jueves, 6 de febrero de 2014

Dating of the MiDDle to Upper paleolithic transition at the abrigo 3 Del hUMo (Málaga, spain)

Dating of the MiDDle to Upper paleolithic
   transition at the abrigo 3 Del hUMo
             (Málaga, spain)
Julián Ramos Fernándezi, Katerina Doukaii, Alistair W. G. Pikeiii,
           Louise Thomasiv, Peter van Calstereniv y João Zilhãov
abstract: The Humo rockshelters contain one of the few long archeological stratigraphies spanning the
Middle-to-Upper Paleolithic transition in southern Iberia. Stalagmite, bone and shell samples from Area 1 of the
Abrigo 3 were collected in order to assess the chronology of the latest Mousterian of this locus. The results indicate
that the sequence is older than hitherto thought, with U-series placing the level 25 stalagmite in MIS 5 and Ra-
diocarbon on mussel shell placing level 18 in the range of 43.6-44.8 thousand calendar years ago. This latter result
provides a terminus post quem for Level 17, the uppermost unit of the Abrigo 3 containing stone tools of Middle
Paleolithic technology.
KeyworDs: Iberia, Middle Paleolithic, Upper Paleolithic, Radiocarbon, U-series.
cronología De la transición Del paleolítico MeDio
al sUperior en el abrigo 3 Del hUMo (Málaga, españa)
resUMen: Los abrigos del Humo contienen una de las pocas estratigrafías arqueológicas abarcando la transición
del Paleolítico Medio al Superior en el sur de la Península Ibérica. Se han obtenido muestras de estalagmita, hueso
y concha del Área 1 del Abrigo 3 para determinar la cronología del último Musteriense de este locus. Los resultados
indican que la secuencia es más antigua que lo hasta ahora pensado, puesto que el método de la serie del uranio sitúa
la estalagmita del nivel 25 en el MIS 5 y el radiocarbono sobre concha de mejillón sitúa el nivel 18 entre hace 43.6
y 44.8 miles de años (de calendario). Este ultimo resultado proporciona un terminus post quem para el nivel 17, la
unidad más alta del Abrigo 3 que aún contiene una industria de tecnología Paleolítico Medio.
Mainake, XXXIII / 2011-2012 / pp. 275-284 / ISSN: 0212-078-X
palabras clave: Península Iberica, Paleolítico Medio, Paleolítico Superior, radiocarbono, serie del uranio.
i
ii
iii
iv
v
Parque Arqueológico de La Araña, 29004 Málaga; Spain; cuevadelhumo@gmail.com.
Research Laboratory for Archaeology and the History of Art, Dyson Perrins Building; University of Oxford; Oxford
OX1 3QY; United Kingdom. katerina.douka@rlaha.ox.ac.uk.
Department of Archaeology & Anthropology, University of Bristol; 43 Woodland Road; Bristol BS8 1UU; United
Kingdom. Alistair.Pike@bristol.ac.uk.
Open University Uranium-Series Laboratory, Earth and Environmental Sciences; The Open University; Milton
Keynes MK7 6AA; United Kingdom. l.e.thomas@open.ac.uk; v.calsteren@open.ac.uk.
ICREA Research Professor, University of Barcelona. Seminari d’Estudis i Recerques Prehistòriques; Departament
de Prehistòria, Història Antiga i Arqueologia; Facultat de Geografia i Història; C/ Montalegre 6; 08001 Barcelona;
Spain. joao.zilhao@ub.edu.


276
Julián Ramos Fernández, Katerina Douka, Alistair W. G. Pike, Louise Thomas, Peter van Calsteren y João Zilhão
1. introDUction
     The Humo rockshelters are located in the
La Araña archeological complex, at the eastern
end of the Málaga city limits (Figure 1). This
karst locality features a number of sites that
have yielded evidence of human settlement
between the Middle Pleistocene and later pre-
historic times, including Paleolithic parietal art
at Cueva Navarro. The area is also of significant
geological and paleontological interest because
of the preservation of paleocliffs, abrasion
platforms, raised beaches and associated conti-
nental deposits that have provided key evidence
for the study of Quaternary sea level change in
the Mediterranean1.
     The line of rockshelters known as Abrigos
del Humo is excavated along the lowermost
paleocliff, where the high sea level stand of
Marine Isotope Stage (MIS) 5 is marked at an
elevation of 0.9 m amsl by a Strombus bubo-
nius beach. Archeological work carried out in
the area since the early 20th century explored
a number of deposits filling cavities along the
base of the paleocliff, where they have been
preserved from the effects of erosion, quarrying
and road construction that, seaward, entailed
their loss. Such deposits thus correspond to the
truncated fills of caves and rockshelters whose
original topographic configurations are in most
cases difficult if not impossible to reconstruct.
Their archeological contents document use for
settlement and ritual purposes in the Middle
Paleolithic, the Solutrean, the Magdalenian, the
Early Neolithic and the Copper Age2.
     Fieldwork in this locality was first carried
out in 1965 by Gálvez Pacheco and has been on-
going since 1980 under the directon of one of us
( J. R. F.). In 2003-04, the deposits preserved in
Fig. 1. Site location. Top: Google Earth view over the La Ara-
ña/El Humo archeological localities. Middle: Plan indicating the
position of the different caves and rockshelters making up the
complex. Bottom: View from the east over the paleocliff, the
     Cueva del Humo and the Abrigo 3 in December 2005.
Mainake, XXXIII / 2011-2012 / pp. 275-284 / ISSN: 0212-078-X
the different areas of the Abrigo 3 were profiled
with standard archeological methodology (Fi-
gure 2). The disturbed sediments were removed
and systematically screened, the in situ deposits
were excavated according to the observed stra-
tigraphy and with plotting of individual finds,
and samples for dating and for geological, paly-
nological and anthracological analyses were
1
2
RAMOS-FERNÁNDEZ, J. y DURÁN, J. J. (1998); FERRE, E., et al. (2004).
RAMOS-FERNÁNDEZ, J. (2005); RAMOS-FERNÁNDEZ, J. et al. (2005).


Dating of the Middle to Upper Paleolithic transition at the Abrigo 3 del Humo (Málaga, Spain)
277
taken, but, so far, only brief descriptions of the
results obtained have been published3. The se-
dimentary fill consists of a succession of sandy
deposits including wall-derived angular clasts
and blocks and featuring interstratified stalag-
mite formations. The better preserved basal
3
4
Idem.
Sensu Claassen: CLAASSEN, C. (1998).
levels contain abundant stone tools associated
with wood charcoal, animal bone and mollusc
shells. In some levels, the latter form lenses of
such density as to qualify for the category of
shell-matrix deposits4; this is namely the case
with Level 19 (Figure 2).
Mainake, XXXIII / 2011-2012 / pp. 275-284 / ISSN: 0212-078-X
Fig. 2. The Abrigo 3. Left: General view from the east of the site in December 2005, with indication of the stalagmites
sampled in the western area. Right, top: Detail of the stratigraphy in the east profile of the western area; note, at the top,
the dense accumulation of osyter and mussel shells in level 19. Right, bottom: The stratigraphic succession observed in
the east profile of the western area, with indication of the correspondence between stalagmite samples and stratigraphic
units of the main shell-matrix deposits of MIS-3 age (levels 17, 19 and 21); the vertical bar marks the span of the deposits
                                                     in the photo above.


278
Julián Ramos Fernández, Katerina Douka, Alistair W. G. Pike, Louise Thomas, Peter van Calsteren y João Zilhão
Mainake, XXXIII / 2011-2012 / pp. 275-284 / ISSN: 0212-078-X
     Of primary concern here is the observation
that, in the western area of the site, the interface
between the Middle and the Upper Paleolithic
is located between Levels 15 and 17: Level 15
yielded Upper Paleolithic artifacts, the material
in Level 16 is of uncertain affinities, upper Level
17 is a sterile eboulis deposit, and the stone tool
component in lower Level 17 is unambiguously
Middle Paleolithic. Unpublished Thermolu-
minescence (TL) results obtained on samples
from stalagmites sandwiching these levels brac-
keted the transition to between ~23 ka (the age
obtained for Level 15) and ~35 ka (the age ob-
tained for Level 23).
     As summarized in a recent review5, the
uppermost levels from the long Middle Paleo-
lithic sequences excavated at Sima de las Pa-
lomas (Murcia), Gorham’s Cave (Gibraltar)
and Gruta da Oliveira (Portugal) are associated
with dates younger than ~35 ka 14C BP (~40 ka
cal BP), while the Protoaurignacian of northern
Spain and France, presumably modern human-
associated, began as early as ~36.5 ka 14C BP
(~41.5 ka cal BP). The TL results reported for
interstratified stalagmites from Humo 3 could
therefore indicate that Levels 17-22 of the site
represented yet another instance of the persis-
tence of Neandertals and the Middle Paleolithic
in southern and western Iberia.
     However, the presence of an Evolved Au-
rignacian in Level 11 of Bajondillo (Torremo-
linos), a rockshelter site located some 10 km to
the southwest, invited caution in the acceptance
of such results. Although the radiocarbon dates
for that level are problematic, as discussed in the
review mentioned above6, elsewhere in Europe
the Evolved Aurignacian begins ~37 ka cal BP
and is associated with modern humans exclu-
sively7. Against this regional background, the
5
6
7
ZILHÃO, J. (2006).
idem.
ZILHÃO, J. (2007); HIGHAM, T. et al. (2011).
terminus post quem for the overlying Middle Pa-
leolithic represented by the unpublished TL age
of ~35 ka for Level 23 of the Abrigo 3 implied
that Middle Paleolithic Neandertals and Upper
Paleolithic modern humans coexisted for se-
veral millennia in the Bay of Malaga; however,
it could also be the case that the true antiquity
of the Middle Paleolithic of the Humo complex
fell significantly beyond the TL-derived esti-
mates. It was in this context that it became im-
portant to assess the chronology of the Abrigo
3 del Humo with alternative techniques. Ac-
cordingly, the site was included in the 2006-07
dating project “The late survival of the Middle
Palaeolithic and Neanderthals in the Iberian Pe-
ninsula” funded in the UK by NERC (Natural
Environment Research Council) and AHRC
(Arts and Humanities Research Council).
2. Materials anD MethoDs
    In March 2006, one of us ( J. R. F.) collected
from the then extant profiles samples spanning
the entire stratigraphic sequence of the Abrigo 3.
In total, 9 speleothem samples and 71 bone and
shell samples were taken from the three areas of
the site. As the archeological stratigraphy was
better defined in its westernmost part, Area 1,
it was decided to focus our efforts on samples
from this area: five speleothem samples were
submitted for dating by U-series at the OUUSL
(Open University Uranium Series Laboratory);
nine bone and nine shell samples and were sub-
mitted for dating by radiocarbon at the ORAU
(Oxford Radiocarbon Accelerator Unit, Uni-
versity of Oxford). The provenience of these
samples is provided in Figure 3.
    At the OUUSL, speleothem samples were
spiked with a mixed 229Th/236U spike and totally


Dating of the Middle to Upper Paleolithic transition at the Abrigo 3 del Humo (Málaga, Spain)
279
Fig. 3. The north profile of the western area of the Abrigo 3 with indication of the position of the samples collected from the profiles
extant in March 2006 and of the approximate age obtained for the dated stratigraphic units. Sample E2, not shown, came from an
                      area of level 25 located ~1.5 m to the east (i.e., to the right) of the profile illustrated here.
dissolved in a sequence of nitric acid and hydro-
fluoric acid. The uranium and thorium fractions
were separated using standard ion exchange
procedures (Calsteren & Thomas 2012). Ura-
nium and thorium isotope ratios were deter-
mined on a Nu Instruments Multi-Collector
Inductively-Coupled Plasma Mass Spectrom-
eter equipped with an RPQ (Retarding Poten-
tial Quadrupole) lens for improved abundance
sensitivity and Ion Counting. Fractionation and
instrument drift were corrected using the data
from the standard-sample bracketing protocol.
Procedure blank contributions on 234U and
230Th were below ppm level. Ages marked ‘cor-
Mainake, XXXIII / 2011-2012 / pp. 275-284 / ISSN: 0212-078-X
rected’ were recalculated from (234Th/234U) and
(234U/238U) ratios where the contribution of a
‘detrital’ component in secular equilibrium used
a 232Th/238U crustal ratio of 3.12 and assumed
that all determined 232Th is detrital. Uncertain-
ties for concentrations and activity ratios are 1
standard error, uncertainties in the calculated
ages are 2 times the propagated uncertainties
of the sample and bracketing standard runs and
the uncertainty in the 232Th concentration for
the detrital corrected age. Decay constants are:
λ 238U=1.55125E-10; λ 232Th=4.94750E-11; λ
234U=2.82629E-06; λ 230Th=9.15771E-06.


280
Julián Ramos Fernández, Katerina Douka, Alistair W. G. Pike, Louise Thomas, Peter van Calsteren y João Zilhão
Mainake, XXXIII / 2011-2012 / pp. 275-284 / ISSN: 0212-078-X
     Three shell samples were dated at the
ORAU using routine pretreatment method-
ology for marine carbonates. The samples were
surface-cleaned with an air abrasive system and
using fine aluminium oxide powder as medium.
Abrasion removed all adhering material and the
first shell layers, until internal, usually opaque,
shell layers were exposed. The cleaned shell un-
derwent mineral staining using Feigl’s solution.
This way aragonite and calcite were visually
separated and a small aragonitic fragment was
sawed off and crushed with a mortar and pestle.
     Approximately 50 mg of shell powder were
reacted with 5 ml of 80% phosphoric acid
(H3PO4) in vacuo. The CO2 gas evolved via this
process was purified in an EA-CF-IRMS system
and was transferred to a graphitization rig where
it was converted to graphite via H2 reduction at
560° C for 6 h, in the presence of 2-2.5 mg of a
pre-cleaned Fe+ catalyst. The graphite was then
pressed into an aluminium target holder, prior
to measurement using an accelerator mass spec-
trometer (HVEE Tandetron AMS system).
     The same procedure was followed for the
reference carbonate sample IAEA C1 Carrara
Marble, a geological sample of infinite age which
is routinely analysed alongside the unknown-
age samples to monitor introduction of modern
contamination during sample pretreatment.
      Despite the fact that other shell species
were available for dating, we chose to date
mussel shells and not limpets or scallops. Mus-
sels are linked to food collection and are easy
to harvest alive, therefore their age-at-death
should reflect the time of collection. Moreover
they are filter-feeders, which means that they
feed on seawater dissolved inorganic carbon
(DIC) and marine particulate organic matter
(POM). Limpets were also collected for food
since the Middle Palaeolithic, however they are
sometimes suspected to incorporate into their
shells dead carbon (containing no 14C) from the
limestone rocks on which they graze. Therefore,
their measured radiocarbon content would
require a particular correction, but the exact
offset is unknown. Scallops, on the other hand,
are deep-sea animals the collection of which for
food purposes is difficult to prove. It is most
likely that these shells represent beach collec-
tion and would therefore suffer from an inbuilt
age. This means that the time of their collec-
tion (the archeological event of interest) and
the time of their death (the event dated when
measuring the shell’s radiocarbon age) may be
several centuries or even millennia apart.
     Alterations in the mineralogy of an ar-
chaeological shell is usually reflected in the
transformation of original aragonite and high-
Mg calcite to secondary low-Mg calcite. Such
diagenetic processes are very likely to affect the
information we want to obtain especially from
samples of such antiquity as the Abrigo del
Humo shells. Limpets and scallops are almost
entirely calcitic in nature, therefore presence of
secondary low-Mg calcite would have been dif-
ficult to detect. Mussels contain both original
aragonite and high-Mg calcite; in this study
the aragonitic phase of the (nacre) was sampled
and dated. This work was undertaken before a
new pretreatment protocol was established at
the ORAU for the physical separation of orig-
inal aragonite from secondary, low-Mg calcite
(CarDS)8. While we did not perform XRD
for the precise identification of the dated min-
eral, staining with Feigl’s solution allowed us to
separate mechanically any potentially contami-
nating phase. Despite our efforts, the presence
of contamination in the mussel nacre cannot
be completely ruled out; we will return to this
point in the next section.
8
DOUKA, K. et al. (2010).


Dating of the Middle to Upper Paleolithic transition at the Abrigo 3 del Humo (Málaga, Spain)
281
Fig. 4. Bayesian model containing the available chronometric data from Abrigo del Humo 3. The radiocarbon mea-
surements were calibrated using Marine09 and a local reservoir offset. The U-Th determinations reflect calendar
ages and do not require calibration. OxA-16805 was flagged an outlier and is marked in red (see text for details).
3. resUlts
    We obtained results for three speleothem
and three shell samples, as reported and dis-
cussed here. The results and associated prove-
nience and chemistry details are reported in
Table 1 for the U-series speleothem dates and in
Table 2 for the radiocarbon shell dates.
    All radiocarbon results were calibrated
using Marine09, which subtracts 400 14C years
from IntCal099. Using OxCal 4.1.710, all dating
results, both U-Th and 14C measurements, were
included in a Bayesian statistical framework
the results of which are presented in Figure 4.
In this age model, the local reservoir offset of
-22 ±35 14C years (CHRONOS database) was
subtracted from each measurement prior to ca-
9 REIMER, P. et al. (2009).
10 BRONK RAMSEY, C. (2009).
Level Sample # Lab #
 18M5316803
 21M7016804
 24M7116805
δ13C
-0.37
 0.06
 0.74
  Age BP
40730 ± 310
42140 ± 360
32580 ± 180
 Age cal BP
43590-44819
44580-45814
36325-37097
Table 1. OxA radiocarbon dates on Mytilus sp. shells from
the Abrigo 3 del Humo. Calibration was carried out with the
    Calib 6.1.0 software using the marine09.14c curve.
Mainake, XXXIII / 2011-2012 / pp. 275-284 / ISSN: 0212-078-X
libration. The calibrated results are reported at
the 68.2% and 95.4% level of confidence (2 and
3σ). The small number of measurements do not
allow any drastic refinement of the chrono-stra-
tigraphic succession, however the model per-
mits some statistically-sound observations to be
made regarding the likely start and end bounda-
ries of each dated layer, as well as on the dura-
tion of each phase (the archaeological levels, or
Layers) and the identification of outliers.


282
Julián Ramos Fernández, Katerina Douka, Alistair W. G. Pike, Louise Thomas, Peter van Calsteren y João Zilhão
Level
25
25
23
Sample
Field #
  E1
  E2
  E3
Weight
 (mg)
405.56
525.31
427.08
        U
     (ppb)
0.34942±0.00095
0.22569±0.00032
0.29711±0.00052
238
      Th
    (ppb)
422.60±76.39
 55.21±9.98
307.47±55.58
232
 (230Th/232Th)
 activity ratio
1.8707±0.0772
7.6020±0.3147
1.9108±0.0806
  (230Th/234U)
 activity ratio
0.6797±0.0059
0.5578±0.0049
0.5950±0.0057
Table 2. OUUSL U-series dates on speleothem samples from the Abrigo 3 del Humo.
  Errors are 2 σ. See text for details on the correction for detritical contamination.
Mainake, XXXIII / 2011-2012 / pp. 275-284 / ISSN: 0212-078-X
     The U-series speleothem dates show that
the deposition of the site’s basal flowstone de-
posits (Levels 23 and 25) occurred significantly
earlier than the unpublished 35 ka estimate
previously available for Level 23 and the ~36.8-
36.5 ka cal BP result (OxA-16805) obtained by
radiocarbon for a mussel shell from Level 24.
OxA-16805 is out of stratigraphic order and
an outlier. It comes from a sample collected >1
m below Level 21, making it highly unlikely
that the anomaly can be explained by post-de-
positional disturbance; nothing was observed
in field that can corroborate such large scale
movement of material. Dating marine carbo-
nates close to the limit of the radiocarbon te-
chnique is not easy and requires careful sample
selection, screening and cleaning. No XRD
analyses were performed for the dated sample;
therefore, diagenetic alterations and incorpora-
tion of younger carbon during recrystallization
processes cannot be ruled out. Although the
sample’s chemistry provides no indication of
contamination by younger carbonates (the δ13C
value does not reveal incorporation of lighter te-
rrestrial carbon), in the context of the ensemble
of results obtained we may conclude that OxA-
16805 represents a minimum age only and, ac-
cordingly, should be interpreted simply as pro-
viding a terminus ante quem of 43.5 ka cal BP
for the deposition of Level 24.
     Even if a significant error is incorporated
in the detrital correction and propagated to
the corrected U-series age, stalagmite sample
E2 (Level 25) remains firmly in the MIS 5 time
bracket. Such an incorporation would imply a
wider bracket for stalagmite sample E3 (Level
23), placing it in the region of 38 to 66 ka, but
the radiocarbon results for overlying Level 21
constrain to 44.6 ka the upper limit of the age
interval of E3. Overall, therefore, the results
show that, in the stratigraphic sequence of
Abrigo 3, only Level 23 and those above it are
of MIS 3 age.
    The ~44.8–43.6 ka cal BP radiocarbon
date for Level 18 provides a terminus post
quem for Level 17, the site’s uppermost Middle
Paleolithic level. However, as U-series dating
of the speleothem sample from Level 15 was
unsuccessful and no mussel shell samples were
available from Level 17 itself, the age of this
level cannot be firmly established at present.
    Where the issue of the late persistence of
the Middle Paleolithic is concerned, our re-
sults for the Abrigo 3 del Humo are, therefore,
inconclusive. However, they corroborate the
evidence concerning the Middle Paleolithic
exploitation of marine resources provided by a
cluster of other sites located along the Medite-
rranean coast of Iberia, between Gibraltar and
Murcia11. The importance of such resources for
the regional populations of Neandertals is at-
tested by the fact that shell accumulations are
important even in Level 24, which dates for
bracketing speleothems place in MIS 4, i.e., in
a glacial maximum, when the coast line was far-
thest from the site. That such accumulations are
11 ZILHÃO, J. et al. (2010); COLONESE, A. C. et al. (2011); CORTÉS-SÁNCHEZ, M. et al. (2011).


Dating of the Middle to Upper Paleolithic transition at the Abrigo 3 del Humo (Málaga, Spain)
283
Table 2 (Cont.).
Level
25
25
23
  (234U/238U)
 activity ratio
1.0705±0.0106
1.0561±0.0031
1.0666±0.0053
      age (ka)
    uncorrected
121981/+ 3180/-3214
88263/+ 2253/- 2273
97422/+ 2767/- 2795
      age (ka)
     corrected
72558/+2721/-2745
79737/+2875/-2908
56214/+2232/-2250
more important in the overlying levels of MIS 3
age (namely, Levels 21, 19 and 17) is therefore
explained as a byproduct of sea level change and
corresponding implications for site function
and resource acquisition patterns. All other
things being equal (namely where technology
and settlement and subsistence systems are con-
cerned), it is to be expected that the accumula-
tion of marine resources at hunter-gatherer sett-
lement sites will vary inversely to the distance
separating them from the coast. The Abrigo 3
evidence fits such an expectation and, in con-
text, strongly suggests that, by early Upper Pleis-
tocene times, marine resources were being routi-
nely exploited, possibly on a seasonal basis, and
played a rather significant role in Neandertal
economies of southwestern Iberia.
acKnowleDgMents
    The work reported here was funded by a
grant in kind from NERC (IP/911/0506) and
AHRC (NF/2006/1/4) to J. Zilhão (Principal
Investigator), Alistair W. G. Pike and Thomas F.
G. Higham.
bibliografía
BRONK RAMSEY, C. (2009): “Bayesian analysis of radiocarbon dates”, Radiocarbon, 51(1): 337-360.
CALSTEREN, P. van, y THOMAS, L. E. (2012): “Quantitation of protactinium, 231Pa in abyssal carbonate”, Journal of
   Analytical Atomic Spectrometry, 27: 952-956.
CLAASSEN, C. (1998): Shells, Cambridge.
COLONESE, A. C., MANNINO, M. A., BAR-YOSEF MAYER, D. E., FA, D. A., FINLAYSON, J. C., LUBELL,
   D. y STINER, M. (2011): “Marine mollusc exploitation in Mediterranean prehistory: An overview”, Quaternary
   International, 239: 86-103.
CORTÉS-SÁNCHEZ, M., MORALES-MUÑIZ, A., SIMÓN-VALLEJO, M. D., LOZANO-FRANCISCO, M. C.,
   VERA-PELÁEZ, J. L., FINLAYSON, C., RODRÍGUEZ-VIDAL, J., DELGADO-HUERTAS, A., JIMÉNEZ-
   ESPEJO, F. J., MARTÍNEZ-RUIZ, F., MARTÍNEZ-AGUIRRE, M. A., PASCUAL-GRANGED, A. J., BER-
   GADÀ-ZAPATA, M. M., GIBAJA-BAO, J. F., RIQUELME-CANTAL, J. M., LÓPEZ-SÁEZ, J. A., RODRIGO-
   GÁMIZ, M., SAKAI, S., SUGISAKI, S., FINLAYSON, G., FA, D. A. y BICHO, N. F. (2011): “Earliest Known
   Use of Marine Resources by Neanderthals”, PLoS ONE, 6 (9): e24026.
DOUKA, K., HEDGES, R. E. M. y HIGHAM, T. F. G. (2010): “Improved AMS 14C dating of shell carbonates using
   high-precision X-Ray Diffraction (XRD) and a novel density separation protocol (CarDS)”, Radiocarbon, 52 (2-3):
   735-751.
FERRE, E., CORTÉS, M., RAMOS, J., SENCIALES, J. M., LOZANO-FRANCISCO, M. C., VERA-PELAEZ, J. L.,
   AGUILERA, R. y NAVARRETE, I. (2004): “El Cuaternario reciente en el sector oriental de la Bahía de Málaga.
   Rasas y depósitos marinos, continentales y arqueológicos”, Cuaternario y Geomorfología, 18 (1-2): 73-93.
HIGHAM, T., JACOBI, R., BASELL, L., BRONK RAMSEY, C., CHIOTTI, L. y NESPOULET, R. (2011): “Pre-
   cision dating of the Palaeolithic: A new radiocarbon chronology for the Abri Pataud (France), a key Aurignacian
   sequence”, Journal of Human Evolution, 61: 549-563.
RAMOS-FERNÁNDEZ, J. (2005): “El Paleolítico Superior en La Araña (Málaga), en P. Cantalejo, (ed.), Arte rupestre y
   sociedades prehistóricas con expresiones gráficas, Málaga, pp. 181-190.
RAMOS-FERNÁNDEZ, J., AGUILERA, R., CORTÉS, M. y BAÑARES, M. M. (2005): “El parque arqueológico de
   La Araña. Una vía para el estudio, la conservación y la puesta en valor de yacimientos paleolíticos”, en M. Santonja, A.
Mainake, XXXIII / 2011-2012 / pp. 275-284 / ISSN: 0212-078-X


284
Julián Ramos Fernández, Katerina Douka, Alistair W. G. Pike, Louise Thomas, Peter van Calsteren y João Zilhão
   Pérez-González y M. Machado, (eds.), Geoarqueología y patrimonio en la Península Ibérica y el entorno mediterráneo,
   Almazán (Soria), pp. 625-638.
RAMOS-FERNÁNDEZ, J. y DURÁN, J. J. (1998): “El Solutrense de La Araña (Málaga)”, en J. L. Sanchidrián y M. D.
   Simón, (eds.), Las culturas del Pleistoceno Superior en Andalucía, Málaga, pp. 63-74.
REIMER, P. J., BAILLIE, M., BARD, E., BAYLISS, A., BECK, J. W., BLACKWELL, P. G., BRONK RAMSEY, C.,
   BUCK, C. E., BURR, C. E., EDWARDS, R. L., FRIEDRICH, M., GROOTES, P. M., GUILDERSON, T. P.,
   HAJDAS, I., HEATON, T. J., HOGG, A. G., HUGHEN, K. A., KAISER, K. F., KROMER, B., MCCORMAC,
   F. G., MANNING, S. W., REIMER, R. W., RICHARDS, D. A., SOUTHON, J. R., TALAMO, S., TURNEY, C.
   S. M., VAN DER PLICHT, J. y WEYHENMEYER, C. E. (2009): “INTCAL09 and MARINE09 radiocarbon age
   calibration curves, 0-50,000 years cal BP”, Radiocarbon, 51: 1111-1150.
ZILHÃO, J. (2006): “Chronostratigraphy of the Middle-to-Upper Paleolithic Transition in the Iberian Peninsula”,
   Pyrenae, 37: 7-84.
ZILHÃO, J. (2007): “The emergence of ornaments and art: an archaeological perspective on the origins of behavioural
   ‘modernity”, Journal of Archaeological Research, 15: 1-54.
ZILHÃO, J., ANGELUCCI, D., BADAL-GARCÍA, E., d’ERRICO, F., DANIEL, F., DAYET, L., DOUKA, K.,
   HIGHAM, T. F. G., MARTÍNEZ-SÁNCHEZ, M. J., MONTES-BERNÁRDEZ, R., MURCIA-MASCARÓS,
   S., PÉREZ-SIRVENT, C., ROLDÁN-GARCÍA, C., VANHAEREN, M., VILLAVERDE, V., WOOD, R. y ZA-
   PATA, J. (2010): “Symbolic Use of Marine Shells and Mineral Pigments by Iberian Neandertals”, Proceedings of the
   National Academy of Sciences USA, 107 (3): 1023-1028.
Mainake, XXXIII / 2011-2012 / pp. 275-284 / ISSN: 0212-078-X