6
Documenta Praehistorica XLVIII (2021)
Population dynamics and human strategies
in Northwestern Patagonia>
a view from Salamanca Cave (Mendoza, Argentina)
Gustavo Neme1, Marcelo Zárate2, María Paz Pompei 3, Fernando Franchetti 1,
Adolfo Gil 1, Miguel Giardina1, Viviana P. Seitz 4, María Laura Salgán 5, Cinthia Abbona1,
and Fernando Fernández 6
1 Instituto de Evolución, Ecología Histórica y Ambiente (IDEVEA), Consejo Nacional de Investigaciones Científicas
y Técnicas (CONICET), Universidad Tecnológica Nacional (UTN), San Rafael, AR
gneme@mendoza-conicet.gob.ar< ferfranchetti@gmail.com< agil@mendoza-conicet.gob.ar<
mgiardina@mendoza-conicet.gob.ar< abbonacinthia@gmail.com
2 Instituto de Ciencias de la Tierra y Ambientales de La Pampa (INCITAP), CONICET, Universidad Nacional de
La Pampa, Santa Rosa, AR< marcelozarate55@yahoo.com.ar
3 Instituto Superior de Estudios Sociales (ISES), CONICET, Universidad Nacional de Tucumán, San Miguel de
Tucumán, AR< paz.pompei@gmail.com
4 Instituto Argentino de Investigaciones de Zonas Áridas (IADIZA), CONICET, Centro Científico Tecnológico
(CCT), Mendoza, AR< vseitz@mendoza-conicet.gob.ar
5 IDEVEA, CONICET, International Center for Earth Science (ICES) Nodo Malargüe, Universidad Nacional de
Cuyo, AR< lsalgan@mendoza-conicet.gob.ar
6 CONICET, Departamento de Química, Facultad de Ingeniería, Universidad de Buenos Aires, AR
fernandezf77@yahoo.com.ar
ABSTRACT – In this paper we evaluate the role of human strategies in the Andean Piedmont from
northern Patagonia across the Holocene. Specifically, we present the analysis of the Early Holocene-
Late Holocene archaeological record of Salamanca cave (Mendoza-Argentina). We identified techno-
logical changes that occurred during the Late Holocene and the implications of a human occupation
hiatus in the Middle Holocene. We follow a multiproxy approach by the analysis of radiocarbon dates,
archaeofaunal remains, ceramic, lithics and XRF obsidian sourcing. We also discuss a detailed stra-
tigraphic sequence by geomorphological descriptions, the construction of a radiocarbon sequence
model and summed probability distributions, compared with other archaeological sites in the region.
We conclude that after the Middle Holocene archaeological hiatus, human populations grew while
guanaco populations dropped. The imbalance between demography and resources boosted the incor-
poration of new technologies such as ceramics and the bow and arrow, allowing people to exploit
lower-ranked resources. 
KEY WORDS – hunter-gatherers; Northern Patagonia; mid Holocene; late Holocene; intensification;
Southern Mendoza
DOI> 10.4312\dp.48.15
IZVLE∞EK – V ≠lanku ocenjujemo vlogo ≠lovekovih strategij v andskem Piedmontu iz severne Pata-
gonije v obdobju holocena. Natan≠neje predstavljamo analizo zgodnje holocenskih do pozno holocen-
skih arheolo∏kih podatkov iz jame Salamanca (pri mestu Mendoza v Argentini). Prepoznali smo
tehnolo∏ke spremembe, ki so se zgodile v ≠asu poznega holocena, ter posledice hiatusa v poselitvi v
≠asu srednjega holocena. Z analizo radiokarbonskih datumov, arheozoolo∏kih ostankov, keramike,
Populacijske dinamike in ;lovekove strategije v severozahodni Patagoniji>
pogled iz jame Salamanca (Mendoza, Argentina)
Population dynamics and human strategies in Northwestern Patagonia> a view from Salamanca Cave (Mendoza, Argentina)
7
presence of different ecosystems with a diverse geo-
logical setting.
A recent systematic survey, which compared the
Highlands, Piedmont and Lowlands with the same
scale and units of analysis (in the Diamante fluvial
basin), proved that the Piedmont was by far the
most intensely exploited biogeographic environ-
ment, with a higher site density, discard rate, and
evidence of longer stays on sites (Franchetti 2019).
Under this scenario, archaeological information avai-
lable in this area also indicates a strong use of rock
shelters, with rich faunal and archaeobotanical as-
semblages, intense obsidian use and scant ceramic
presence, in comparison with northern and western
archaeological records (Durán 2000; Durán et al.
1999; Neme et al. 2011; Sugrañes, Franchetti 2012).
In the year 2007, the ‘Salamanca cave’archaeological
site was found as part of a cultural resource manage-
ment study in the ‘Puesto Rojas’ oil field (Malargüe-
Mendoza province, Argentina). The excavation made
it possible to identify a human occupation sequence
of more than 7000 14C years BP, with a dense cultu-
ral deposit. The study of Salamanca cave allows us
to add new information, improving the chronology
and regional sequence, as well as aspects of the die-
tary, technology and cultural adaptations. In this
paper we report the stratigraphy of the cave filling,
along with the radiocarbon chronology and its as-
sociated archaeological record. We discuss the re-
sults obtained through different lines of evidence
such as archaeofaunal remains, ceramic, lithic and
XRF obsidian sourcing, framed into a regional archa-
eological context, and a geoarchaeological perspec-
tive. 
Background
The archaeological record of Gruta del Indio and El
Chancho sites, located at the Piedmont (Fig. 1), indi-
Introduction
The archaeological record of arid environments pre-
sents specific challenges for preservation and site
formation. At the same time, it allows us to acquire
a better comprehension of human adaptations to
these ecosystems. In South America, the Patagonia
region became one of the last refuges of a hunter-
gather way of life in the world (Bettinger 2015; Bor-
rero 1994–1995; 2001; Morgan et al. 2017). In Men-
doza province, the Patagonia phytogeographic unit
extends across the Piedmont of the Andes Cordille-
ra, and widens southwards including the volcanic
district of Payunia (Roig 2000; Villagra et al. 2010;
Oyarzabal et al. 2018). Several archaeological sites
have been found and studied, leading to debates and
discussions in the Andes Piedmont from northwest-
ern Patagonia, Argentina (Gambier 1985; Durán
2000; Gil et al. 2005; Neme 2007; Neme et al. 2011;
Barberena et al. 2010). These discussions were re-
lated to the age of the first human occupations in the
region and their probable coexistence with megafau-
na (Forasiepi et al. 2010; Neme, Gil 2012; Barbere-
na et al. 2010), and also to the occurrence of a Mid-
dle Holocene (7000–4000 14C years BP) human oc-
cupation hiatus (Gil et al. 2005; Neme, Gil 2008a;
2009; Barberena et al. 2015a; Durán et al. 2016).
During the Late Holocene there was an intensifica-
tion process in northern Patagonia due to a demo-
graphic increase, which triggered important dietary,
technological and social changes (Neme 2007; Ne-
me, Gil 2008a). However, hunter-gatherer groups
kept inhabiting this area, functioning as a barrier to
the farmers settled to the north (Lagiglia 1968; Gil
et al. 2009; 2011).
Most of the archaeological sites excavated in the An-
des Piedmont of northwestern Patagonia are caves
and rock shelters, which record the adaptation of
small-scale groups to desert environments. The eco-
tonal nature of the study area, strongly modelled by
altitude, provided a wide set of resources due to the
kamnitih orodij in iskanja izvora obsidiana z metodo XRF smo sledili pristopu z ve≠ kriteriji. Raz-
pravljamo tudi o podrobnem stratigrafskem zaporedju s pomo≠jo geomorfolo∏kih opisov, pa tudi o
izgradnji modela radiokarbonske sekvence in o porazdelitvah povzete verjetnosti, vse to pa primer-
jamo ∏e z drugimi arheolo∏kimi najdi∏≠i v regiji. Zaklju≠ujemo, da se je ∏tevilo ljudi po obdobju sred-
njega holocena pove≠alo, medtem ko se je ∏tevilo gvanakov zmanj∏alo. Neravnovesje med demogra-
fijo in viri je spodbudilo vklju≠evanje novih tehnologij kot so lon≠arstvo ter lok in pu∏≠ice, kar je lju-
dem omogo≠ilo izkori∏≠anje manj pomembnih naravnih virov.
KLJU∞NE BESEDE – lovci-nabiralci; severna Patagonija; srednji holocen; pozni holocen; intenzivi-
ranje; ju∫na Mendoza
G. Neme, M. Zárate, M. P. Pompei, F. Franchetti, A. Gil, M. Giardina, V. P. Seitz, M. L. Salgán, C. Abbona, and F. Fernández
8
cates that the first human occupations in southern
Mendoza date to c. 11 000 14C years BP (Lagiglia
1968; Neme, Gil 2012). In the Andean Piedmont
from northwestern Patagonia, the earliest evidence
of human occupation comes from the Grande River
valley and dates back to c. 9000 14C years BP (Gam-
bier 1985; Durán 2000; Diéguez, Neme 2004; Ne-
me et al. 2011; Barberena et al. 2010). The first ex-
cavated site in the study area was El Manzano cave
(Gambier 1985), dated 8100 14C years BP, with a
c. 7000–2100 14C years BP archaeological hiatus
(Neme et al. 2011). In the same valley, other exca-
vated rock shelters dated a Late Holocene archaeo-
logical record (Durán 2000; 2002; Durán, Ferrari
1991; Durán et al. 1999; Campos et al. 2006). These
archaeological sites include Cañada de Cachi, Pue-
sto Carrasco, Cueva de Luna and Caverna de las Bru-
jas (Fig. 1). The most important changes identified
by researchers are the technological innovations, in-
cluding the incorporation of the bow and ceramics
that occurred c. 2000 14C years BP (Durán 2000).
Moreover, a group of cemeteries (Ojo de Agua, Baja-
da de las Tropas, and Las Chacras) were excavated
(Fig. 1) in the surroundings of Salamanca cave, dated
in the last c. 3000 years BP (Novellino, Neme 1999;
Gil et al. 2011; Salgán et al. 2012).
After the first arrival of human populations, towards
the Late Pleistocene-Early Ho-
locene (Gradín et al. 1984;
Durán 2002; Berón 2006; Gil
2006; Neme, Gil 2008a; Bar-
berena et al. 2010; 2015a;
Tripaldi et al. 2017), occupa-
tions increased between 9000
and 7000 years BP. However,
the majority of environmen-
tal eco-zones show a temporal
hiatus during almost the en-
tire Middle Holocene, particu-
larly between 7000 and 4000
years BP (Berón 2006; Gil
2006; Gil et al. 2005; Neme,
Gil 2009; 2010; Barberena et
al. 2010; 2015a).
The meaning of this Middle
Holocene hiatus is a topic of
strong debate. The most ac-
cepted hypothesis to explain
the absence of radiocarbon
dates is a drop in human po-
pulation as a consequence of
environmental deterioration.
An increase in the aridity affected the location and
effective population size, provoking aggregation of
human groups across the region (Gil et al. 2005; Ne-
me, Gil 2012). Volcanic activity is an alternative ex-
planation, and some archaeologists suggest that vol-
canic eruptions may have been of major magnitude
both spatially and temporarily (Durán, Mikkan
2009; Durán et al. 2016), and that such events could
affect the environmental availability of the land and
its productivity.
In contrast, other researchers question the existence
of the hiatus, assuming sampling problems or biases
(García 2005; Garvey, Bettinger 2018). Finally, ano-
ther hypothesis is that the lack of archaeological re-
cord is a consequence of anthropic behaviour, such
as the cleaning of the occupation floors or pit excava-
tion, which resulted in low visibility of the archaeo-
logical record (García et al. 1999). At a macro regio-
nal scale, a similar Holocene hiatus was also identi-
fied with a different magnitude in neighbouring
areas (north of Neuquén and central Chile) and other
South American regions (Barberena et al. 2015b;
2017; Méndez et al. 2014; Santoro et al. 2016; Ri-
ris, Arroyo-Kalin 2019).
Strong and increasing evidence of occupations oc-
curred after 4000 years BP in the northern Patago-
Fig. 1. Salamanca cave location, obsidian sources and the other archaeo-
logical sites mentioned in the text.
Population dynamics and human strategies in Northwestern Patagonia> a view from Salamanca Cave (Mendoza, Argentina)
9
nia region. During the last 2000 years BP those hu-
man occupations spread to all environments includ-
ing marginal settings, while the dietary breadth ex-
panded (Gil 2006; Neme 2007; Neme, Gil 2012). In
Patagonia’s northern boundary, this process led the
human population to incorporate domestic plants,
but always as a small part of the diet (Gil et al. 2006;
2011; 2020). While the Patagonian human popula-
tions continued with a hunter-gatherer way of life,
they adopted a more diverse technological set, more
distant social networks, and probably a diminishing
residential mobility (Gil et al. 2011; Neme, Gil 2008a;
2008b; 2012; Ugan et al. 2012).
The archaeological site
Salamanca cave (35°17’4.11”S/69°41’33.34”W) is
located in the Andean Piedmont at the rim of a mesa
formed by several erosional episodes from the Plio-
cene to the Present. Following the regional stratigra-
phic scheme (Nullo et al. 2005), the mesa is made
up of a basaltic lava flow (Pleistocene) that buries a
1–2 meter-thick conglomerate (Pleistocene) that un-
conformably overlies continental Cretaceous reddish
sandstones. (Fig. 2a). 
The mesa makes up the south valley wall of a minor
creek along c. 400m. The stream discharge comes
from a water spring situated c. 500m upstream of
the site (Fig. 2.c). The Salamanca cave is developed
in the basaltic lava that caps the mesa (Fig. 2.b). It
is a chamber-like void (c. 8m
deep, 1.5m high, 3m wide) of
fairly regular morphology
(spheroidal in shape), and the
entrance is an ample opening
(3.5m high, 2.5m wide) (Fig.
2.d). The cave walls are cover-
ed by ferric and manganese
oxide stains, and show minor
spalling. The chamber-like
morphology of the cave
points to a primary cavity of
the basalt formed while the
lava flowed on the aggrada-
tional land surface made up
of conglomerates. During the
widening of the valley margin
by lateral erosion the volcanic
chamber void was gradually
opened, and exposed to exo-
genous processes; eventually
it became an available locus
for human occupation.
During January 2010 we excavated a 2x1m unit (A-
1) at the west part of the cave, close to the wall
(Fig. 3). The excavation unit was subdivided into
two 1x1m subunits (north and south sector). We re-
moved 20 artificial levels, five centimetres each, fol-
lowing the natural slope of the surface floor. The
excavation reached the weathering bedrock of the
cave. We found several medium and small blocks
from the cave ceiling at the surface and also in the
sedimentary filling.
Chronostratigraphy and archaeological com-
ponents
The cave is filled with c. 1.3m of fine sand that is
the sedimentary matrix of the archaeological record;
a secondary fraction of blocks (main axis c. 10cm
long) is present, mainly at the lowermost section of
the sequence. Several bioturbation structures were
identified throughout the sequence (e.g., rodents
caves, roots alteration, insect activity) along with hu-
man structures (hearth and trash middens) which
became the major hindrance in the stratigraphic
analysis of the sedimentary filling. Hence, difficulties
were apparent when trying to correlate the artificial
excavation layers, the radiocarbon dates and the ar-
chaeological record.
The sedimentary filling is subdivided into five stra-
tigraphic units (1, 2, 3, 4, 5 from top to bottom) on
the basis of the occurrence of discontinuity surfaces
of sedimentation (Tab. 1). The NE angle of the exca-
Fig. 2. Landscape at Salamanca cave and the surrounding area: a gene-
ral view towards the north of the mesa; b Salamanca cave at the foot of
the basaltic lava flow; c water spring upstream of Salamanca cave; d en-
trance and the excavation unit.
G. Neme, M. Zárate, M. P. Pompei, F. Franchetti, A. Gil, M. Giardina, V. P. Seitz, M. L. Salgán, C. Abbona, and F. Fernández
10
vation is the less disturbed
and relatively complete sec-
tion (Fig. 4). Units were de-
scribed following a pedosedi-
mentary approach, a combin-
ed sedimentological (grain
size, sedimentary structure)
and soil description (bounda-
ries, Munsell soil colours).
The lower boundary of Unit 2
exhibits an irregular shape
(deep pocket), the upper part
of Unit 3 (erosional surface).
In turn, at the NW angle of the excavation the lower
boundary of Unit 3 is a well-defined erosional sur-
face carved in Unit 4. It is a c. 40cm deep and 50cm
wide depression (trash hole, see below) which lat-
erally grades into a depositional boundary (Fig. 4).
Twelve radiocarbon dates were obtained from char-
coal samples (Tab. 2). The uncalibrated median chro-
nology ranges from 7335 to 1055 14C years BP. Con-
sidering the stratigraphy and the radiocarbon dates,
we defined three archaeological components (youn-
ger to older: A, B, C; Tab. 3).
The younger component (A) includes the stratigra-
phic units 1, 2 and 3. Unit 3 intrudes the NW part of
the stratigraphy; the boundary shape and the filling
with a chaotic and high concentration of bones along
with lithic, charcoal and ceramic remains make it
possible to interpret it as a trash hole between 1055
and 1516 14C years BP (Fig. 4).
The middle component (B) includes the upper sec-
tion of stratigraphic unit 4 with radiocarbon dates
spanning from 1516 and 2200 14C years BP (Fig. 4).
The lower component (C) comprises the lower sec-
tion of unit 4. No sedimentary differences were ob-
served between components B and C. However, the
presence of much older radiocarbon dates (7100 and
7400 14C years BP), and differences in the archaeo-
logical record (projectile point shape, raw materials,
exploited taxa and absence of pottery), lead us to
differentiate components B and C.
In the case of component A, even when the radio-
carbon dates had ages from the last two millennia,
several modern historic remains are present (e.g.,
goat bones, etc.) in the uppermost 20cm. This indi-
cates that part of this component corresponds to
occupations of the late historic herders (Fig. 4). In
component C, a radiocarbon date yielded 1870±20
Fig. 3. Salamanca cave view on floor.
Tab. 1. Lithological features and archaeological remains of Salamanca cave stratigraphic units.
Unit Thickness Lithological features and archaeological remains Lower contact
Light brownish gray (10YR 6\2) fine sands< moderately compacted to
1 0–15cm loose,massive< well preserved volcanic ash lenses (1932 Quiza-Pú sharp, irregular
eruption), thinly laminated on the W wall profile of the excavation.
15–30cm Grayish brown (10YR 5\2) fine sands, massive< goat dung, extensive
2 maximum, average hearth (20cm long, 12cm thick) in the E wall profile. sharp, undulating
thickness 15 cm
3 30–50cm
Grayish brown (10YR 5\2) fine sands with coarse clasts< very frequent sharp, undulating<
archaeological remains (organic fragments and lithic artifacts). sharp, smooth
Grayish brown (10YR 5\2) fine sands with basaltic clasts< friable, sedi-
mentary beds dipping towards the central part of the cave< some rodent
4
(25 to 50cm–100cm)
caves. Abundant archaeological material including organic remains
sharp, smooth~70cm maximum
(e.g. vegetal macro-remains, vertebrate bones) lithic artifacts, as well as
thickness
charcoal fragments, several hearths< dark brown (10YR 3\3) silty sand
layer, 2cm thick including numerous mm long charcoal fragments.
Light brownish gray (10YR 6\2) very fine sand including coarse basaltic
5 (100–110cm) angular fragments< archaeologically sterile. It overlies a weathered sharp, irregular
basaltic surface.
Population dynamics and human strategies in Northwestern Patagonia> a view from Salamanca Cave (Mendoza, Argentina)
11
14C years BP. However, considering
the identified bioturbation processes,
and the resulting discontinuity of la-
yers throughout the sequence, it is
highly probable that the charcoal frag-
ment comes from younger layers.
We calibrated the 14C dates in OxCal
4.3 (Bronk Ramsey 2009) using the
SHCal13 curve, and modelled them as
a phased stratigraphic sequence (Tab.
4). These 14C dates are listed in stra-
tigraphic order from top (younger
ages) to bottom (older ages) (Fig. 5).
Cultural horizons were modelled as
phases assuming that they do not
overlap and deposited sequentially,
and therefore we relied on our strati-
graphic interpretations to organize
the dates accordingly. We modelled
the dates within components (A, B
and C) using the command phase and estimated its
duration with the command span. In addition, we
separated each phase with the command boundary,
assuming they were single events and that they oc-
curred between the different cultural occupations.
We used the command interval to estimate the du-
ration of the Middle Holocene hiatus. The model de-
monstrated an overall agreement index of 100%, in-
dicating a good fit for the data used (Fig. 5).
The archaeological record 
Zooarchaeological record
A total of 31 035 bone specimens were analysed, of
which 5928 were assigned to some taxonomic cate-
gory. The results show a wide range of species ex-
ploitation, dominated by the Camelidae, Chinchil-
lidae and Chlamyphoridae families (Tab. 5). Tapho-
nomic observation (weathering, fresh fractures, cut
marks, burnt surfaces) allowed us to discriminate
between the species exploited by humans from
those introduced by other predators. We confirm
that bone specimens of Didelphidae, Caviidae, Cte-
nomyidae, Cricetidae and Lestodelphys sp. as well
as suboscine birds were incorporated by the preda-
tory activity of the common owl (Tyto furcate).
Component A is the most diverse considering the ex-
ploited species (NTAXA=10), even when their sam-
ple size (NISP=431) is not much larger than that of
Fig. 4. Stratigraphic units in profile of north wall and the radio-
carbon dates.
Tab. 2. Radiocarbon dates from Salamanca cave. The calibration was performed by the OxCal program
with the southern hemisphere calibration curve (Hogg et al. 2013).
Sample Deep Context Stratigraphic Archaeological Material Lab code 14C age Calib. δ13C
ID (cm) unit component (yrs. BP) (‰)
CS-1 35 North profile 2 A| Charcoal AA-94003 1352±35 1284±19 –23.1
CS-2 25 North profile 2 A| Charcoal AA-94004 1516±38 1425±58 –23.2
CS-3 39 North profile
Trash top 3
A Charcoal AA-94005 1055±35 982±38 –22.9
3
CS-4 39 North profile A Charcoal AA-94006 1366±35 1297±16 –22.8
4
CS-5 50 North profile
Trash floor B
Charcoal AA-94007 1621±36 1497±54 –23.4
3
CS-6 60 East profile 4 B Charcoal AA-94002 1561±38 1461±47 –22.9
CS-8 75 Level 15 4 B Charcoal UGAMS-7006 2200±20 2234±60 –23.7
CS-9 65 Level 13 4 B Charcoal UGAMS-7007 1920±20 1867±23 –22.1
CS-10 95 Level 19 4 C Charcoal UGAMS-7004 7140±25 7973±17 –22.9 
CS-11 95-100 Level 19–20 4 C Charcoal UGAMS-7005 1870±20 1815±41 –23.8 
CS-7 100 East profile 4 C Charcoal AA-94001 7335±51 8135±67 –24.2
G. Neme, M. Zárate, M. P. Pompei, F. Franchetti, A. Gil, M. Giardina, V. P. Seitz, M. L. Salgán, C. Abbona, and F. Fernández
12
component B (NISP=405 and NTAXA=7). On the
other hand, component C is the less diverse (NTAXA
=5) but it must be mediated by the sample size (NISP
=66). With the exception of the domestic taxa (Ca-
prininae) registered and the smaller NTAXA, few dif-
ferences occur between components A and B (Fig. 6).
The most important changes take place between
components C and B, which reflect a more intensive
exploitation of Chlamyphoridae during the earlier
occupation (component C), and a dominance of Chin-
chillidae and Camelidae families in component B.
The latter trend is maintained in component A (Fig.
6). The mollusc fragments identified in component
A correspond to marine taxa, probably from central
Chile. This is a common scenario identified in the
southern Mendoza archaeological record, where we
observe molluscs used as tools, containers, and neck-
laces beads (Lagiglia 1977; Neme 2007).
Nine camelid bone samples were analysed to extract
ancient DNA. Two of them were not considered due
to the low coverage. The remaining seven samples
confirmed our identification as Lama guanicoe spe-
cimens (Abbona et al. 2020) in the three defined
Fig. 5. Chronological model for 11 14C dates from Salamanca cave. The grey lines show probability esti-
mates trimmed or shifted based on constraints of the Bayesian chronological model developed here. The
overall agreement index for this model is strong (Amodel:100).
Population dynamics and human strategies in Northwestern Patagonia> a view from Salamanca Cave (Mendoza, Argentina)
13
components (A, B and C). It is important to discard
the proposed hypothesis about the existence of do-
mestic camelid (Lama glama) in the southern Men-
doza region (Bárcena 1997). 
Lithic analysis
The lithic analysis includes 4987 artefacts of which
2331 (46%) belong to component A, 2608 (53%) to
component B and 48 (1%) to component C. Silica or
cryptocrystalline was the most used raw material at
the site (57%), followed by obsidian (39%) and then
basalt (3%). By silica we mean rocks with microcry-
stalline structure, composed primarily of quartz and
including chert, chalcedony, agate, and silcrete. How-
ever, their use over time shows changes – silica do-
minated component C (%), while obsidian was do-
minant in component A (%) (Fig. 7). The low-quali-
ty raw material (mostly basalt) decreased over time
(3% A, 4.5% B and 6.5% C). Debitage is the most
frequent category, with an increase of the discard
rate of tools by year during the last two millennia,
from 0.1 in the component C to 3 in the component
A (Tab. 6). Projectile points and scrapers are the best
represented tool category (Fig. 9). The projectile
points are the regional triangular types related to
Late Holocene human occupation, however in the
older component a stemmed one was found (Fig.
8.1). This last point type has no similarities with any
other projectile point in the regional context.
Another difference is the increase in the number and
diversity of tools over time. Component C includes
only one tool; component B encompasses 28 tools
within five tool classes; and component A, presents
49 tools within four tool classes. The differences are
not due to the sample size. On the contrary, compo-
Tab. 3. Archaeological components defined in the
paper.
Archaeological Excavation Stratigraphic Chronology
component units units (14C years BP)
A 1 to 9 1, 2 and 3 1055–1516
B 10 to 16 4 1561–2200
C 17 to 21 4 7100–7400
Date code – Commands Modeled BP
2σ cal range
Boundary End Salamanca cave 1043–415
Span Phase A 277–543
R_Date AA-94005 1050–809
R_Date AA-94003 1302–1119
R_Date AA-94006 1301–1182
R_Date AA-94004 1396–1299
Boundary 1460–1313
Span Phase B 608–903
R_Date AA-94002 1525–1360
R_Date AA-94007 1545–1381
R_Date UGAMS-7007 1879–1745
R_Date UGAMS-7006 2305–2064
Boundary 3552–2070
Interval Middle Holocene 1837– 5850
Boundary 8010–4529
Span Phase C 0–291
R_Date UGAMS-7004 7998–7854
R_Date AA-94001 8185–7972
Boundary Start Salamanca cave 10 233– 7952
Tab. 4. Modelled radiocarbon sequence for Sala-
manca cave.
Tab. 5. Faunal taxonomic list and abundance
(NISP) in Salamanca cave. *Rheidae eggshell frag-
ments in brackets; # Consumed species.
Archaeological components
A B C
Eudromia elegans # 0 1 0
Phalacrocorax # 1 0 0
Anatidae # 1 0 0
Suboscine 1 0 0
Rheidae # 7 (561)* 8 (180)* 0 (26)*
Rhea pennata # 0 0 1
Thylamys pallidior 92 83 7
Lestodelphys halli 1 11 0
Chlamyphoridae # 2 4 0
C. vellerosus # 0 5 6
C. villosus # 14 65 11
Z. pichiy # 12 7 205
Hystricomorpha indet. 252 265 32
Ctenomys sp. 23 53 6
Galea leucoblephara 3 0 1
cf Galea musteloides 1 0 0
Microcavia australis 9 16 1
Chinchillidae # 148 160 0
Lagidium sp # 0 0 1
Lagidium viscacia # 0 1 1
Cricetidae indet. 1468 175 152
Akodon iniscatus 9 16 9
Paynomys macronix 1 1 0
Phylotis xantopygus 71 90 1
Calomis musculinus 1 14 0
Eligmodontia sp. 161 254 23
Reitrodon auritus 40 50 3
Euneomys petersoni 17 45 2
Carnivora # 0 0 1
Lycalopex griseus # 2 0 0
Lycalopex sp. # 0 5 0
Puma concolor # 2 1 0
Puma cf P. yaguaroundi # 2 0 0
Artiodactila # 90 29 6
Camelidae # 5 7 0
Lama sp. # 99 110 15
Lama guanicoe # 43 49 3
Caprininae # 18 0 0
Mollusk 4 0 0
Total 2600 3099 487
G. Neme, M. Zárate, M. P. Pompei, F. Franchetti, A. Gil, M. Giardina, V. P. Seitz, M. L. Salgán, C. Abbona, and F. Fernández
14
nent B consists of more arte-
facts but less tools and catego-
ries, while component A has
less artefacts but more tools
and categories (Tab. 7). In ad-
dition, the artefact discard rate
increases over time (from one
to 46). In component B, there
is a higher debitage/instrument
index, which implies increased
transport of raw materials to
the site and a possible increase
in manufacturing. On another
hand, component A shows an
increased discard of tools (Tab.
7). Finally, no cores were re-
corded in the sequence, and
93% of the debitage has no cor-
tex reserve. 
XRF analysis
A total of 88 obsidian artefacts
were analysed by XRF follow-
ing the methodology of the Uni-
versity of Missouri Research Reactor Laboratory
(MURR) (Glascock, Ferguson 2012). The results
were compared using the database available in
MURR, including seven obsidian sources from north-
ern Patagonia (Fig. 9). We could identify four known
obsidian sources and two unknown ones (Tab. 8).
The most represented obsidian source is Coche Que-
mado (72%), located on the south bank of the Gran-
de river, 100km to the south (Salgán et al. 2020),
followed by Las Cargas source (11%), located c.
70km at the Argentina-Chile border in the Andes
(Salgán et al. 2015) (Fig. 1). The other obsidian
sources identified are El Peceño (3%), located 100km
away, on the plains (Salgán, Pompei 2017), and La-
guna del Maule (3%), 100km away on the Argenti-
nean-Chilean border in the Andes (Barberena et al.
2019). The unknown source represents 10% of the
samples.
The Coche Quemado and Las Cargas sources are pre-
sent in the three components, in categories asuch as
debitage, bifacial artefacts, and artefacts with com-
plementary traces. There was no evidence of cores
or unifacial artefacts (Tab. 8). In component B, un-
known Group A (Giesso et al. 2011) as well as unas-
signed artefacts are added, both present as debi-
tage. In component A, the evi-
dence of sources located at di-
stances greater than 100km
(e.g., El Peceño and Laguna
del Maule) is recorded as bi-
facial artefacts and debris.
Tools are always projectile
points, as reported in la Payu-
nia and the Atuel river basin
(Salgán et al. 2012; 2014;
Pompei 2019).
Ceramic analysis
The ceramic record (Tab. 9)
includes 66 sherds from com-
ponent A (N=62) and B (N=4).
The high frequency of sherds
Fig. 6. Proportion of each consumed animal taxa in the Salamanca cave
components
Fig. 7. Raw material by component in Salamanca cave (BAS basalt; OBS
obsidian; RHY rhyolite; SIL silica).
Population dynamics and human strategies in Northwestern Patagonia> a view from Salamanca Cave (Mendoza, Argentina)
15
assigned to the Chilean typology highlights the pre-
sence of PAT (Early Pottery Period) and ‘Marrón Pu-
lido’ (Brown polished) types (Figs. 10 and 11). Those
Chilean traditions are well dated in Chile between
1800 and 800 years BP (Falabella et al. 2016). PAT
style is highly polished black ceramics with utilita-
rian forms, with conspicuously reduced firing, and
fine-grained temper size. Brown polished style is a
utilitarian ceramic with brown colours, sometimes
with incised decoration, and commonly with a pol-
ished surface treatment. Temper size is well select-
ed, consisting of fine grains of sands. The red pol-
ished style has similar characteristics to the brown
polished style, the only differences is the use of co-
lour. Overo style is a utilitarian ceramic with brown
and grey colours, sometimes with incised decoration
in the neck section, smoothed surface treatment, oxi-
dized and oxidized incomplete firing. Temper size is
poorly selected, medium to large and generally con-
sists of sand.
Only a small part of the ceramic record (16%) be-
longs to the local ‘Overo’ typology (Lagiglia 1997),
suggesting that ceramics were mainly acquired by
extra-local exchange systems, in agreement with
other sites of the region (Sugrañes 2019). The small
size of the ceramic component reinforces this idea,
and is coherent with the mobility constraints on
north Patagonia hunter-gathers way of life. The ce-
ramic record presents thin sherds (average 4.9mm),
predominantly polished surface treatment (79%),
fine temper size (63%) and reduce firing (40%).
We consider some macroscopic characteristics in the
ceramics as a proxy of labour investment. The use
of fine temper size impacts the wall thickness and
therefore implies more labour and skill by the pot-
ter to produce a piece (Eerkens 2008). Reduced fir-
ing involves more investment, as the process needs
more supervision and precision to acquire the ex-
pected effect. A polished surface treatment involves
more labour and time once the piece has dried.
Discussion
The Salamanca cave archaeological record suggests
a long human occupation sequence at the Andean
Piedmont, starting 7335 14C years BP, with a noto-
Tab. 6. Artefact classes by raw material in each component. References, other raw materials: andesite,
quartz, vulcanite and indeterminate raw material. 
Component Raw material Debitage
Bifacial Unifacial Non formal
Other art.
Total
tool tool tool classes
Cryptocristalline 96.2% (628) 1.4% (9) 2.1% (14) 0.3% (2) 653
Obsidian 95.3% (688) 2.5% (18) 0.7% (5) 1.4% (10) 0.1% (1) 722
A Basalt 84.8% (28) 15.2% (5) 33
Rhyolite 100% (6) 6
Others RM 100% (5) 5
Subtotal A 95.14% (1350) 1.90% (27) 1.33% (19) 0.85% (12) 0.77% (11) 1419
Cryptocristalline 98.3% (1173) 0.5% (6) 0.7% (9) 0.3% (3) 0.2% (2) 1193
Obsidian 98.3% (525) 1.5% (8) 0.2% (1) 534
B Basalt 89.9% (62) 1.4% (1) 8.7% (6) 69
Rhyolite 100% (7) 7
Others RM 50% (3) 50% (3) 6
Subtotal B 97.84% (1770) 0.83% (15) 0.50% (9) 0.22% (4) 0.61% (11) 1809
Cryptocrystalline 94.7% (18) 5.3% (1) 19
C Obsidian 100% (10) 10
Basalt 100% (2) 2
Subtotal C 96.77% (30) 3.23% (1) 31
Total (n) 3150 43 28 16 22 3259
Fig. 8. Projectile points in Salamanca cave: 1 com-
ponent C; 2 component B; 3, 4 component A.
G. Neme, M. Zárate, M. P. Pompei, F. Franchetti, A. Gil, M. Giardina, V. P. Seitz, M. L. Salgán, C. Abbona, and F. Fernández
16
rious Middle to Late Holocene hiatus from 7140 to
2200 14C years BP. The sedimentary analysis and
the numerical ages indicate major, important post-
depositional processes affecting the archaeological
record. We defined three archaeological components
to identify changes or continuities in the archaeolo-
gical record. The earliest component (C), radiocar-
bon dated between 7335 and 7140 years BP, is cha-
racterized by a low archaeological depositional rate
(0.1lithic/year), with a narrow set of subsistence
activities that involved hunting, cooking, tool mak-
ing and maintenance. This evidence suggests that
the cave was occupied by small hunter-gatherer
bands during short stays. Chlamyphoridae (armadil-
los) and Camelidae (guanaco) were the most impor-
tant faunal resources exploited, followed by some
few Rheidae and Chinchillidae specimens. Local raw
materials represent c. 70% of the lithic record in
component C, coming from a less than 20km ra-
dius. Obsidian represents less than 30% of the total
used raw material. The sourcing analysis indicates
that it was obtained from sources located 70–100km
away (Tab. 8 and Fig. 1).
The radiocarbon dates discontinuity of more than
5000 years (from 7973 to 2234 cal years BP) is si-
milar to the gap identified in El Manzano cave (from
7907 and 2105 cal years BP; Neme et al. 2011) and
Huenul cave (Barberena et al. 2015a; 2015b) lo-
cated 70km and 180km away, respectively, in the
Andean Piedmont. A sum probability curve (Fig. 12)
was done using the radiocarbon calibrated dates
from Salamanca cave (Tab. 2) and from the two
other cave sequences (Neme et al. 2011; Barberena
et al. 2015b). It shows a bimodal curve with Early
and Late Holocene occupations. At the same time, it
defines a significant occupational gap, including most
of the Middle Holocene and the Early-Late Holocene
from 7400 to 2400 cal years BP, suggesting the exis-
tence of a very similar regional pattern.
In the case of Salamanca cave, this discontinuity ap-
pears to be chronological and stratigraphical (Veth
et al. 2017). The obtained radiocarbon dates indi-
cate that there may be a stratigraphic gap with no
sedimentation between 2200 and 7140 years BP
(see Fig. 4). This absence of sedimentation makes it
difficult to make definitive assumptions related to
the Middle Holocene gap regionally considered.
Different explanations have been proposed to un-
derstand this regional gap. Plausible explanations
include a regional depopulation, settlement reorga-
nization, changes in mobility, or in site size (Neme,
Gil 2009), although aridization is accepted as the
most plausible (Gil et al. 2005; Neme, Gil 2008a;
Barberena et al. 2017). Other researchers discuss
the significance of local volca-
nism and glacier events with
regard to the regional pro-
ductivity (Durán, Mikkan
2009; Durán et al. 2016).
However, the human res-
ponse to drier climatic condi-
tions remains undetermined
and debatable. Based on ob-
sidian hydration data, mito-
chondrial DNA and radiocar-
bon trends, other authors sug-
gest a continuity in human
occupation throughout north-
ern Patagonia (Garvey, Bet-
tinger 2018; Gordón et al.
2019; Pérez et al. 2016a).
Nonetheless, the Patagonian
human demography seemed
to be non-homogeneous, with
non-inhabited areas (Pérez et
al. 2016b).
Tab. 7. Lithic discard rates in Salamanca cave.
Fig. 9. XRF obsidian source analysis from Salamanca cave site artefacts.
Ellipses are the 90% confidence interval around each source.
Component A B C
Debitage\tools 29.3 73.8 30.0
Temporal span (years) 460 640 200
Artifact discard rate\years 3.1 2.8 0.2
Population dynamics and human strategies in Northwestern Patagonia> a view from Salamanca Cave (Mendoza, Argentina)
17
The analysis of Neuquén province (northern Patago-
nia) fauna shows an increase in guanaco depen-
dence during the Middle Holocene (Rindel 2017),
similar to the Middle Holocene pattern of southern
Mendoza (Neme, Gil 2009). The guanaco demogra-
phic curve, reconstructed using mitochondrial DNA
from the entire Patagonia, indicate a population in-
crease during the beginning of the Middle Holocene
c. 7500 years BP (Pérez et al. 2017; Moscardi et al.
2020; Abbona et al. 2021), which could explain the
human focus on this resource. Guanacos are adapt-
ed to cold and grassland environments (Puig et al.
2011), and changes in the environmental conditions
that favour guanaco populations can reduce plant
productivity in environments outside the Andes, as
well as smaller animal populations, leading humans
to focus on this target.
The suggestive contemporaneity in the Middle Holo-
cene radiocarbon hiatus among all the Andean Pied-
mont from northern Patagonia Piedmont sites (El
Manzano cave, Huenul cave and Salamanca cave),
needs to be explored and explained. More important-
ly, this trend is also identified at a larger regional
scale in other archaeological sites with a noticeable
decrease in the archaeological record (Neme, Gil
2009; 2012). Even if we consider the obsidian hy-
dration from the area in El Manzano cave (Garvey,
Bettinger 2018), the number of obsidian samples as-
signed to the Middle Holocene is much smaller than
those compared with the Early and Late Holocene
from the same site. This suggests a human depopu-
lation process or changes in the settlement pattern
related to a rearrangement to the new environmen-
tal conditions.
After this long discontinuity in the Salamanca cave
archaeological record, the human occupations rise
again during component B with an increase in the
faunal and lithic diversity as well as in the deposi-
tion rate (2.8 lithic artefacts discarded by year), and
probably more reoccupation events. In the lithic re-
cord the use of local raw materials in older compo-
nents is also recorded. It is associated with a Late
Holocene increase in the use of obsidian and an in-
crease in the number of obsidian sources exploited.
In El Manzano cave, the Laguna del Maule source is
recorded in the oldest components (Barberena et
al. 2019), while in the Late Holocene there is a sig-
nificant increase in obsidian over local resources, in-
corporating the Las Cargas obsidian source (Neme
et al. 2011). In Huenul cave, the infrequent use of
Laguna del Maule subtypes 1 and 2 is incorporated
in the Late Holocene, together with a Huenul local
source (Fernández et al. 2017).
Two technological innovations are evident in com-
ponent B: ceramics and the bow and arrow, as sug-
gested by the projectile point size (Martínez 2003;
Ratto 2003; Banegas et al. 2014; Castro 2015). The
projectile points change from stemmed to triangu-
lar notched types (Fig. 9). A more diverse use of raw
material is apparent, with silica increasing and ba-
Tab. 8. Artefact classes present in the obsidian components analysed from XRF Salamanca cave.
Components Artifact classes
Coche
Las Cargas El Peceño 1
Laguna del
Unassigned
Unknown
Quemado Maule-2 group A
Debitage 39% (11) 11% (3) 4% (1) 11% (3)
A
Bifacial tool 25% (7) 4% (1)
Non formal tool 7% (2)
Total 71% (20) 11% (3) 7% (2) 11% (3)
Debitage 57% (16) 4% (1) 4% (1)
B Bifacial tool 32% (9) 4% (1)
Total 89% (25) 4% (1) 4% (1) 4% (1)
Debitage 70% (7) 10% (1)
C
Bifacial tool 10% (1)
Non formal tool 10% (1)
Total 80% (8) 20% (2)
Fig. 10. Percentages of ceramic types in Salamanca
cave site.
G. Neme, M. Zárate, M. P. Pompei, F. Franchetti, A. Gil, M. Giardina, V. P. Seitz, M. L. Salgán, C. Abbona, and F. Fernández
18
salt decreasing. Four ceramic
sherds were recovered in this
component. But even when
the chronology is coherent
with the regional arrival of ce-
ramics dated at c. 2000 years
BP (Marsh 2017; Sugrañes
2017), these sherds, which
share the same characteris-
tics, are more likely the result
of vertical migration by some
post-depositional disturbance.
The exploited faunal resources increase from NTAXA
=5 in component C to NTAXA=7 in component B.
There is an increase in chinchilla hunting, a smaller
species, uncommon in the regional zooarchaeologi-
cal record. Framed within the diet breadth model,
chinchilla is lower ranked among the rest of species
in the region (Corbat et al. 2021). This taxon re-
places the Chlamyphoridae as the most abundant
small mammal on the site. This increase in the num-
ber of consumed taxa and the incorporation of small-
er animals agrees with the regional northern Pata-
gonia trends (Neme 2007; Otaola et al. 2015; Rin-
del 2017). However, the guanaco continues as the
main faunal resource at Salamanca cave and in the
other sites of the region.
Component A is the most diverse in terms of fauna
exploitation, lithic and ceramic types, with the high-
est depositional rate of the sequence (3 lithic arte-
facts discarded by year) and a decrease in the waste
discard rate in relation to the tools. This implies a
change in the mode of transport of raw materials,
from preform in component B to the transport of
tools in component A. This evidence is the result of
a higher reoccupation pattern and a larger number
of activities carried on. The faunal record shows a
strong dependence on guanaco and chinchilla, but
several carnivores, birds and armadillos could be
part of the diet too, as the taphonomic observations
suggest. Only the small rodents (Cricetidae and Sig-
modontinae) were clearly introduced to the site by
avian predators (mainly owls).
Components A and B are difficult to discriminate
due to the stratigraphic complexity, the absence of
a clear unconformity, and the presence of crossing
radiocarbon dates. Alternatively, both units (A and
B) might be a single component instead of two. How-
ever, on the basis of the presence of older radiocar-
bon dates, and the absence of historic remains in
component B as well as other archaeological aspects
(e.g., zooarchaeology, proportion of lithics raw ma-
terials) support the identifica-
tion of two components.
Obsidian trace element ana-
lysis indicates an increase in
the use of raw material sour-
ces during the Late Holocene.
Component C has the pres-
ence of Coche Quemado and
Las Cargas sources, while
components A and B incor-
porate at least four obsidian
sources in low frequencies.
Specifically, in component A,
obsidian artefacts from La-
guna del Maule 2 are record-
ed. This source was located in
the Barrancas and Colorado
rivers at a distance between
100 and 160km (Barberena
Fig. 11. Ceramic sherds from Salamanca cave. Types: PAT, Early Ceramic
Period (left), Brown Polished (centre), and Overo (right).
Tab. 9. Ceramic sherd properties from the Salamanca cave site. Referen-
ces: F fine; M medium; P polished; S smoothed; PAT Early Ceramic Pe-
riod; OV Overo; MP brown polished; RP red polished; O oxidized; OI oxi-
dized incomplete; R reduced; I incised; P painted, in section S sherd, R rim.
Variable Style
PAT MP OV RP
Thickness
4.8 5.6 5.8 4.3average mm
Residues
3 1 1 0
Count % Count % Count % Count %
Temper F 18 64.3 8 72.7 4 50 0 0
size M 10 35.7 3 27.3 4 50 1 100
Surface S 2 6.3 2 14.3 7 77.8 0 0
treatment P 30 93.8 12 85.7 2 22.2 1 100
Section
S 27 84.4 11 78.6 9 100 1 100
R 5 15.6 3 21.4 0 0 0 0
Decoration
I 9 90 0 0 0 0 0 0
P 1 10 0 0 0 0 0 0
O 2 6.3 1 7.1 1 11.1 1 100
Firing OI 13 40.6 7 50 4 44.4 0 0
R 17 53.1 6 42.9 4 44.4 0 0
Population dynamics and human strategies in Northwestern Patagonia> a view from Salamanca Cave (Mendoza, Argentina)
19
et al. 2019). The presence of debitage from this
source could indicate a home range increase, or
greater human interaction with populations located
to the south.
The ceramic sherds appear from level one to level
twelve, radiocarbon dated in 1530 years BP. The
presence of ceramics in the latter components could
reinforce the idea of a higher number of subsistence
activities, reflected in the mentioned number and
diversity of both the faunal and lithic records. In ad-
dition, the ceramics recovered show high manufac-
ture inversion, as reflected in the average thickness
(4.9mm), surface treatment (79% polished), temper
size (63% fine) and firing (40% reduced). Following
Sturm et al. (2016) this conjunction of traits on the
ceramic assemblage suggests a high reuse of vessels,
probably for several years. The high proportion of
non-local ceramic (mainly from the Chilean side of
the Andes) indicates a high value of this technology
in the area, not only for subsistence but also for
trade networks. These networks could have worked
like the Hxaro system (Weisner 2020), in which the
interchange of gifts assured the access to comple-
mentary environments when conditions of rainfall
and resources got harsh (Neme, Gil 2005; Eerkens
2011). El Manzano cave (90km to the south) regis-
ters a highly diverse ceramic assemblage made up
of non-local types, suggesting that the northwestern
Patagonia ceramic assemblages were obtained most-
ly from neighbouring societies. The sites located in
the Andean Piedmont from northwestern Patagonia
Fig. 12. Permutation tests showing variation between regional population growth. Grey areas represent
the confidence envelope for the null model, red areas represent where the empirical SPD significantly
deviates from the null model. The black line represents the empirical SPD for each archaeological site
(Crema, Bevan 2021).
G. Neme, M. Zárate, M. P. Pompei, F. Franchetti, A. Gil, M. Giardina, V. P. Seitz, M. L. Salgán, C. Abbona, and F. Fernández
20
are among the more diverse in faunal taxa, and pro-
bably plant resources too (Neme et al. 2011; Llano
2011; Llano, Barberena 2013; Otaola et al. 2015).
The Late Holocene archaeological record in Sala-
manca cave (components A and B) suggests a major
change after the Middle Holocene gap. The described
species faunal consumption in component C indi-
cates a narrow dietary focus on guanaco and arma-
dillos. These two species are placed at the top of the
regional faunal ranking (Corbat et al. 2021), and
represent more than 90% of the total NISP in com-
ponent C, while in later components A and B, those
species represent only 50% and 40%, respectively.
This change shows a broader diet later, and an in-
crease in the consumption of lower return rate spe-
cies such as chinchilla, birds and probably carnivo-
res too. This change in favour of a broader diet in
later components is in agreement with the intensifi-
cation process proposed for the region for the last
2000 years BP (Neme 2007; Neme, Gil 2008a). This
intensification process was criticized by latter re-
search, arguing that it could be an effect of the spa-
tial scale considered, instead of the diminution in
guanaco population by human hunting pressure
(Otaola et al. 2015; Wolverton et al. 2015). How-
ever, the faunal changes that took place from the
same sequence in Salamanca cave are in agreement
with the intensification process. A recent ancient
DNA guanaco study (including some samples from
Salamanca cave sequence) also demonstrate the exi-
stence of a decline in guanaco population in north-
western Patagonia during the las 2500 years BP
(Abbona et al. 2021), reinforcing the proposed re-
gional intensification process.
The implications of these intensification processes
move beyond the faunal, and are in line with other
changes identified in Salamanca cave. Among them
are the increase in obsidian use, the presence of Pa-
cific mollusc shells, and the introduction of Chilean
pottery during the last 2200 years. Those records
show the inclusion of more diverse and distant sites
of acquisition, including new obsidian sources, mol-
luscs, and Chilean ceramics, suggesting the establish-
ment of a stronger regional network, which allowed
regional hunter-gatherers access to non-locally avai-
lable resources. The increase in utilitarian pottery,
especially in component A, indicates new resources
being processed (including plants not yet analysed
in Salamanca). The human occupations are more re-
dundant and possibly longer than those from ear-
lier times, which is concordant with the increase in
the discard rate mentioned in this paper.
The presence of shelters, water, raw materials, and
the biological diversity could favour human occu-
pation in places like Salamanca cave, El Manzano
cave, Cueva de Luna, El Chacay, Cañada de Cachi or
Puesto Carrasco. There is an important plant record
in the sequence, some burnt, but not analysed yet.
Perhaps, the vegetal resources at the site can ex-
plain the occurrence of ceramic sherds, with the ce-
ramics used to process, transport or store seeds.
Final remarks
The Andean Piedmont in northwestern Patagonia
has diverse and rich resources. Located between
the Andean foothills and the Lowlands, it allows
access to wide and varied faunal, plants and lithic
resources. The predominance of Patagonia phyto-
geography province ensures an abundant guanaco
presence, as well as armadillos, and Rheidae, the
three top ranking animals of the region. The charac-
teristics of the Salamanca archaeological record
agree with most of the Piedmont excavated caves,
which shows a wide diversity in their faunal, lithic
and ceramic record, and intense signal of human oc-
cupations (Gambier 1985; Durán et al. 1999; Du-
rán 2000; Neme et al. 2011; Barberena 2015). The
chronological sequence follows a common pattern
among the caves in the region. After a first increase
of human occupation, during the colonization c.
7300 years BP, the archaeological record disconti-
nues until c. 2200 years BP.
In Salamanca cave, the observations made on the
stratigraphy shows different post-depositional pro-
cesses that altered the sediments and disposition of
archaeological materials. Human groups carried out
a diverse set of activities on the site, including those
related to food (animal and plant) acquisition and
processing, as well as lithic tool preparation and
maintenance. The ceramic diversity indicates that
cooking, transport and storage were part of the sub-
sistence activities.
The fauna analysis shows a strong exploitation of
middle size mammals at the beginning of the se-
quence, especially armadillos, which changes in
favour of chinchilla towards the last two millennia.
Beyond this trend, as in many other northwestern
Patagonian environments, camelids were the base of
human subsistence.
The lithic assemblage shows an increasing use of
non-local raw materials, consistent with the obser-
vations in other sites from the region. The artefac-
Population dynamics and human strategies in Northwestern Patagonia> a view from Salamanca Cave (Mendoza, Argentina)
21
tual categories suggest the presence of the last sta-
ges of the operative sequence, formalisation and re-
activation of tools. The few ceramic sherds and their
attribution to styles from the western side of the
Andes reinforce the idea of the high investment and
low production scale for this type of technology, in-
corporated mainly from exchange.
Finally, the archaeological information from Sala-
manca cave indicates that the site was occupied by
hunter-gatherer groups which colonized and occu-
pied the Andean Piedmont environment, facing en-
vironmental constraints, especially during the Mid-
dle Holocene. The lack of a Middle to Late Holocene
sedimentary record is a hindrance to interpreting
cultural discontinuities (Veth et al. 2017). Salaman-
ca cave also provides insights on the important chal-
lenge related to the resource structure and demogra-
phic dependent factors, and especially the intensi-
fication process, which forced changes in resource
use, technology and human relations with neigh-
bouring populations, increasing long distance ex-
change. Future excavations, as well as the use of
new methodology, will allow us to confirm and re-
ject some of the ideas presented here, and improve
our comprehension of the complex site formation
processes.
This work was supported by the Agencia Nacional
de Promoción Científica y Técnica (Republic of
Argentina): Grant numbers PICT-2013-0881 and
PICT-2016-2667; and CONICET by Grant PIP-0342.
To Jorge Zagal who kindly allowed us to work in his
lads. To the Municipalidad de Malargüe by their
housing support. To Cerutti and Neme families for
their help and support during the field work. To the
reviewers for their significant comments and cor-
rections that helped us to improve the manuscript.
ACKNOWLEDGEMENTS
Abbona C., Neme G., Johnson J., Kim T., Gil A., and Wol-
verton S. 2020. Were domestic camelids present on the
prehispanic South American agricultural frontier? An an-
cient DNA study. PLoSONE 15(11): e0240474.
https://doi.org/10.1371/journal.pone.0240474
Abbona C., Neme G., Johnson J., +4 authors, and Wolver-
ton S. 2021. Sustainable harvest or resource depression?
Using ancient DNA to study the population dynamics of
guanaco in western Argentina during the Holocene. Jour-
nal of Archaeological Science 129: 105355.
https://doi.org/10.1016/j.jas.2021.105355. 
Banegas A., Gómez Otero J., Goye S., and Ratto N. 2014.
Cabezales líticos del Holoceno tardío en Patagonia Meri-
dional: diseños y asignación funcional. Magallania 42(2):
155–174.
Barberena R. 2015. Cueva Huenul 1 archaeological site
(northwestern Patagonia, Argentina): initial colonization
and Middle Holocene demographic retraction. Latin Ame-
rican Antiquity 26: 304–318.
Barberena R., Pompei M. P., Otaola C., +4 authors, and
Hoguin R. 2010. Pleistocene-Holocene transition in north-
ern Patagonia: Evidence from Huenul Cave (Neuquén, Ar-
gentina). Current Research in the Pleistocene 27(1): 5–8.
Barberena R., Borrazzo K., Rughini A., + 13 authors, and
Quiroga M. 2015a. Perspectivas arqueológicas para Pata-
gonia septentrional: sitio Cueva Huenul 1 (provincia del
Neuquén, Argentina). Magallania 43(1): 137–163.
Barberena R., Prates L., and de Porras M. E. 2015b. The
human occupation of northwestern Patagonia (Argentina):
paleoecological and chronological trends. Quaternary In-
ternational 356: 111–126.
Barberena R., Méndez C., and de Porras M. E. 2017. Zoom-
ing out from archaeological discontinuities: The meaning
of Middle-Holocene temporal troughs in South American
deserts. Journal of Antropological Archaeology 46: 68–81.
Bárcena R. 1997. Malargüe Historia y Perspectivas. In P.
Lacoste (ed.), Prehistoria. UNO. Mendoza: 6–15.
Barberena R., Fernández M. V., Rughini A., + 10 authors,
and Glascock M. D. 2019 Deconstructing a complex obsi-
dian ‘source-scape’: a geoarchaeological and geochemical
approach in northwestern Patagonia. Geoarchaeology 34:
30–41. https://doi.org/10.1002/gea.21701
Berón M. 2006. Dinámica poblacional y estrategias de
subsistencia de poblaciones prehispánicas de la cuenca
Atuel- Salado- Chadileuvú- Curacó, Provincia de La Pam-
pa. Revista Arqueología 12: 347–352. 
Bettinger R. L. 2015. Orderly Anarchy: Sociopolitical
Evolution in Aboriginal California. University of Califor-
nia Press. Berkeley.
References
∴
G. Neme, M. Zárate, M. P. Pompei, F. Franchetti, A. Gil, M. Giardina, V. P. Seitz, M. L. Salgán, C. Abbona, and F. Fernández
22
Crema, E., Bevan, A. 2021. Inference from Large Sets of
Radiocarbon Dates: Software and Methods. Radiocarbon
63(1): 23–39. https://doi.org/10.1017/RDC.2020.95
Borrero L. A. 1994–1995. Arqueología de la Patagonia.
Palimpsesto 4: 9–69.
2001. El poblamiento de la Patagonia. Emecé. Buenos
Aires.
Bronk Ramsey C. 2009. Bayesian analysis of radiocarbon
dates. Radiocarbon 51(1): 337–360.
https://doi.org/10.1017/S0033822200033865 
Campos G., Cortegoso V., and Mazzanti C. 2006. Estrate-
gias tecnológicas y cambios en el uso de los recursos líti-
cos. Caverna de las Brujas, Malargüe, Mendoza. Anales
de Arqueología y Etnología 61: 33–52.
Castro S. 2015. Explotación de recursos líticos durante
el Holoceno en el límite de vertientes cordilleranas.
Valle del río de Las Taguas, San Juan, Argentina. Un-
published PhD thesis. Universidad Nacional de Córdoba.
Córdoba. 
Corbat M., Gil A., Neme G., and Bettinger R. 2021. Subsi-
stencia humana el Centro Occidente de Argentina: impli-
cancias del ranking de recursos. Latin American Anti-
quity. In press.
Diéguez S., Neme G. 2004. Geochronology of the archae-
ological site Arroyo Malo 3 and the first human occupa-
tions in the North Patagonia Early Holocene. In L. Miotti,
M. Salemme, and N. Flegenheimer (eds.), Ancient evi-
dence for paleo South Americans: From where the
south winds blows. Center for the Study of the First Ame-
ricans. Texas A&M University Press. Texas: 87–92. 
Durán V. 2000. Poblaciones indígenas de Malargüe.
CEIDER, Serie libros No. 1. Facultad de Filosofía y Letras.
Universidad Nacional de Cuyo. Mendoza. 
2002. Nuevas consideraciones sobre la problemática
arqueológica del río Grande (Malargüe, Mendoza). In A.
Gil, G. Neme (eds.), Entre Montañas y Desiertos: Ar-
queología del Sur de Mendoza. Sociedad Argentina de
Antropología. Buenos Aires: 103–118.
Durán V., Ferrari J. 1991. El proceso de araucanización
del sur mendocino desde una perspectiva arqueológica.
In Actas de XI Congreso Nacional de Arqueología Chi-
lena. Santiago de Chile: 155–187.
Durán V., Mikkan R. 2009. Impacto del volcanismo Holo-
cénico sobre el poblamiento humano del sur de Mendoza
(Argentina). Intersecciones en Antropología 10: 295–310.
Durán V., Neme G., and Gil A. 1999. Algunos problemas
relacionados con el registro arqueológico de alero Puesto
Carrasco (curso medio del valle del río Grande, Malargüe,
Mendoza). Soplando el Viento. In Actas de las Terceras
Jornadas de Arqueología de la Patagonia. Puerto Ma-
dryn: 333–356.
Durán V., Winocur D., Stern C., Garvey R., Barberena R.,
Peña Monné J. L., and Benítez A. 2016. Impacto del volca-
nismo y glaciarismo holocénicos en el poblamiento huma-
no de la cordillera sur de Mendoza (Argentina): una per-
spectiva geoarqueológica. Intersecciones en Antropolo-
gía. Volumen especial 4: 33–46. 
Eerkens J. 2011. Pot Conveyance, Design Characteristics,
and Precontact Adaptations to Arid Environments. In R.
E. Hughes (ed.), Perspectives on Prehistoric Trade and
Exchange in California and the Great Basin. University
of Utah Press. Utah: 135–147.
2008. Nomadic Potters Relationships between Ceramic
Technologies and Mobility Strategies. In H. Barnard, W.
Wendrich (eds.), The archaeology of mobility: Old
World and New World nomadism (Vol. 4). Cotsen In-
stitute of Archaeology Press at UCLA. Los Angeles.
Falabella F., Pavlovic D., Planella M. T., and Sanhueza L.
2016. Diversidad y heterogeneidad cultural y social en
Chile central durante los períodos Alfarero temprano e
Intermedio tardío (300 años a.C. a 1450 años d.C.). In F.
Falabella, M. Uribe, L. Sanhueza, C. Aldunate, and J. Hidal-
go (eds.), Prehistoria en Chile. Editorial Universitaria.
Santiago: 365–399. 
Fernández M. V., Barberena R., Rughini A., + 7 authors,
and Glascock M. D. 2017. Obsidian geochemistry, geoar-
chaeology, and lithic technology in northern Patagonia
(Argentina). Journal of Archaeological Science: Reports
13: 372–381. https://doi.org/10.1016/j.jasrep.2017.04.009
Forasiepi A. M., Martinelli A. G., Gil A., Neme G., and Cer-
deño E. 2010. Fauna extinta y ocupaciones humanas en
el Pleistoceno final-Holoceno temprano del centro occi-
dental argentino. In M. De Nigris, P. M. Fernández, M.
Giardina, +4 editors, and H. D. Yacobaccio (eds.), Zooar-
queología a principios del Siglo XXI: Aportes Teóricos,
Metodológicos y Casos de Estudio. Ediciones El Espinillo.
Buenos Aires: 219–229. 
Franchetti F. 2019. Hunter-gatherer adaptation in the
deserts of northern Patagonia. Unpublished PhD thesis.
University of Pittsburgh. Pittsburgh. http://d-scholarship.
pitt.edu/37315/7/Franchetti%20ETD%20Final.pdf
Gambier M. 1985. La Cultura de los Morrillos. Instituto
de Investigaciones Arqueológicas y Museo. Facultad de
Population dynamics and human strategies in Northwestern Patagonia> a view from Salamanca Cave (Mendoza, Argentina)
Filosofía. Humanidades y Artes. Universidad Nacional de
San Juan. San Juan.
García A. 2005. Human occupation of the Central Andes
of Argentina (32°–34°S) During the Middle Holocene.
Quaternary International 132(1): 61–70.
https://doi.org/10.1016/j.quaint.2004.07.015
García A., Zárate M., and Páez M. 1999. The Pleistocene/
Holocene transition and human occupation in the Central
Andes of Argentina: Agua de la Cueva locality. Quater-
nary International 53: 43–52.
https://doi.org/10.1016/S1040-6182(98)00006-8
Garvey R., Bettinger R. 2018. A regional approach to pre-
historic landscape use in West-Central Argentina. Journal
of Archaeological Science: Reports 19: 846–855.
https://doi.org/10.1016/j.jasrep.2017.03.013
Giesso M., Durán V., Neme G., + 3 authors, and Sanhueza
L. 2011. A study of obsidian source usage in the Central
Andes of Argentina and Chile. Archaeometry 53(1): 1–
21. https://doi.org/10.1111/j.1475-4754.2010.00555.x
Gil A. 2006. Arqueología de la Payunia (Mendoza, Ar-
gentina). El poblamiento humano en los márgenes de
la agricultura. BAR International Series 1477. BAR Publi-
shing. Oxford. https://doi.org/10.30861/9781841719122
Gil A., Zárate M., and Neme G. 2005. Middle Holocene
paleoenvironments and the archaeological record of south-
ern Mendoza, Argentina. Quaternary International 132:
81–94. https://doi.org/10.1016/j.quaint.2004.07.014
Gil A., Tykot R. H., Neme G., and Shelnut N. 2006. Maize
on the frontier: isotopic and macrobotanical data from
centralwestern Argentina. In J. Staller, R. H. Tykot, and B.
Benz (eds.), Histories of Maize. Academic Press. London:
199–214.
Gil A., Neme G., Tykot R. H., Novellino P., Cortegoso V.,
and Durán V. 2009. Stable isotopes and maize consum-
ption in central western Argentina. International Journal
of Osteoarchaeology 19(2): 215–236.
https://doi:10.1002/oa.1041 
Gil A., Neme G., and Tykot R. H. 2011. Stable Isotopes and
Human Diet in Central Western Argentina. Journal of
Archaeological Science 38(7): 1395–1404.
https://doi.org/10.1016/j.jas.2011.01.010
Gil A., Villalba R., Franchetti F., + 3 authors, and Neme G.
2020. Between Foragers and Farmers: Climate Change
and Human Strategies in Northwestern Patagonia. Qua-
ternary 3(2): 17. https://doi:10.3390/quat3020017 
Glascock M. D., Ferguson J. R. 2012. Report on the Ana-
lysis of Obsidian Source Samples by Multiple Analytical
Methods. Available from the Archaeometry Laboratory.
University of Missouri. Columbia. 
Gordón F., Béguelin M., Rindel D., + 5 authors, and Ber-
nal V. 2019. Estructura espacial y dinámica temporal de
la ocupación humana de Neuquén (Patagonia Argentina)
durante el Pleistoceno final-Holoceno. Intersecciones en
Antropología 20(1): 93–105. 
Gradin C., Vayá C., Quintana M., Nami H., Salvino A., Be-
rón M., and Aguerre A. 1984. Investigaciones arqueoló-
gicas en Casa de Piedra. Dirección General de Cultura de
La Pampa. Santa Rosa.
Hogg A. G., Hua Q., Blackwell P. G., + 7 authors, and Tur-
ney C. S. 2013. SHCal13 Southern Hemisphere Calibra-
tion, 0–50,000 Years cal BP. Radiocarbon 55(4): 1889–
1903. https://doi.org/10.2458/azu_js_rc.55.16783
Lagiglia H. 1968. Secuencias culturales del Centro Oeste
Argentino: Valles del Atuel y Diamante. Revista Científica
de Investigaciones 1(4): 159–174.
1977. Arqueología y ambiente natural de los valles
del Atuel y del Diamante. Unpublished PhD thesis. Uni-
versidad Nacional de La Plata. La Plata. 
1997. Arqueología de Cazadores-Recolectores Cordil-
leranos de Altura. ICN Ediciones Ciencia y Arte. San
Rafael.
Llano C. 2011. Aprovechamiento de los Recursos Vege-
tales entre las Sociedades Cazadoras-Recolectoras del
Sur de Mendoza. Unpublished PhD thesis. Universidad
Nacional del COMAHUE. San Carlos de Bariloche.
Llano C., Barberena R. 2013. Explotación de especies ve-
getales en la Patagonia septentrional: el registro arqueo-
botánico de Cueva Huenul 1 (provincia de Neuquén, Ar-
gentina). Darwiniana, nueva serie 1(1): 5–19.
Marsh E. 2017. La fecha de la cerámica más temprana en
los Andes sur. Una perspectiva macrorregional mediante
modelos bayesianos. Revista del Museo de Antropología
Suplemento Especial 1: 83–94.
Martínez J. 2003. Ocupaciones humanas tempranas y
tecnología de caza en la microrregión de Antofagasta
de la Sierra (10000–7000 AP). Unpublished PhD thesis.
Universidad Nacional de Tucumán. San Miguel de Tucu-
mán.
Méndez C., Gil A., Neme G., +4 authors, and Maldonado
A. 2014. Middle Holocene radiocarbon ages in the Subtro-
pical Andes (~29°–35°S), climatic change and implica-
tions for human space organization. Quaternary Interna-
tional 356: 15–26.
http://repositorio.uchile.cl/handle/2250/132257
23
G. Neme, M. Zárate, M. P. Pompei, F. Franchetti, A. Gil, M. Giardina, V. P. Seitz, M. L. Salgán, C. Abbona, and F. Fernández
24
Morgan C., Tushingham S., Garvey R., Barton L., and Bet-
tinger R. 2017. Hunter-Gatherer Economies in the Old
World and New World. Oxford Research Encyclopedia of
Environmental Science. Oxford.
Moscardi B., Rindel D. D., and Pérez S. I. 2020. Human
diet evolution in Patagonia was driven by the expansion
of Lama guanicoe after megafaunal extinctions. Journal
of Archaeological Science 115: 105098.
https://doi.org/10.1016/j.quaint.2014.06.059
Neme G. 2007. Cazadores-Recolectores de Altura en
Los Andes Meridionales: El Alto Valle del Río Atuel.
BAR International Series 1591. BAR Publishing. Oxford.
2005. Aportes para la discusión del intercambio en el
sur de Mendoza. In Actas del XVI Congreso Nacional
de Arqueología Chilena. Sociedad Chilena de Arqueo-
logía. Santiago de Chile: 317–326.
2008a. Biogeografía Humana en los Andes Meridiona-
les: Tendencias Arqueológicas en el Sur de Mendoza.
Chungara 40(1): 5–18.
2008b. Faunal exploitation and agricultural transitions
in the South American agricultural limit. International
Journal of Osteoarchaeology 18: 293–306.
https://doi.org/10.1002/oa.944
2009. Human occupation and increasing Middle Holo-
cene aridity. Current Anthropology 50(1): 149–163.
https://doi.org/10.1086/596199 
Neme G., Gil A. 2010. What happened during the mid-Ho-
locene in arid western Argentina? Current Anthropology
51: 417–419. https://doi.org/10.1086/652644
2012. El Registro Arqueológico del sur de Mendoza en
Perspectiva Biogeográfica. In G. Neme, A. Gil (eds.),
Paleoecología Humana en el Sur de Mendoza: Per-
spectivas Arqueológicas. Sociedad de Antropología Ar-
gentina. Buenos Aires: 254–279. 
Neme G., Gil A., Garvey R., + 4 authors, and Michieli C.
2011. El Registro Arqueológico de la Gruta de El Manza-
no y sus implicancias para la Arqueología de Nordpatago-
nia. Magallania 39: 245–268.
Novellino P., Neme G. 1999. Hallazgos de restos esquele-
tales humanos en Laguna Blanca y El Sosneado (sur de
Mendoza). In Actas del XII Congreso Nacional de Ar-
queología Argentina. Tomo II. La Plata: 256–260.
Nullo F. E., Stephens G., Combina A., + 3 authors, and Za-
nettini J. C. M. 2005. Hoja Geológica 3569-III, Malar-
güe, provincia de Mendoza. Instituto de Geología y Re-
cursos Minerales, Servicio Geológico Minero Argentino.
Boletín 346. Buenos Aires.
Otaola C., Wolverton S., Giardina M., and Neme G. 2015.
Geographic Scale and Zooarchaeological Analysis of Late
Holocene Foraging Adaptations in Western Argentina.
Journal of Archaeological Science 55: 16–25.
https://doi.org/10.1016/j.jas.2014.12.004
Oyarzabal M., Clavijo J., Oakley L., + 8 authors, and León
R. 2018. Unidades de vegetación de la Argentina. Ecolo-
gía Austral 28: 40–63.
https://doi.org/10.25260/EA.18.1.0.399
Pérez S. I., González P., and Bernal V. 2016a. Past popu-
lation dynamics in Northwest Patagonia: an estimation
using molecular and radiocarbon data. Journal of Archa-
eological Science 65: 154–160.
https://doi.org/10.1016/j.jas.2015.11.013 
Pérez S. I., Postillone M. B., Rindel D., Gobbo D., Gonzá-
lez P. N., and Bernal V. 2016b. Peopling time, spatial oc-
cupation and demography of Late Pleistocene-Holocene
human population from Patagonia. Quaternary Interna-
tional 425: 214–223. https://doi.org/10.1016/j.quaint 
Pérez S. I., Postillone M. B., and Rindel D. 2017. Domes-
tication and human demographic history in South Ame-
rica. American journal of physical anthropology 163
(1): 44–52. https://doi.org/10.1002/ajpa.23176
Pompei M. P. 2019. Organización de la tecnología lítica
en la cuenca del río Atuel (sur de la provincia de Men-
doza) durante el Holoceno tardío. Unpublished PhD the-
sis. Universidad Nacional de La Plata. La Plata.
Puig S., Rosi M., Videla F., and Méndez E. 2011. Summer
and winter diet of the guanaco and food availability for
a High Andean migratory population (Mendoza, Argenti-
na). Mammalian Biology (Zeitschrift für Säugetierkun-
de) 76: 727–734.
https://doi.org/10.1016/j.mambio. 2011.07.001
Ratto N. 2003. Estrategias de caza y propiedades del re-
gistro arqueológico en la Puna de Chaschuil (Departa-
mento Tinogasta, Catamarca). Unpublished PhD thesis.
Universidad de Buenos Aires. Buenos Aires.
http://cambiocultural.homestead.com/ratto.html
Rindel D. 2017. Explorando la variabilidad en el registro
zooarqueológico de la Provincia del Neuquén: tendencias
cronológicas y patrones de uso antrópico. In F. Gordón,
R. Barberena, and V. Bernal (eds.), El poblamiento huma-
no del norte de Neuquén. Estado actual del conocimiento
y perspectivas. Aspha Ediciones. Buenos Aires: 101–122. 
Population dynamics and human strategies in Northwestern Patagonia> a view from Salamanca Cave (Mendoza, Argentina)
25
Riris P., Arroyo-Kalin M. 2019. Widespread population de-
cline in South America correlate with Middle Holocene cli-
mate change. Scientific Reports 9(1): 1–10.
https://doi.org/10.1038/s41598-019-43086-w
Roig F. A. 2000. Vegetación de la Provincia de Mendoza.
In F. A. Roig, E. Martínez Carretero, and E. Méndez (eds.),
Recurso y problemas ambientales de la zona árida. Pro-
vincias de Mendoza, San Juan y La Rioja. Junta de Go-
bierno de Andalucía-Universidades y Centros de Investi-
gación de la Región Andina. Mendoza: 29–48. 
Salgán M. L., Pompei M. P. 2017. Fuente de obsidiana El
Peceño: primeros resultados de su abordaje tecnológico,
geoquímico y espacial. Revista del Museo de Antropolo-
gía. Suplemento Especial 1: 51–58.
Salgán M. L., Tucker H., Luna L., Aranda C., and Gil A.
2012. Estudios arqueológicos y bioarqueológicos en la
cuenca media del río Malargüe (provincia de Mendoza): el
sitio Bajada de las Tropas 1. Relaciones de la Sociedad
Argentina de Antropología XXXVII(1): 113–135.
Salgán M. L., Gil A., and Neme G. 2014. Movilidad, aprovi-
sionamiento y uso de obsidiana en El Payén, sur de la pro-
vincia de Mendoza, Argentina. Comechingonia 18: 33–50.
Salgán M. L., Garvey R., Neme G., Gil A., Giesso M., Glas-
cock M. D., and Durán V. 2015. Las Cargas: Characteriza-
tion of a Southern Andean Obsidian Source and Its Prehi-
storic Use. Geoarchaeology 30(2): 139–150.
https://doi.org/10.1002/gea.21502
Salgán M. L., Pompei M. P., Diéguez S., Glacock M., Neme
G., and Gil A. 2020. Geoarchaeology and Spatial Distri-
butions of the “Coche Quemado” Obsidian Source in
Northwestern Patagonia. Archaeometry 62(2): 232–246.
https://doi.org/10.1111/arcm.12526 
Santoro C. M., Capriles J. M., Gayo E. M., + 10 authors, and
Marquet P. A. 2016. Continuities and discontinuities in
the socio-environmental systems of the Atacama Desert
during the last 13,000 years. Journal of Anthropological
Archaeology 46: 28–39.
https://doi.org/10.1016/j.jaa.2016.08.006
Sturm C., Clark J. K., and Barton L. 2016. The logic of ce-
ramic technology in marginal environments: implications
for mobile life. American Antiquity 81(4): 645–663.
https://doi.org/10.1017/S0002731600101015
Sugrañes N. 2017. La tecnología cerámica y su relación
con las estrategias de subsistencia y movilidad de pob-
laciones humanas en la cuenca del Atuel (sur de Men-
doza), durante el Holoceno tardío. Unpublished PhD
thesis. Universidad del Centro de la Provincia de Buenos
Aires. Olavarría. 
2019. Diversidad cerámica del sur de Mendoza. Discu-
tiendo tipologías. Relaciones de la Sociedad Argenti-
na de Antropología XLIV(1): 57–76.
Sugrañes N., Franchetti F. 2012. Antecedentes, problemas
y perspectivas del análisis cerámico en el sur de Mendo-
za. In G. Neme, A. Gil (eds.), Paleoecología humana en
el sur de Mendoza: perspectivas arqueológicas. Sociedad
Argentina de Antropología. Buenos Aires: 229–254. 
Tripaldi A., Zárate M., Neme G., Gil A., Giardina M. A., and
Salgán M. L. 2017. Archaeological site formation proces-
ses in northwestern Patagonia, Mendoza Province, Argen-
tina. Geoarchaeology 32(6): 605–621.
https://doi.org/10.1002/gea.21632
Ugan A., Neme G., Gil A., Coltrain J., Tykot R., and Novel-
lino P. 2012. Geographic variation in bone carbonate and
water δ18O values in Mendoza, Argentina and their rela-
tionship to prehistoric economy and settlement. Journal
of Archaeological Science 39(8): 2752–2763.
https://doi.org/10.1016/j.jas.2012.04.013
Veth, P., Ward I., Ditchfield K. 2017. Reconceptualising
Last Glacial Maximum discontinuities: A case study from
the maritime deserts of north-western Australia. Journal
of Anthropological Archaeology 46: 82–91.
https://doi.org/10.1016/j.jaa.2016.07.016
Villagra P., Cesca E., Álvarez J., Rojas F., Bourguet M., Ru-
bio C., and Mastrángelo P. 2010. Anexo II: Documento
de Ordenamiento de Bosques Nativos de la Provincia
de Mendoza. Secretaría de Medio Ambiente. Dirección de
Recursos Naturales y Renovables. Gobierno de la Provin-
cia de Mendoza. Mendoza.
Wiessner P. 2002. Hunting, healing, and hxaro exchange:
A long-term perspective on! Kung (Ju/’hoansi) large-game
hunting. Evolution and Human Behavior 23(6): 407–
436. https://doi.org/10.1016/S1090-5138(02)00096-X
Wolverton, S., Otaola, C., Neme, G., Giardina, M., and Gil
A. 2015. Patch choice, landscape ecology, and foraging
efficiency: the zooarchaeology of late Holocene foragers
in western Argentina. Journal of Ethnobiology 35: 499–
518. https://doi.org/10.2993/etbi-35-03-499-518.1