 
at Riparo Tagliente (Verona, Italy). In: Molines, N., 
Moncel M.H., Monnier J.L. (Eds), Données récents 
sur les modalités de peuplement et sur le cadre 
chronostratigraphique. Géologique et paléogéo-
graphique des industries du Paléolithique inferieur 
et moyen en Europe. B.A.R. International Series, 
Oxford, 261-268. 

Tostevin G.B., Skrdla P. (2006) - New Excavation at 
Bohunice and the Question of the uniqueness of 
the Type-Site for the Bohunician Industrial Type. 
Anthropologie XLV/1, 31-48. 

Uthmeier T. (2004) - Micoquien, Aurignacien und Gra-
vettien in Bayern: eine regionale Studie zum Über-
gang vom Mittel- zum Jungpaläolithikum (PhD 
Dissertation). Habelt, Bonn. 

Viganò A., Bressan G., Ranalli G., Martin S. (2008) - 
Focal mechanism inversion in the Giudicarie-
Lessini seismotectonic region (Southern Alps,  
Italy): insights on tectonic stress and strain. Tecto-
nophysics, 460, 106-115. 

Zampieri D. (1995) - Tertiary extension in the southern 
Trento Platform, Southern Alps, Italy. Tectonics, 
14, 645-657. 

Zampieri D. (2000) - Segmentation and linkage of the 
Lessini Mountains normal faults, Southern Alps, 
Italy. Tectonophysics, 319, 19-34. 

Zanferrari A., Bollettinari G., Carobene L., Carton A., 
Carulli G. B., Castaldini D., Cavallin A., Panizza 
M., Pellegrini G. B., Pianetti F., Sauro U. (1982) - 
Evoluzione neotettonica dell'Italia nord-orientale. 
Memorie Scienze Geologiche, 35, 355-376. 

Zorzi F., Pasa A. (1944-1945) - Il deposito quaternario 
di Villa di Quinzano. Bullettino di Paletnologia Ita-
liana, VIII (1), 1-52.  

Zilhão J. (2005) - Burial evidence for social differentia-
tion of age classes in the Early Upper Paleolithic, 
in: Vialou, J., Renault-Miskowsky, J., Patou Ma-
this, M. (Dir.), Comportements des hommes du 
Paléolithique moyen et supérieur en Europe: terri-
toires et milieux. ERAUL, Liège, 231-241. 

 
 
Foliate tools from a Middle Palaeolithic site in Italy 

 
 
29 

Ms. received: September 1, 2021 Revised: December 2, 2021 
Accepted: December 2, 2021 Avaiable online: February 1, 2022 



   

 

 
 
30