Diet of Upper Paleolithic Modern Humans: Evidence From Microwear Texture Analysis

Abstract

This article presents the results of the occlusal molar microwear texture analysis of 32 adult Upper Paleolithic modern humans from a total of 21 European sites dating to marine isotope stages 3 and 2. The occlusal molar microwear textures of these specimens were analyzed with the aim of examining the effects of the climatic, as well as the cultural, changes on the diets of the Upper Paleolithic modern humans. The results of this analysis do not reveal any environmentally driven dietary shifts for the Upper Paleolithic hominins indicating that the climatic and their associated paleoecological changes did not force these humans to significantly alter their diets in order to survive. However, the microwear texture analysis does detect culturally related changes in the Upper Paleolithic humans' diets. Specifically, significant differences in diet were found between the earlier Upper Paleolithic individuals, i.e., those belonging to the Aurignacian and Gravettian contexts, and the later Magdalenian ones, such that the diet of the latter group was more varied and included more abrasive foods compared with those of the former. Am J Phys Anthropol, 2014. © 2014 Wiley Periodicals, Inc.

Full text

Diet of Upper Paleolithic Modern Humans: Evidence
From Microwear Texture Analysis
Sireen El Zaatari,
1,2
* and Jean-Jacques Hublin
1
1
Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology, Leipzig D-04103, Germany
2
Department of Early Prehistory and Quaternary Ecology, Paleoanthropology, Senckenberg Center for Human
Evolution and Paleoecology, Eberhard Karls University of T€
ubingen, T€
ubingen 72070, Germany
KEY WORDS Aurignacian; Gravettian; Magdalenian
ABSTRACT This article presents the results of the
occlusal molar microwear texture analysis of 32 adult
Upper Paleolithic modern humans from a total of 21
European sites dating to marine isotope stages 3 and 2.
The occlusal molar microwear textures of these speci-
mens were analyzed with the aim of examining the
effects of the climatic, as well as the cultural, changes on
the diets of the Upper Paleolithic modern humans. The
results of this analysis do not reveal any environmentally
driven dietary shifts for the Upper Paleolithic hominins
indicating that the climatic and their associated paleoeco-
logical changes did not force these humans to signifi-
cantly alter their diets in order to survive. However, the
microwear texture analysis does detect culturally related
changes in the Upper Paleolithic humans’ diets. Specifi-
cally, significant differences in diet were found between
the earlier Upper Paleolithic individuals, i.e., those
belonging to the Aurignacian and Gravettian contexts,
and the later Magdalenian ones, such that the diet of the
latter group was more varied and included more abrasive
foods compared with those of the former. Am J Phys
Anthropol 000:000–000, 2014. V
C2014 Wiley Periodicals, Inc.
The availability of food resources and the possession of
successful strategies to access these resources are the
first requirements for the survival of any species, includ-
ing hominins. Using occlusal molar microwear texture
analysis, this study examines the effects of the climati-
cally induced changes in food supply and the develop-
ment of new technology on the diets of European Upper
Paleolithic modern humans. The Upper Paleolithic is the
period associated with the first modern human presence
in Europe. This period begins at the time of the severe
climatic fluctuations of marine isotope stage (MIS) 3,
continues through the extreme conditions of the Last
Glacial Maximum (LGM) in MIS 2, and ends with the
warming trend at the end of MIS 2. In terms of technol-
ogy, the Upper Paleolithic period is characterized by
unprecedented advancements. Its technological com-
plexes are mostly distinguished from those of earlier
periods through their considerable increase in artifact
and raw material diversity and their unparalleled rate
of development, innovation, and change (e.g., Klein,
2009). Indeed, over the course of only 30 ka, a succession
of techno-economic variants are recognized among which
the Aurignacian, the Gravettian, and the Magdalenian
stand out as major traditions.
The first of these traditions, the Aurignacian, is charac-
terized by the increased use of organic material such as
bone and antler as raw materials for the production of
formal tools, e.g. Mladec
ˇpoints, split-based points (Lio-
lios, 2006; Tartar and White, 2013). The lithic assemb-
lages, on the other hand, include both large blades and
bladelets produced from prismatic, carinated endscrapers
and burin-like cores. These blanks were used for the
manufacture of a retouched tool-kit including Aurigna-
cian blades and Dufour bladelets (e.g., Brooks, 1982;
White, 1993; Blades, 1999). During this period, ornamen-
tation becomes systematic and mobile art objects and
parietal art are also well-documented (e.g., Vanhaeren
and d’Errico, 2006). This techno-complex is generally
believed to have been brought into Europe by the earliest
modern humans to enter the continent around 43 ka Cal
BP [an even earlier date is suggested by archaeological
evidence from western Eurasia (Svoboda and Bar-Yosef,
2003; Hublin, 2012)], i.e., during the extreme climatic
instability of MIS 3 (60–24 ka Cal BP) (e.g., Hublin,
2013). This climatic instability is clearly visible in various
environmental proxies as severe millennial scale climatic
fluctuations between cold phases, when conditions
approached those prevailing during glacial times, and
warm phases, when conditions were similar to those of
the Holocene (e.g., Shackleton, 1977; Barron and Pollard,
2002). These climatic oscillations greatly affected vegeta-
tion cover in Europe which, in response, fluctuated
between polar desert and shrub tundra in the northern
latitudes south of the Fennoscandia ice-sheet, between
cold steppe-tundra and conifer woodland in the central
latitudes, and between a mix of grasslands and forests
and more forested yet still open vegetation in the south-
ern/Mediterranean areas (van Andel and Tzedakis, 1996).
Additional Supporting Information may be found in the online
version of this article.
Grant sponsors: Max Planck Society and the Hunt Post-Doctoral
Fellowship; Grant number: 8554 (to S.E.Z.); Grant sponsor:
National Science Foundation; Grant numbers: 0452155 (to F.E.G.
and S.E.Z.) and 0315157 (to P.S.U.); Grant sponsor: LSB Leakey
Foundation; Grant numbers: 800320 (to F.E.G. and S.E.Z.).
*Correspondence to: Sireen El Zaatari, R€
umelinstr. 23, D 72070
T€
ubingen, Germany. E-mail: sireen.el-zaatari@uni-tuebingen.de
Received 15 July 2013; accepted 12 December 2013
DOI: 10.1002/ajpa.22457
Published online 00 Month 2014 in Wiley Online Library
(wileyonlinelibrary.com).
Ó2014 WILEY PERIODICALS, INC.
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 00:00–00 (2014)

During these unstable conditions modern humans contin-
ued to adapt their technology while present in Europe
and, towards the end of MIS 3, i.e., at approximately 35
ka Cal BP, around the time of the onset of the sharp drop
in temperature ultimately leading to the LGM, the
archaeological record attests to the Gravettian, a new
techno-complex which became the second major wide-
spread Upper Paleolithic tradition in Europe (e.g., Otte,
1981; Goutas et al., 2011; Moreau, 2012). This industry is
characterized by a systematic use of abrupt retouch for
shaping bladelets and backed point tools, the presence of
specific types of tanged points and burins, and the wide
distribution of female figurines (e.g., Otte, 1981; Goutas
et al., 2011).
During MIS 2, conditions became more intense. The
Fennoscandia ice-sheet which had remained limited to
only a part of Fennoscandia during MIS 3, extended
southward, and by around 21 ka Cal BP at the peak of
the LGM it covered all of Scandinavia, the Baltic, and
most of the British Isles (COHMAP, 1988). The Alps
were also covered with glaciers while winter sea ice
extended as far south as the coast of France (COHMAP,
1988). During this time, most of Europe was generally
bare of vegetation with permafrost and polar deserts
spreading across most of the northern and central parts
of the continent (Maarleveld, 1976; Zagwijn, 1992). Vege-
tation cover in southern Europe was also sparse with
steppe taxa dominating and montane vegetation
descending to the coastal Mediterranean areas (Garcia
and Arsuaga, 2003). The extreme conditions of the LGM
had a profound effect on the human populations as well.
They led to their dispersal and, most likely, even led to
the extinction of populations in the northern latitudes
(e.g., Finlayson, 2004).
After the LGM, the climate oscillated between cold
and temperate periods until the cold climatic conditions
of the Younger Dryas (12.8–11.5 ka Cal BP) marked the
end of the Pleistocene. Even though the period extend-
ing from the end of the LGM to the onset of the Younger
Dryas witnessed fluctuations between several stadials
and interstadials, there was a general warming trend.
Open vegetation continued to prevail in most of the
European continent, yet the overall warming trend led
to the gradual replacement of tundra and steppe vegeta-
tion of northern and central Europe by spruce forests
and birch-conifer woodlands (Huntley and Birks, 1983).
With the progressive retreat of the Scandinavian ice-
sheet following the LGM and the change in vegetation
cover, humans expanded and reoccupied most of Europe.
This time period also witnessed the spread of a new cul-
ture, the Magdalenian (20–13 ka Cal BP). Compared
with all the previous hominin cultures, the Magdalenian
is more diverse. It is a period of remarkable economic
innovation, complex social interactions, and elaborate
artistic expression. The Magdalenian is characterized by
the standardization and systematization of the bladelet
technology, with small blanks, which were produced in
various ways, transformed into backed bladelets (e.g.,
Pigeot, 2004). A spectacular development of mobile art is
observed and parietal art is widely represented (e.g.,
Jochim, 2011).
Amidst the fluctuating climates, Upper Paleolithic
modern humans persisted and flourished in Europe.
This study employs occlusal molar microwear texture
analysis to examine the effects of both the changing
environmental conditions and the changing cultures on
the diets of European Upper Paleolithic humans. This
study aims to answer two questions: 1) how did modern
human populations adapt to the changing paleoecologi-
cal conditions, and 2) to what extent did their cultural
features play a role in their adaptation and, thus, sur-
vival? It is suggested that the development of the Upper
Paleolithic technological complexes was at least in part
triggered by climatic changes and that these complexes
served to aid modern humans in their adaptation to the
different climatic periods in Europe at the later part of
the Late Pleistocene (e.g., Straus, 1995; Goebel, 1999;
Hoffecker, 2005). This notion is reinforced by the broad
temporal correlation between three major climatic peri-
ods during that time, the period of millennial-scale cli-
matic fluctuations spanning the middle part of MIS 3,
the period of sharp decrease in temperature covering the
final part of MIS 3, and the period of an overall warming
trend stretching from the end of the LGM to the begin-
ning of the Younger Dryas stadial in MIS 2, and the
three main technological phases of the Upper Paleolithic,
the Aurignacian, the Gravettian, and the Magdalenian,
respectively. It should be noted that these major techno-
logical traditions of the Upper Paleolithic were geo-
graphically wide-ranging, spreading across the
European continent. It should also be noted that the var-
ious parts of this continent were affected differently by
the climatic conditions resulting in geographically dis-
tinct biomes during any single climatic period (e.g., van
Andel and Tzedakis, 1996). Thus, if indeed these techno-
logical developments aided Upper Paleolithic humans in
their survival, they would have had to have aided them
across the different local biomes in which they lived, giv-
ing them freedom from the prevailing local environmen-
tal constraints and allowing them to gain access to
various food resources. The detection among Upper Pale-
olithic human groups of significant differences in diet
that are correlated with differences in technological com-
plexes, but that are uncorrelated with differences in
local biomes, would reflect the far-reaching importance
of the various Upper Paleolithic techno-complexes and
therefore, would support such a scenario. Also, in such a
scenario, it would be expected to observe the biggest
shift in diet during the Magdalenian time since this
period witnessed the most revolutionary and profound
cultural change in comparison with the earlier periods.
If, on the other hand, it were the case that technological
innovations had not played a substantial role in the die-
tary adaptations of these hominins, then a lack of signif-
icant correlation between cultural and dietary shifts
would be expected, and instead a link between biome
type and diet might emerge reflecting significant effects
of the distinct local paleoecological conditions on the
diets of Upper Paleolithic humans. Such a link would be
an indication of environmentally driven dietary patterns
where these early humans, regardless of their cultural
association, would have had to have substantially differ-
ent diets in the different biomes in order to survive
across the European continent.
MATERIALS AND METHODS
This study was based on the analysis of high resolu-
tion dental casts of a total of 32 Upper Paleolithic adult
individuals (i.e., whose third molars had erupted or were
in the process of eruption at the time of death) from 14
European sites (Table 1). These specimens are associated
with different Upper Paleolithic technological cultures
(i.e., Aurignacian, Gravettian, or Magdalenian) and
2S. EL ZAATARI AND J.-J. HUBLIN
American Journal of Physical Anthropology

TABLE 1. List of specimens sampled in this study
Specimen
a
Tooth Location Technological complex Vegetation cover
b
Farincourt 1 RM
1
Western Europe Magdalenian (David,
1996; Joffroy and
Mouton, 1957)
Open-steppe vegetation—
Faunal data (Joffroy
and Mouton, 1946;
Joffroy and Mouton,
1956)
Saint Germain La
Rivie
`re 1970-7-6
RM
2
Western Europe Magdalenian (Blanchard
et al., 1972)
Open-steppe vegetation –
Faunal data (Blanchard
et al., 1972; Vanhaeren
and d’Errico, 2005)
Abri Pataud 1 LM
1
Western Europe Magdalenian (Movius,
1975)
Forest-steppe vegetation
– Pollen data (Donner,
1975)
Lachaud 3 RM
2
Western Europe Magdalenian (Cheynier,
1953, 1965)
Open-steppe vegetation –
Faunal data (Cheynier,
1953, 1965)
Isturitz 115 LM
2
Western Europe Gravettian (de Saint
P
erier 1930, 1936; de
Saint P
erier and de
Saint P
erier 1952)
c
Open-steppe vegetation –
Pollen data (Leroi-
Gourhan, 1959)
Doln
ıV

estonice 13 LM
1
Eastern Europe Gravettian (Svoboda,
2006)
Forest-steppe vegetation
– Floral and faunal
data (Manson et al.,
1994; Musil, 2010;
Svobodov
a, 1991; West,
2001)
Doln
ıV

estonice 15 RM
1
Doln
ıV

estonice 16 LM
2
Doln
ıV

estonice 31 RM
3
Pavlov 1 RM
2
Eastern Europe Gravettian (Klima, 1959) Forest-steppe vegetation
– Floral and faunal
data (Manson et al.,
1994; Musil, 2010;
Svobodov
a, 1991; West,
2001)
P
redmost
ı21 RM
1
Eastern Europe Gravettian (Svoboda,
2008)
Forest-steppe vegetation
– Faunal data (Musil,
2010)
Abri Labatut 1 LM3 Western Europe Gravettian (de
Sonneville-Bordes,
1960)
unavailable
Cro-Magnon 2 RM
1
Western Europe Gravettian (Henry-
Gambier, 2002)
Open-steppe vegetation –
Faunal data (Lartet,
1868)
Barma Grande 1 LM
2
Western Europe Gravettian (Formicola
et al., 2004)
Forest-steppe vegetation
– Pollen data
a
(Follieri
et al., 1998; Watts
et al., 1996)
Barma Grande 2 RM
2
Les Rois R50 #31 RM
2
Western Europe Aurignacian (Vallois,
1958)
Open-steppe vegetation –
Faunal data (Michel
et al., 2008; Mouton
and Joffroy, 1958)
Mladec
ˇ1RM
1
Eastern Europe Aurignacian (e.g., Bayer,
1925)
Forest-steppe vegetation
– Faunal data (Pacher,
2006; Svoboda, 2001)
Mladec
ˇ2LM
1
Mladec
ˇ8LM
2
Abri Blanchard 1 RM
3
Western Europe Aurignacian (Ferembach,
1958)
Open-steppe vegetation –
Faunal data, pollen
data and charcoal data
e
(Pelegrin and O’Farell,
2005)
a
The following specimens were excluded from further analyses due to postmortem taphonomic damage to their occlusal molar
surfaces: Isturitz 106 and 107, Saint-Germain-la-Riviere B1, 1970-7-1, 1970-7-2, and 1970-7-21, Doln
ıV

estonice 3, 32, and 38, Pav-
lov 2, and 3, and Barma Grande 5.
b
Detailed descriptions of the biome reconstructions are available in the Supplementary Material Section.
c
There are some indications that human remains attributed to the Gravettian context including the specimen analyzed here might
come from the Magdalenian context (Gambier, 1990).
d
Paleoecological reconstructions are unavailable from the site of Barma Grande itself. But, pollen analyses from other Italian sites
(i.e., Lago Grande di Monticchio, Lagaccione, Vico, and Valle di Castiglione) show that during that time the area was covered by a
mix of open and forested vegetation (Watts et al., 1996; Follieri, 1998). Similar conditions were also detected in the deep sea cores
from the Mediterranean (Genty et al., 2005).
e
No paleoclimatic data is available from the site of Blanchard itself. But, climatic reconstructions are available from the Aurigna-
cian layers of Abri Castanet which is very close to Abri Blanchard. The analysis of charcoal and pollen samples as well as faunal
assemblages from Abri Castanet show the prevalence of cold, open vegetation at the time (Pelegrin and O’Farell, 2005).
UPPER PALEOLITHIC MODERN HUMANS’ DIET 3
American Journal of Physical Anthropology

different paleoecological conditions (i.e., open-steppe or a
mix of forest-steppe vegetation). The dental casts were
prepared following established procedures (Teaford and
Oyen, 1989). The teeth were cleaned with cotton swabs
soaked with water and, in some cases, with acetone and
ethyl alcohol as well to remove preservatives, when pres-
ent on the teeth. Negative molds were then made using
President MicroSystemTM (Colte
`ne-Whaledent) impres-
sion material and positive casts were poured using Epo-
Tek 301 epoxy resin and hardener (Epoxy Technology).
A Sensofar PlmConfocal Imaging Profiler (Solarius
Development, Inc.) with a 1003objective was used to
examine the occlusal surfaces of the molar casts. Speci-
mens that showed postmortem taphonomic artifacts (12
in total) were excluded from further analysis (see El
Zaatari, 2010 for examples of such damage). For the
specimens that showed well preserved premortem micro-
wear free of postmortem defects (n520), four adjoining
scans covering a total area of 276 3204 lm of a crush-
ing/grinding facet of one molar per individual were
taken. Surfaces were scanned with a lateral sampling
interval of 0.18 mm and a vertical resolution of 0.005 mm.
Photosimulations and 3D images were generated using
the Solarmap Universal software (Solarius Development
Inc., Sunnyvale, CA). The resulting scans were then
characterized with Toothfrax and SFrax software (Surf-
ract) using scale sensitive fractal analysis following Scott
et al. (2005, 2006). Five variables were generated: com-
plexity (Asfc), anisotropy (epLsar), scale of maximum
complexity (Smc), textural fill volume (Tfv), and hetero-
geneity (HAsfc). Detailed descriptions of these variables
and their computations can be found in Scott et al.
(2005, 2006) and Ungar et al. (2007). Briefly, complexity
(Asfc) reflects the change in surface roughness across
different scales of observation. Anisotropy (epLsar)
reflects the orientation of wear striations. The scale of
maximum complexity (Smc) reflects the size of the
microwear features. Textural fill volume (Tfv) reflects
the geometric shape and depth of wear features. Median
values for Asfc,epLsar,Smc, and Tfv were calculated
from the four scans to produce a single value for each
variable for each tooth, and therefore, each individual.
Finally, heterogeneity (HAsfc) reflects the variability in
complexity across the surface. It should be noted that
the individual heterogeneity values used in this study
were calculated using the four scans for each tooth with-
out further splitting single scans into smaller
subregions.
For the purpose of statistical analyses, each individual
was assigned to a technological complex (Aurignacian,
Gravettian, or Magdalenian), geographic location (east-
ern or western Europe), and paleoecological category
(open-steppe vegetation or forest-steppe vegetation). Sta-
tistical analyses focused on assessing differences in the
five variables among the Upper Paleolithic populations
when the individuals were grouped based on their tech-
nological complex, geographical location, and paleoeco-
logical category. All data were first rank-transformed to
reduce the possible effects of violating assumptions asso-
ciated with parametric statistical tests (Conover and
Iman, 1981). Then, data for the five variables were com-
pared among the different groups using a multivariate
analysis of variance model (MANOVA) (Neff and Marcus,
1980). Single classification ANOVAs on each variable,
along with multiple comparisons tests, i.e., Fisher’s least
significant difference (LSD) and Tukey’s honestly signifi-
cant difference (HSD) post hoc tests, were used to deter-
mine sources of significant differences when present
(Sokal and Rohlf, 1995; Cook and Farewell, 1996).
For the interpretation of the microwear textures of the
Upper Paleolithic groups, hierarchical cluster analysis,
using Euclidean distance and complete linkage (Forte-
lius and Solounias, 2000), was conducted on the Upper
Paleolithic groups with different technological industries
as well as on four recent hunter-gatherer groups with
diverse yet known diets. The recent hunter-gatherer
groups used for this analysis are: the Fuegians whose
diet consisted almost exclusively of meat (e.g., Bridges,
1885; Yesner et al., 2003), the Chumash who relied pre-
dominantly on fish and marine mammals for their sub-
sistence, but who also supplemented their diet with
some terrestrial animals and several kinds of plants
(e.g., Timbrook, 1993; Walker, 1996), the Khoesan whose
diet consisted of a mix of animal and plant foods (e.g.,
Lee, 1979; Sealy, 2006), and the Tigara whose diet con-
sisted predominantly of meat of marine mammals conta-
minated by large amounts of sand and grit (e.g., Larsen
and Rainey, 1948; Giddings, 1967). Detailed information
about these groups and their microwear analyses can be
found in El Zaatari (2010, in press). The five microwear
variables were used in the hierarchical cluster analysis,
thus the values were first standardized using z-scores to
correct for their different scales of measurement.
RESULTS
Tables 2 and 3 presents individual raw data as well as
summary statistics for the microwear texture variables.
Example photosimulations and three-dimensional
images are illustrated in Figure 1. Tables 4 to 8 present
the results of the statistical analyses and these results
are illustrated in Figure 2. Table 9 presents summary
statistics for the comparative recent hunter-gatherer
groups. Finally, Figure 3 illustrates the results of the
hierarchical cluster analysis including the fossil and the
recent human groups.
The MANOVA results indicate significant differences
in the model among the three Upper Paleolithic groups
when the individuals are grouped based on their techno-
logical complex but not when they are grouped based on
TABLE 2. Individual data of specimens used in this study
Specimen Asfc epLsar Smc Tfv HAsfc
Farincourt 1 2.695 0.0022 0.209 13041.8 0.157
Saint Germain La
Rivie
`re 1970-7-6
3.608 0.0019 0.151 15737.8 0.157
Abri Pataud 1 3.564 0.0013 0.150 8585.5 0.058
Lachaud 3 3.295 0.0021 0.150 10514.1 0.411
Isturitz 115 1.420 0.0028 0.417 7238.5 0.062
Doln
ıV

estonice 13 1.451 0.0009 0.267 5079.1 0.092
Doln
ıV

estonice 15 1.450 0.0050 0.150 7997.9 0.110
Doln
ıV

estonice 16 1.532 0.0031 0.150 14524.1 0.154
Doln
ıV

estonice 31 1.150 0.0037 0.941 446.8 0.056
Pavlov 1 1.984 0.0011 0.267 14458.6 0.132
P
redmost
ı 21 0.745 0.0033 0.341 8600.6 0.233
Abri Labatut 1 1.923 0.0014 0.342 979.7 0.019
Cro-Magnon 2 1.868 0.0039 0.433 10170.1 0.125
Barma Grande 1 1.129 0.0025 0.341 2347.2 0.190
Barma Grande 2 0.626 0.0042 262.147 13430.9 0.417
Les Rois R50 #31 1.368 0.0020 0.150 8823.0 0.179
Mladec
ˇ1 1.405 0.0023 0.267 1677.7 0.173
Mladec
ˇ2 0.952 0.0026 0.267 7597.7 0.165
Mladec
ˇ8 1.824 0.0009 0.208 6621.3 0.237
Abri Blanchard 1 1.625 0.0022 0.267 1425.4 0.129
4S. EL ZAATARI AND J.-J. HUBLIN
American Journal of Physical Anthropology

their geographic location or paleoecological conditions
(Tables 4–6). When the specimens are grouped based on
their technological complex, individual ANOVAs show
significant variation in surface complexity (Asfc) and
scale of maximum complexity (Smc) (Table 7). No signifi-
cant differences are detected for the remaining varia-
bles, i.e., anisotropy (epLsar), textural fill volume (Tfv),
and heterogeneity (HAsfc). Post hoc tests indicate that
the Upper Paleolithic individuals from the Magdalenian
context have significantly higher mean complexity value
(Asfc) compared with those of the Gravettian and the
Aurignacian contexts (Table 8). It should be noted that
the significant differences in scale of maximum complex-
ity detected between the individuals from the Gravettian
context on the one hand and those from the Magdale-
nian and Aurignacian contexts on the other are mostly
driven by a single outlier in the Gravettian sample
which has an extremely high scale of maximum com-
plexity value. This high scale of maximum complexity
reflects the fact that the scans of this individual were
dominated by a few very large pits most likely resulting
from a single ingestion of a relatively big abrasive item.
This interpretation is supported by the low Asfc value
for this individual which indicates an otherwise non-
abrasive diet. As this appears to be an isolated event,
not much weight should be given to this elevated Smc
value for this individual. Finally, it should also be noted
that Fischer’s post hoc test show that individuals of
Magdalenian context have significantly higher mean tex-
tural fill volume (Tfv) value compared with those of
Aurignacian contexts.
The results of hierarchical cluster analysis show that
the sample of Aurignacian Upper Paleolithic humans
clusters closest to the Chumash. The Fuegians form a
second-order cluster to that containing the Aurignacian
individuals and the Chumash. A third-order cluster sep-
arates the Gravettian group from these first and second
order clusters. In a distinct cluster, the Magdalenian
group clusters closest to the Khoesan hunter-gatherers
and a second-order cluster separates the Tigara from the
Magdalenian and the Khoesan.
DISCUSSION
The link between microwear textures and diet has
been established across different extant taxa (e.g., Scott
et al., 2006, 2012; Ungar et al., 2007; El Zaatari, 2010;
El Zaatari et al., 2011). Of particular relevance to the
current study is the significant correlation between the
microwear texture variables and ethnographically docu-
mented or archeologically inferred differences in the
diets of recent hunter-gatherer populations as illustrated
by El Zaatari (2010, in press) and El Zaatari et al.
(2011). Specifically, the complexity values have been
found to be positively correlated with the levels of
ingested abrasives, whether dietary or extraneous (i.e.,
environmental abrasives such as sand, grit, etc, that get
attached to the food items as a result of food preparation
techniques) in nature. To this end, the Tigara popula-
tion, known to ingest very high amounts of environmen-
tal abrasives, has a very high complexity average value
compared with the remaining groups; whereas, the Fue-
gians whose diet was almost free of hard brittle items
has the lowest complexity mean value. The anisotropy
values on the other hand have been found to be posi-
tively correlated with the level of consumption of tough
items. This is illustrated in the high average value of
anisotropy for the Fuegian sample compared with the
rest of the modern human samples analyzed. The Fue-
gians, unlike the other groups, based their subsistence
almost exclusively on meat (a relatively tough dietary
item) that was generally free of dietary and environmen-
tal abrasives. The Khoesan who had a mixed diet which
consisted of the least percentage of meat among the
modern groups have, on average, the lowest anisotropy
values among the groups analyzed. The scale of maxi-
mum complexity was found to be correlated with the
size of the wear causing particles. The Fuegians whose
diet was almost free of small abrasive particles have the
highest scale of maximum complexity mean value
reflecting the presence of wear features resulting almost
exclusively from the tough meat diet. On the other
hand, the Khoesan and Chumash whose diets consisted
of relatively small plant food abrasives were found to
have the lowest scale of maximum complexity mean
value among the groups analyzed. Textural fill volume
values are most likely affected by two factors, the
amount and size of the abrasive particles. A high tex-
tural fill volume value would indicate a substantial con-
sumption of abrasive particles or a consumption of very
hard abrasive particles that would cause deep features
on the occlusal molar surface and vice versa. This is
TABLE 3. Summary statistics for the microwear texture variables
Group means N Asfc epLsar Smc Tfv HAsfc
By geographic location
Eastern Europe 9 Mean 1.388 0.0025 0.318 7444.9 0.150
SD 0.393 0.0014 0.242 4866.7 0.061
Western Europe 11 Mean 2.102 0.0024 24.069 8390.4 0.173
SD 1.031 0.0009 78.962 5002.2 0.131
By vegetation cover
Open-steppe vegetation 7 Mean 2.268 0.0024 0.254 9564.4 0.174
SD 0.925 0.0007 0.125 4540.1 0.111
Forest-steppe vegetation 12 Mean 1.484 0.0026 22.124 7613.9 0.168
SD 0.769 0.0013 75.587 4788.8 0.986
By technological context
Magdalenian 4 Mean 3.291 0.0019 0.165 11969.8 0.196
SD 0.421 0.0004 0.029 3104.8 0.151
Gravettian 11 Mean 1.389 0.0029 24.163 7752.1 0.145
SD 0.449 0.0013 78.931 5153.3 0.109
Aurignacian 5 Mean 1.435 0.0020 0.232 5229.0 0.177
SD 0.326 0.0007 0.052 3447.6 0.039
UPPER PALEOLITHIC MODERN HUMANS’ DIET 5
American Journal of Physical Anthropology

illustrated by the Tigara having the highest textural fill
volume mean value whereas the Fuegians having the
lowest. The heterogeneity variable is associated with the
level of individual dietary variability. Ingesting different
kinds of particles would increase the variability in
microwear texture across the tooth surface, and, thus,
Fig. 1. Two dimensional and three-dimensional representations of microwear surfaces for Upper Paleolithic individuals: A)
Lachaud 3 (Magdalenian); B) Cro-Magnon 2 (Gravettian); C) Mladec
ˇ1 (Aurignacian). Each image represents an area 138 lm3
104 lm on the original tooth surface. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
TABLE 4. MANOVA results by geographic divisions
Value FHypothesis df Error df p
Wilks’ Lambda 0.748 0.945 5.000 14.000 0.483
Pillai’s Trace 0.252 0.945 5.000 14.000 0.483
Hotelling’s
Trace
0.337 0.945 5.000 14.000 0.483
Significant pvalues (p<0.05) are represented in bold.
TABLE 5. MANOVA results by paleoecological divisions
Value FHypothesis df Error df p
Wilks’ Lambda 0.746 0.887 5.000 13.000 0.517
Pillai’s Trace 0.254 0.887 5.000 13.000 0.517
Hotelling’s
Trace
0.314 0.887 5.000 13.000 0.517
Significant pvalues (P<0.05) are represented in bold.
6S. EL ZAATARI AND J.-J. HUBLIN
American Journal of Physical Anthropology

would lead to a higher heterogeneity value as is the case
for the Khoesan who had the most dietary variability
among the groups analyzed. The Fuegians on the other
hand have the lowest heterogeneity values reflecting the
uniformity in their diet. The fact that the microwear tex-
ture variables were found to correlate well with the
known diets of recent hunter-gatherer groups allows for
the use of these variables for the examination of the
diets of Upper Paleolithic humans.
Archaeological and fossil evidence indicates that mod-
ern humans first entered Europe during MIS 3. Their
successful settlement in this continent from then on
shows that the Upper Paleolithic people did not only
manage to cope with the challenges brought about by
entering a new territory, but that they were also able to
cope with the fast changing and sometimes extreme cli-
matic conditions affecting the continent during MIS 3-2.
Marine isotope stage 3 witnessed severe short-lived cli-
matic oscillations where conditions rapidly fluctuated
between cold stadial and warm interstadial phases
(Dansgaard et al., 1993). Some of these oscillations,
namely the several sudden warming Dansgaard-
Oeschger events, were extremely abrupt, such that the
Greenland ice deposits record risings of average temper-
atures by up to 15C over a period of few decades, i.e.
within an individuals’ lifetime (Severinghaus et al.,
1998; Huber et al., 2006). The following stage, MIS 2,
included the Last Glacial Maximum, the coldest phase of
the Last Glacial Period. The changing environmental
conditions during the Upper Paleolithic Period greatly
affected food availability as food sources, whether plant
or animal, responded to climatic fluctuations. These
responses also differed spatially resulting in the forma-
tion of different biomes across the European continent
during any single major climatic period. The Upper Pale-
olithic specimens examined in this study have been
attributed to two different biomes, i.e., open-steppe or
forest-steppe, each of which would have supported differ-
ent kinds of plant and animal communities. Yet, it does
not appear to be the case that these differences in food
sources had significant effects on the diets of the Upper
Paleolithic humans as indicated by the results of the
occlusal molar microwear texture analysis. Indeed, these
results do not show any paleoecologically driven dietary
shifts for these humans suggesting that differences in
climatic conditions did not force them to significantly
alter their diets. Thus, the results of this study show
that Upper Paleolithic humans were to some extent free
from the environmental constraints. But what could
have given them this freedom?
The Upper Paleolithic modern humans had novel and
elaborate technological complexes which significantly
changed over time, such that it is now possible to
TABLE 6. MANOVA results by technological division
Value FHypothesis df Error df p
Wilks’ Lambda 0.197 3.258 10.000 26.000 0.007
Pillai’s Trace 1.098 3.405 10.000 28.000 0.005
Hotelling’s Trace 2.582 3.098 10.000 24.000 0.011
Significant pvalues (P<0.05) are represented in bold.
TABLE 7. Individual ANOVAS by technological division
Sum of squares df Mean square Fp
Surface complexity (Asfc) Effect 320.654 2 160.327 7.915 0.004
Error 344.345 17 20.256
Anisotropy (epLsar) Effect 127.131 2 63.565 2.011 0.164
Error 537.370 17 31.610
Scale of maximum complexity (Smc) Effect 272.414 2 136.207 6.005 0.011
Error 385.586 17 22.682
Textural fill volume (Tfv) Effect 155.564 2 77.782 2.596 0.104
Error 509.436 17 29.967
Heterogeneity (HAsfc) Effect 81.414 2 40.707 1.187 0.329
Error 583.086 17 34.299
Significant pvalues (<0.05) are represented in bold.
TABLE 8. Multiple comparisons tests (matrices of pairwise mean differences) by technological division
Aurignacian Gravettian
pp
Value Tukey’s Fisher’s Value Tukey’s Fisher’s
Magdalenian
Asfc 10.30 0.009 0.003 9.86 0.004 0.002
epLsar 1.63 0.903 0.672 5.85 0.205 0.092
Smc 3.35 0.558 0.309 8.93 0.013 0.005
Tfv 8.30 0.089 0.037 5.32 0.247 0.114
HAsfc 2.35 0.823 0.558 2.43 0.760 0.487
Gravettian
Asfc 0.44 0.982 0.859
epLsar 4.23 0.366 0.181
Smc 5.58 0.105 0.044
Tfv 2.98 0.581 0.327
HAsfc 4.78 0.309 0.148
Significant pvalues (<0.05) are represented in bold.
UPPER PALEOLITHIC MODERN HUMANS’ DIET 7
American Journal of Physical Anthropology

classify these complexes into several distinct industries
within the relatively short time period of the Upper
Paleolithic. If Upper Paleolithic humans were using
their tool kits to access their preferred dietary items
and, to some level, free themselves from the environ-
mental constraints, it would be expected to find evidence
for temporal dietary shifts correlated with changes in
technological complexes. Such evidence is revealed
through the microwear analyses. Specifically, the micro-
wear data show significant differences in the microwear
textures, and thus diets, between earlier and later
Upper Paleolithic humans, i.e., between those associated
with Aurignacian and Gravettian cultures on the one
hand and those associated with the Magdalenian culture
Fig. 2. Plots of the means (dots) and one standard deviation ranges (bars) for the occlusal molar microwear texture variables
for the three Upper Paleolithic groups, the Magdalenian, the Gravettian, and the Aurignacian.
8S. EL ZAATARI AND J.-J. HUBLIN
American Journal of Physical Anthropology

on the other. These differences are observed mostly in
the complexity variable which is positively correlated
with the level of abrasiveness in the diet. The lack of
significant differences across the other variables (with
the exception of the difference in scale of maximum com-
plexity which is driven by the outlier in the Gravettian
sample) indicates that the most significant difference in
the diet of the Upper Paleolithic groups pertains to the
difference in amount of abrasives in the diet with the
Magdalenian diet being significantly more abrasive than
the Aurignacian and Gravettian diets.
When compared with the recent hunter-gatherer
groups with known but diverse diets and whose micro-
wear textures have been found to be highly correlated
with their documented diets (El Zaatari, 2010; El Zaa-
tari et al., 2011), two distinct clusters reflecting the dif-
ferences in the microwear signatures of the Upper
Paleolithic groups clearly appear. The Aurignacian and
Gravettian groups cluster close to the mostly meat eat-
ing hunter-gatherer groups, the Fuegians and the Chu-
mash. It should be noted though that neither of these
Upper Paleolithic groups cluster very close to the Fue-
gians, whose diet consisted of almost exclusively meat,
with plants forming less than <15% of the total diet
(e.g., Bridges, 1885; Murdock, 1962). However, a very
tight cluster contains the Aurignacian group and the
Chumash. The diet of the Chumash consisted mostly of
fish and marine mammals, but was also supplemented
with some terrestrial animals and several kinds of
plants (e.g., Timbrook, 1993; Walker, 1996). Ethno-
graphic data show that plants formed 36 to 45% of the
diet of the mainland Chumash groups but that this per-
centage was a bit lower for the islands’ groups to which
the specimens included in this study belong (Murdock,
1964; Erlandson et al., 2009). These results indicate that
the diets of the early Upper Paleolithic modern humans
did not consist almost entirely of meat, but that they
were somewhat varied including plants, as well as possi-
bly other kinds of foods such as marine foods, birds, etc.
Archaeological evidence supports the exploitation of
plants in the earlier part of the Upper Paleolithic, i.e.,
during MIS 3. Plant remains and plant processing
equipment have been found at several Upper Paleolithic
sites dating to MIS 3 (e.g., Mason et al., 1994; Svoboda
et al., 2000; Aranguren et al., 2007; Revedin et al.,
2010). Also, evidence for the ingestion of marine foods
and other small animals such as birds and small verte-
brates by early Upper Paleolithic humans is available
from various dietary reconstruction sources (e.g., Bosin-
ski, 2000; Hahn, 2000; Svoboda et al., 2000; Stiner,
2001; Richards, 2009).
Compared with the earlier part of the Upper Paleo-
lithic, the microwear data detects a clear shift in
humans’ diet during the later part of this period. This
shift is illustrated through the close clustering of the
Magdalenian group with the Khoesan hunter-gatherers
who had a mixed diet consisting of both animal and
plant foods, with the latter forming between 60 and 80%
of their diet (Lee, 1979; Sealy, 2006). These results are
in agreement with those from various other dietary
reconstruction techniques that point to a diversification
of dietary resources, with a decreased dependence on
large mammalian herbivores and an increased focus on
small animals, birds and marine resources, as well as an
increased exploitation of plant foods, during the later
part of Upper Paleolithic compared with the earlier part
(e.g., Hansen and Renfrew, 1978; Straus, 1985; Straus
and Clark, 1986; Le Gall, 1992; Aura et al., 1998; Badal,
1998; Holliday, 1998; Badal, 2001; Stiner and Munro,
2002; Vaquero et al., 2002; Almeida et al., 2004; Finlay-
son, 2004; Haws, 2004; Haws and Valente, 2006; Manne
and Bicho, 2009; Richards and Trinkaus, 2009).
Finding evidence of significant differences in diet
between the Magdalenian group and the earlier Upper
Paleolithic groups is not surprising, but rather expected
in the light of the other notable changes in terms of
human culture, technology, social structure, and popula-
tion density during the Magdalenian period (e.g.,
Bocquet-Appel and Demars, 2000; Schwendler, 2012).
However, what appears to be counterintuitive is the
overall dietary diversification as well as the increased
reliance on plant foods by the Magdalenian people. The
Magdalenian culture developed at the end of LGM in
MIS 2 when conditions were still relatively cold and
open vegetation prevailed in many parts of Europe.
Under such conditions, plant and animal communities
would not have been as diverse as in warmer and more
wooded habitats. Yet, modern humans were expanding,
occupying new territories, and supporting higher popula-
tion densities. This shows that these humans were suc-
cessful in getting access to enough food recourses in
spite of the relatively low environmental carrying
Fig. 3. Hierarchical cluster analysis using euclidean dis-
tance and complete linkage.
TABLE 9. Microwear texture mean values for the recent human comparative samples
Group NTribe/site, area, country Asfc epLsar Smc Tfv HAsfc
Fuegians 6 Yamana Tribe, Tierra
del Fuego, Argentina
AD 1880 Mean 0.948 0.0044 0.400 5,224.8 0.109
SD 0.291 0.0014 0.135 3,522.5 0.027
Chumash 13 Santa Cruz, California, USA 4000–5000 BP Mean 2.787 0.0023 0.19 6,635.9 0.191
SD 2.344 0.0007 0.055 3,191.8 0.121
Khoesan 9 Oakhurst Shelter, South Africa 9000–5000 BP Mean 3.548 0.0020 0.176 9,272.0 0.325
SD 1.601 0.0010 0.051 6,051.2 0.208
Tigara 27 Point Hope, Alaska, USA AD 1200–1700 Mean 6.406 0.0031 0.215 12,434.1 0.286
SD 5.615 0.0016 0.070 4,946.7 0.204
Data for Fuegians, Chumash, and Khoesan are from El Zaatari (2010) and data for Tigara is from El Zaatari (in press).
UPPER PALEOLITHIC MODERN HUMANS’ DIET 9
American Journal of Physical Anthropology

capacity at the time. Thus, it seems that the cultural
boom of the Magdalenian allowed these humans to con-
quer the environmental stresses and be more in control
of their lives and survival.
CONCLUSION
The Upper Paleolithic Period which is associated with
the first presence of modern humans in Europe is
marked by a major change in hominin material culture.
This period began during the severe climatic fluctuations
of MIS 3, when climatic conditions witnessed an overall
cooling trend leading to the Last Glacial Maximum dur-
ing MIS 2. Yet, modern humans were able to survive
these extreme climatic conditions. Using occlusal molar
microwear texture analysis, this study examined the
relationships between climatic and cultural changes and
the diets of Upper Paleolithic humans in Europe. The
results of this study do not reveal any environmentally
driven dietary shifts for the Upper Paleolithic hominins.
This indicates that these humans were not forced to
alter their diets in a very significant way in response to
changes in paleoecological conditions which in turn mir-
rored climatic changes in order to survive, but that they
had some other means of adaptation to these changes.
The microwear texture analysis suggests that culture
might have given these humans such freedom. The
results of this study detect significant differences in the
diets of individuals associated with different Upper
Paleolithic cultures. More specifically, the microwear
signatures clearly distinguish specimens of Aurignacian
and Gravettian contexts from those of Magdalenian
contexts. In comparison to recent hunter-gatherer
groups with known yet diverse diets, the microwear
textures of the Aurignacian and Gravettian individuals
were found to be similar to those whose diet consisted
mostly of meat whereas the Magdalenian individuals
have microwear textures that are similar to those of
mixed diet hunter-gatherers.
ACKNOWLEDGMENTS
The authors thank the following people for making fos-
sil specimens available for this study: H. de Lumley, M-A.
de Lumley, D. Grimaud-Herv
e,A.Vialet,S.Renault,P.
Mennecier,V.Laborde,L.Huet,J.Svoboda,M.Oliva,M.
Doc
ˇkalov
a, T. Sojkova, P. Neruda, A. Arellano, P.-E.
Moull
e, A. Del Lucchase, V. Formicola, M. Teschler-
Nicola. The authors are grateful for P. Ungar for provid-
ing access to the microscope facilities at the University of
Arkansas. The authors also thank two anonymous
reviewers for their valuable comments on an earlier
version of this article.
LITERATURE CITED
Almeida F, Angelucci D, Gameiro C, Correia J, Pereira T. 2004.
Novos dados parao Paleol
ıtico Superior final da Estremadura
portuguesa: resultados preliminaresdos trabalhos arqueol
ogicos
de 1997–2003 na Lapa dos Coelhos (Casais Mar-tanes, Torres
Novas). Promontoria 2:157–192.
Aranguren B, Becattini R, Mariotti Lippi M, Revedin A. 2007.
Grinding flour in Upper Palaeolithic Europe (25 000 years
bp). Antiquity 81:845–855.
Aura JE, Jord
a Pardo JF, Gonz
alez-Tablas J, B
ecares J,
Sanchidri
an J. 1998. Secuencia Arqueol
ogica de la Cueva de
Nerja: la Sala del Vest
ıbulo. In: Sanchidri
an JL, Sim
on MD,
editors. Las culturas del Pleistoceno Superior en Andaluc
ıa.
M
alaga: Patronato de la Cueva de Nerja. p 217–236.
Badal E. 1998. El inter
es econ
omico del pino pi~
nonero para los
habitantes de la Cueva de Nerja. In: Sanchidri
an JL, Sim
on
MD, editors. Las Culturas del Pleistoceno Superior en
Andaluc
ıa. M
alaga: Patronatodel la Cueva de Nerja. p 287–300.
Badal E. 2001. La recolecci
on de pi~
nas durante la prehist
oria
en la Cueva de Nerja (M
alaga). In: Villaverde V, editor. De
Neandertales a Croma~
nones: El Inicio del Poblamiento
Humano en las Tierras Valencianas. Vale
`ncia: Universitat de
Vale
`ncia. p 101–104.
Barron E, Pollard D. 2002. High-resolution climate simulations
of Oxygen Isotope Stage 3 in Europe. Quat Int 58:296–309.
Bayer, J. 1925. Die €
altere Steinzeit in den Sudentenl€
andern.
Anstalt f€
ur Sudentendeutsche Heimatforschung (Vorge-
schichtliche Abteilung) 2:73–77.
Blades BS. 1999. Aurignacian lithic economy and early modern
human mobility: new perspectives from classic sites in the
V
eze
`re valley of France. J Hum Evol 37:91–120.
Blanchard R, Peyrony D, Vallois HV. 1972. Le gisement et le
squelette de Saint-Germain-la-Rivie
`re. Paris: Archives de
l’Institut de Pal
eontologie Humaine M
emoire 34.
Bocquet-Appel J-P, Demars PY. 2000. Neanderthal contraction
and modern human colonization of Europe. Antiquity 74:544–
552.
Bosinski G. 2000. The period 30,000–20,000 BP in the Rhine-
land. In: Roebroeks W, Mussi M, Svoboda J, Fennema K, edi-
tors. Hunters of the Golden Age: the Mid Upper Paleolithic of
Eurasia 30,000–20,000 BP. Leiden: University of Leiden.
p 271–280.
Bridges T. 1885. The Yahgans of Tierra del Fuego. J R Anthro-
pol Inst 14:288–289.
Brooks A. 1982. Aurignacian assemblages from Abri Pataud
(Dordogne, France). Aurignacien et P
erigordien en Europe.
Lie
`ge: 
Etudes et Recherches Arch
eologiques de l’Universit
e
de Lie
`ge. p 92–104.
Cheynier A. 1953. Stratigraphie de l’abri Lachaud et les cultures
des bords abattus. Archivo de Prehistoria Levantina 4:25–55.
Cheynier A. 1965. L’abri Lachaud aTerrasson (Dordogne).
Paris: Presses Universitaires de France.
COHMAP M. 1988. Climate changes of the last 18,000 years:
observations and model simulations. Science 241:1043–1052.
Conover WJ, Iman RL. 1981. Rank transformations as a bridge
between parametric and nonparametric statistics. Am Stat
35:124–129.
Cook RJ, Farewell VT. 1996. Multiplicity considerations in the
design and analysis of clinical trials. J R Stat Soc Ser A 159:
93–110.
Dansgaard W, Johnsen SJ, Clausen HB, Dahl-Jensen D,
Gundestrup NS, Hammer CU, Hvidberg CS, Steffensen JP,
Sveinbjornsdottir AE, Jouzel J, Bond GC. 1993. Evidence for
general instability of past climate from a 250 kyr ice core
record. Nature 364:218–220.
David S. 1996. La fin du Pal
eolithique sup
erieur en Franche-
Comt
e: environnement, cultures, chronologie. Gallia
pr
ehistoire 38:111–248.
de Sonneville-Bordes D. 1960. Le Pal
eolithique sup
erieur en
P
erigord. Bordeaux: Delmas.
Donner JJ. 1975. Pollen composition of the Abri Pataud sedi-
ments. In: Movius HLJ, editor. Excavation of the abri Pataud,
les Eyzies, Dordogne. Cambridge: Peabody Museum of
Archaeology and Ethnology, Harvard University. p 160–174.
El Zaatari S. 2010. Occlusal microwear texture analysis and the
diets of historical/prehistoric hunter-gatherers. Int J Osteoar-
chaeol 20:67–87.
El Zaatari S. in press. The diets of the Ipiutak and Tigara
(Point Hope, Alaska): evidence from occlusal molar microwear
texture analysis. In: Hilton C, Auerbach B, Cowgill L, editors.
The Foragers of Point Hope: the biology and archaeology of
humans on the edge of the Alaskan arctic. Cambridge: Cam-
bridge University Press.
El Zaatari S, Grine FE, Ungar PS, Hublin JJ. 2011. Ecogeo-
graphic variation in Neandertal dietary habits: evidence from
occlusal microwear texture analysis. J Hum Evol 61:411–424.
Erlandson JM, Rick TC, Braje TJ. 2009. Fishing up the food
web? 12,000 years of maritime subsistence and adaptive
10 S. EL ZAATARI AND J.-J. HUBLIN
American Journal of Physical Anthropology

adjustments on California’s Channel Islands. Pacific Sci 63:
711–724.
Ferembach D. 1958. Note sur la dent Aurignacienne trouv
ee
dans l’Abri Blanchard. Bull Mem Soc Anthropol Paris 9:199–
202.
Finlayson C. 2004. Neanderthals and modern humans: an eco-
logical and evolutionary perspective. Cambridge: Cambridge
University Press.
Follieri M, Giardini M, Magri D, Sadori L. 1998. Palynostratig-
raphy of the last glacial period in the volcanic region of Cen-
tral Italy. Quat Int 47:317–335.
Formicola V, Petitt PB, Ded Lucchese A. 2004. A direct AMS
radiocarbon date on the Barma grande 6 upper paleolithic
skeleton. Curr Anthropol 45:114–118.
Fortelius M, Solounias N. 2000. Functional characterization of
ungulate molars using the abrasion-attrition wear gradient: a
new method for reconstructing paleodiets. Am Mus Novit
3301:1–36.
Gambier D. 1990. Les vestiges humains du gisement d’Isturitz
(Pyr
en
ees-Atlandiques): 
etude anthropologique et analyse des
traces d’action humaine intentionnelle. Antiquit
es Nationales
22-23:9–26.
Garcia N, Arsuaga JL. 2003. Late Pleistocene cold-resistant fau-
nal complex: lberian occurrences. In: Blanca Ruiz Zapata M,
Dorado Valino M, Valdeolmillos Rodriguez A, Gil Garc
ıa MJ,
Bardaj
ı Azc
arate T, Bustamante Guti
errez I, Mart
ınez Men-
diz
abal I, editors. Quaternary Climatic Changes and Environ-
mental Crises in the Mediterranean Region. Madrid:
Universidad de Alcala de Henares. p 149–159.
Genty D, Combourieu-Nebout N, Hatt
e C, Blamart D, Ghaleb
B, Isabello L. 2005. Rapid climatic changes of the last 90 kyr
recorded on the European continent. C R Geosci 337:970–982.
Giddings JL. 1967. Ancient men of the Arctic. London: Secher
and Warburg.
Goebel T. 1999. Pleistocene human colonization of Siberia and
peopling of the Americas: an ecological approach. Evol
Anthropol 8:208–227.
Goutas N, Klaric L, Resesse D, Guillermin P. 2011. 
A la recher-
che des identit
es gravettiennes : actualit
es, questionnements
et perspectives: actes de la table ronde sur le Gravettien en
France et dans les pays limitrophes, Aix-en-Provence, 6–8
octobre 2008: M
emoire LII de la Soci
et
ePr

ehistorique
Franc¸aise.
Hahn J. 2000. The Gravettian in southwest Germany - environ-
ment and economy. In: Roebroeks W, Mussi M, Svoboda J,
Fennema K, editors. Hunters of the Golden Age: the Mid
Upper Paleolithic of Eurasia 30,000–20,000 BP. Leiden: Uni-
versity of Leiden. p 249–256.
Hansen J, Renfrew JM. 1978. Palaeolithic-Neolithic seed
remains at Franchthi Cave, Greece. Nature 271:349–352.
Haws J. 2004. An Iberian perspective on Upper Paleolithic
plant consumption. Promontoria 2:49–106.
Haws JA, Valente MJ. 2006. Animal carcass utilization during
the late Upper Paleolithic occupation of Lapa do Su~
ao (Portu-
gal). In: Haws JA, Brugal J-P, Hockett BS, editors. Paleolithic
Zooarchaeology in Practice. Oxford: BAR International Series.
p 29–37.
Henry-Gambier D. 2002. Les fossiles de Cro-Magnon (Les
Eyzies-de-Tayac, Dordogne): Nouvelles donn
ees sur leur posi-
tion chronologique et leur attribution culturelle. Bull Mem
Soc Anthropol Paris 14:89–112.
Hoffecker, JF. 2005. Innovation and technological knowledge in
the Upper Paleolithic of northern Eurasia. Evol Anthropol 14:
186–198.
Holliday TW. 1998. The ecological context of trapping among
recent hunter-gatherers: implications for subsistence in ter-
minal Pleistocene Europe. Curr Anthropol 39:711–719.
Huber C, Leuenberger M, Spahni R, Fluckiger J, Schwander J,
Stocker TF, Johnsen SJ, Landais A, Jouzel J. 2006. Isotope
calibrated Greenland temperature record over Marine Isotope
Stage 3 and its relation to CH4. Earth Planet Sci Lett 243:
504–519.
Hublin J-J. 2012. The earliest modern human colonization of
Europe. Proc Natl Acad Sci USA 109:13471–13472.
Hublin J-J. 2013 The Makers of the Early Upper Paleolithic in
Western Eurasia. In: Smith FH, Ahern JCM, editors. Origin
of modern humans: biology reconsidered. Hoboken: Wiley-
Blackwell. p 223–252.
Huntley B, Birks HJB. 1983. An atlas of past and present pol-
len maps of Europe. Cambridge: Cambridge University Press.
Jochim M. 2011. The Upper Paleolithic. In: Milisauskas S, edi-
tor. European prehistory: a survey. New York: Springer Scien-
ce 1Business Media. p 67–124.
Joffroy R, Mouton AP. 1946. La station magdal
enienne de Far-
incourt (H.-M.). Bulletin de la Soci
et
ePr

ehistorique Franc¸aise
43:91–100.
Joffroy R, Mouton AP. 1956. Pr
ecisions nouvelles sur les sta-
tions magdal
eniennes de Farincourt (Haute-Marne). Revue
arch
eologique de l’Est et du Centre-Est 7:193–223.
Joffroy R, Mouton AP. 1957. 
Etude des vestiges osseux humains
des grottes pr
ehistoriques de Farincourt (Haute-Marne,
France). Arch Suisses Anthropol Gen 22:6–37.
Klein R. 2009. The Human Career: Human Biological and Cul-
tural Origins. Chicago: University of Chicago Press.
Klima B. 1959. Zur problematik des Aurignacian und Gravet-
tian in Mittel-Europa. Archaeologia Austriaca 26:35–51.
Larsen H, Rainey FG. 1948. Ipiutak and the Arctic whale hunt-
ing culture. New York: Anthropological Papers of the Ameri-
can Museum of Natural History.
Lartet L. 1868. Une s
epulture des Troglodytes du P
erigord
(cr^
anes des Eyzies). Bull Mem Soc Anthropol Paris 3:335–
349.
Le Gall O. 1992. Les Magdaleniens et l’ichtyofaune
dulc¸aquicole. In: Rigaud J-P, Laville H, Vandermeersch B,
editors. Le Peuplement Magdal
enien. Paris: Editions du
CTHS. p 277–285.
Lee RB. 1979. The! Kung San. Men, women and work in a for-
aging society. Cambridge: Cambridge University Press.
Leroi-Gourhan A. 1959. R
esultats de l’analyse pollinique de la
grotte d’Isturitz. Bulletin de la Soci
et
ePr

ehistorique de
France 56:619–624.
Liolios D. 2006. Reflections on the role of bone tools in the defi-
nition of the Early Aurignacian. In: bar-Yosef O, Zilh~
ao J, edi-
tors. Towards a definition of the Aurignacian: Proceedings of
the Symposium held in Lisbon, Portugal, June 25–30, 2002.
Lisbon: Instituto Portugu^
es de Arqueologia. p 37–51.
Maarleveld GC. 1976. Periglacial phenomena and the mean
annual temperature during the last glacial time in the Neth-
erlands. Biuletyn Peryglacjalny 26:57–78.
Manne T, Bicho N. 2009. Vale Boi: rendering new understand-
ings of resource intensification and diversification in south-
western Iberia. Before Farm 2, article 1.
Manson SLR, Hather JG, Hillman GC. 1994. Preliminary inves-
tigation of the plant macroremains from Dolni Vestonice II
and its implications for the role of plant foods in Paleolithic
and Mesolithic Europe. Antiquity 68:48–57.
Mason S, Hather J, Hillman G. 1994. Preliminary investiga-
tions of the plant macro-remains from Dolni Vestonice II and
its implications for the role of plant foods in Palaeolithic and
Mesolithic Europe. Antiquity 68:48–57.
Michel A, d’Errico F, Lenoble A, Vanhaeren M, Ramirez Rozzi
F, Grootes P, Renou S, Rendu W, Daulny L, Verna C, Guy R.
2008. Nouvelles fouilles sur le site aurignacien Chez les Rois
(Mouthiers-sur-Boeme, Charente). In: Jaubert J, Bordes J-G,
Ortega I, editors. Les soci
et
es du pal
eolithique dans un Grand
Sud-Ouest de la France : nouveaux gisements, nouveaux
r
esultats, nouvelles m
ethodes: M
emoire XLVII de la Soci
et
e
pr
ehistorique franc¸aise. p 289–299.
Moreau L. 2012. The early Gravettian of Central Europe: work-
ing out and perspectives. L’Anthropologie 116:609–638.
Mouton AP, Joffroy R. 1958. Le Gisement Aurignacien des Rois
a Mouthiers (Charente). Paris: Centre National de la Recher-
che Scientifique.
Movius HLJ. 1975. Summary of the stratigraphic sequence. In:
Movius HLJ, editor. Excavation of the Abri Pataud Les Eyzies
(Dordogne). Cambridge: Peabody Museum of Archaeology and
Ethnology, Harvard University. p 7–18.
Murdock GP. 1962. Ethnographic atlas. Ethnology 1:113–134.
UPPER PALEOLITHIC MODERN HUMANS’ DIET 11
American Journal of Physical Anthropology

Murdock GP. 1964. Ethnographic atlas. Ethnology 3:199–217.
Musil R. 2010. Palaeoenvironment at Gravettian sites in central
Europe with emphasis on Moravia (Czech Republic). Quart€
ar
57:95–123.
Otte M. 1981. Le gravettien en Europe centrale. Bruge: Disser-
tationes Archaeologicae Gandenses de Tempel.
Pacher M. 2006. Large mammal remains from the Mladec
ˇ
Caves and their contribution to site formation processes. In:
Teschler-Nicola M, editor. Early modern humans at the Mora-
vian Gate: the Mladec
ˇcaves and their remains. Wien:
Springer-Verlag. p 99–148.
Pelegrin J, O’Farell M. 2005. Les lamelles retouch
ees ou uti-
lis
ees de Castanet. In: Le Brun-Ricalens F, editor. Productions
lamellaires attribu
ees 
a l’Aurignacien : cha^
ınes op
eratoires et
perspectives technoculturelles. Luxembourg: Mus
ee National
d’Histoire et d’Art. p 103–121.
Pigeot N, editor. 2004. Les derniers magdal
eniens d’Etiolles:
perspectives culturelles et pal
eohistoriques (l’unit
e d’habita-
tion Q31). Paris: CNRS.
Revedin A, Aranguren B, Becattini R, Longo L, Marconi E,
Mariotti Lippi M, Skakun N, Sinitsyn A, Spiridonova E,
Svoboda J. 2010. Thirty thousand-year-old evidence of plant
food processing. Proc Natl Acad Sci USA 107:18815–18819.
Richards M, Trinkaus E. 2009. Isotopic evidence for the diets of
European Neanderthals and early modern humans. Proc Natl
Acad Sci USA 106:16034–16039.
Richards MP. 2009. Stable Isotope Evidence for European
Upper Paleolithic Human Diets. In: Hublin JJ, Richards MP,
editors. The evolution of hominin diets: integrating
approaches to the study of Palaeolithic subsistence. Dor-
drecht: Springer Science 1Business Media B.V. p 251–257.
Schwendler RH. 2012. Diversity in social organization across
Magdalenian Western Europe ca. 17-12,000 BP. Quat Int 272-
273:333–353.
Scott RS, Ungar PS, Bergstrom TS, Brown CA, Childs BE,
Teaford MF, Walker A. 2006. Dental microwear texture analy-
sis: technical considerations. J Hum Evol 51:339–349.
Scott RS, Ungar PS, Bergstrom TS, Brown CA, Grine FE,
Teaford MF, Walker A. 2005. Dental microwear texture analy-
sis shows within-species diet variability in fossil hominins.
Nature 436:693–695.
Sealy JC. 2006. Diet, mobility, and settlement pattern among
Holocene hunter-gatherers in Southernmost Africa. Curr
Anthropol 47:569–595.
Severinghaus JP, Sowers T, Brook EJ, Alley RB, Bender ML.
1998. Timing of abrupt climate change at the end of the
Younger Dryas interval from thermally fractionated gases in
polar ice. Nature 391:141–146.
Shackleton NJ. 1977. The oxygen isotope stratigraphic record of
the Late Pleistocene. Philos Trans R Soc Lond B Biol Sci 280:
169–182.
Sokal RR, Rohlf FJ. 1995. Biometry. New York: W.H. Freeman
and Company.
Stiner MC. 2001. Thirty years on the “broad spectrum revolu-
tion” and Palaeolithic demography. Proc Natl Acad Sci USA
98:6993–6996.
Stiner MC, Munro ND. 2002. Approaches to prehistoric diet
breadth, demography, and prey ranking systems in time and
space. J Archaeol Method Th 9:181–214.
Straus LG. 1985. Stone age prehistory of northern Spain. Sci-
ence 230:501–507.
Straus LG. 1995. The Upper Paleolithic of Europe: an overview.
Evol Anthropol 4:4–16.
Straus LG, Clark G. 1986. La Riera Cave. Tempe: Arizona State
University.
Svoboda J. 2001. Mladec
ˇand other caves in the middle Danube
region: early modern humans, late Neandertals, and projec-
tiles. In: Zilh~
ao, J, Aubry, T, Carvalho, F, editors. Les pre-
miers hommes modernes de la P
eninsule Ib
erique. Actes du
colloque de la Commission VIII de l’UISPP. Lisbon: Instituto
Portugue
`s de Arqueologia. p 45–60.
Svoboda J. 2006. The archaeological framework. In: Trinkaus E,
Svoboda J, editors. Early modern human evolution in Central
Europe. The people of Doln
ıV

estonice and Pavlov. New York:
Oxford University Press. p 6–8.
Svoboda J, Klima B, Jaro
sov
aL, 
Skrdla P. 2000. The Gravettian
in Moravia: climate, behavior and technological complexity.
In: Roebroeks W, Mussi M, Svoboda J, Fennema K, editors.
Hunters of the Golden Age: the Mid Uppe Paleolithic of Eura-
sia 30,000–20,000 BP. Leiden: University of Leiden. p 197–
217.
Svoboda JA. 2008. The Upper Paleolithic burial area at
Pr
edmost
ı: ritual and taphonomy. J Hum Evol 54:15–33.
Svoboda JA, Bar-Yosef O. 2003. Str
ansk
ask

ala. Origins of the
Upper Paleolithic in the Brno Basin, Moravia, Czech Repub-
lic. Cambridge: Peabody Museum of Archaeology and Ethnol-
ogy, Harvard University.
Svobodov
a H. 1991. Pollen analysis of Upper Palaeolithic triple
burial at Dolni Vestonice. Archeologick
e rozhledy 43:505–510.
Tartar E, White R. 2013. The manufacture of Aurignacian split-
based points: an experimental challenge. J Archaeol Sci 40:
2723–2745.
Teaford MF, Oyen OJ. 1989. Live primates and dental replica-
tion: new problems and new techniques. Am J Phys Anthro-
pol 80:73–81.
Timbrook J. 1993. Island Chumash ethnobotany. In: Glassow
MA, editor. Archaeology on the Northern Channel Islands of
California: studies of subsistence, economics and social orga-
nization. Salinas: Coyote Press. p 47–62.
Ungar PS, Merceron G, Scott RS. 2007. Dental microwear tex-
ture analysis of Varswater bovids and early Pliocene paleoen-
vironments of Langebaanweg, Western Cape Province, South
Africa. J Mamm Evol 14:163–181.
Vallois HV. 1958. Les restes humains d’^
age aurignacien de la
grotte des Rois, Charente. Bull Mem Soc Anthropol Paris 9:
138–159.
van Andel T, Tzedakis PC. 1996. Paleolithic landscapes of
Europe and environs 150,000–25,000 years ago: an overview.
Quat Sci Rev 15:481–500.
Vanhaeren M, d’Errico F. 2006. Aurignacian ethno-linguistic
geography of Europe revealed by personal ornaments.
J Archaeol Sci 33:1105–1128.
Vanhaeren M, d’Errico F. 2005. Grave goods from the Saint-
Germain-la-Rivie
`re burial: Evidence for social inequality in
the Upper Palaeolithic. J Anthropol Archaeol 24:117–134.
Vaquero M, Esteban M, Allue E, Vallverdu J, Carbonell E,
Bischoff JL. 2002. Middle Palaeolithic refugium, or archaeo-
logical misconception? A new U-Series and radiocarbon chro-
nology of Abric Agut (Capellades, Spain). J Archaeol Sci 29:
953–958.
Walker PL. 1996. Integrative approaches to the study of ancient
health: an example from the Santa Barbara channel area of
Southern California. In: P
erez-P
erez A, editor. Notes on popu-
lation significance of paleopathological conditions: health, ill-
ness and death in the past. Barcelona: Fundaci
o Uriach.
p 97–105.
Watts WA, Allen JRM, Huntley B. 1996. Vegetation history and
palaeoclimate of the last glacial period at Lago Grande di
Monticchio southern Italy. Quat Sci Rev 15:133–153.
West D. 2001. Analysis of the fauna recovered from the 1986/
1987 excavations at Dolni Vestonice II, Western Slope.
Pam
atky Archeologick
e 92:98–123.
White R. 1993. Technological and social dimensions of
‘‘Aurignacian-age’’ body ornaments across Europe. In: Knecht
H, Pike-Tay A, White R, editors. Before Lascaux: the complex
record of the Early Upper Paleolithic. Boca Raton: CRC Press.
p 277–299.
Yesner DR, Figuerero Torres MJ, Guichon RA, Borrero LA.
2003. Stable isotope analysis of human bone and ethnohis-
toric subsistence patterns in Tierra del Fuego. J Anthropol
Archaeol 22:279–291.
Zagwijn W. 1992. Migration of vegetation during the quaternary
in Europe In: von Koenigswald W, Werdelin L, editors. Mam-
malian migration and dispersal events in the European Qua-
ternary. Frankfurt: Courier Forschungsinstitut Senckenberg.
p 9–20.
12 S. EL ZAATARI AND J.-J. HUBLIN
American Journal of Physical Anthropology