Palaeoecological implications of the Lower Pleistocene phytolith

Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
Contents lists available at ScienceDirect
Palaeogeography, Palaeoclimatology, Palaeoecology
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p a l a e o
Palaeoecological implications of the Lower Pleistocene phytolith record from the
Dmanisi Site (Georgia)
E. Messager a,b,⁎, D. Lordkipanidze b, C. Delhon c, C.R. Ferring d
a
UMR 7194 CNRS Département de Préhistoire du Muséum national d'Histoire naturelle, 1, rue René Panhard, 75013 Paris, France
Georgian National Museum, 3, Rustaveli Avenue 0105 Tbilisi, Georgia
UMR 6130 CNRS CEPAM, 250 rue Albert Einstein, Sophia Antipolis, F-06650 Valbonne, France
d
Department of Geography, University of North Texas, Denton, TX, 76203-3078, USA
b
c
a r t i c l e
i n f o
Article history:
Received 18 March 2009
Received in revised form 15 January 2010
Accepted 17 January 2010
Available online 25 January 2010
Keywords:
Phytolith analysis
Dmanisi archaeological site
Water stress indices
Palaeoenvironment
Pleistocene
Caucasus
a b s t r a c t
Archaeological investigations of the lower Pleistocene deposits at Dmanisi (Lesser Caucasus, Georgia) have
yielded an assemblage of hominin and faunal remains within a well-dated context. Although abundant
vertebrate fossils have been recovered, paleobotanical studies have been limited. To address this, phytolith
analysis has been conducted on two sections in order to reconstruct the distribution and evolution of
vegetation throughout the entire sedimentary sequence. Large concentrations of phytoliths were recovered
and analysed, permitting the reconstruction of climatic indices. The environmental data obtained from these
phytolith assemblages are consistent with other palaeoecological data (i.e. geological, faunal and other
archaeobotanical records). When considered together, they indicate an environment in which grasses were
well-represented. In addition, the climatically important water stress indices derived from Dmanisi's
phytolith assemblages suggest a period of increased aridity in the middle part of the stratigraphic sequence,
which is contemporaneous with human occupations of the site.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
Discoveries of Early Pleistocene hominid remains Eurasia provide
evidence for an expansion out of Africa earlier than previously
assumed. Present evidence suggests that these dispersals occurred
several times since the beginning of the Pleistocene. New data are
refining the chronological framework in which these first Eurasian
occupations took place (Swisher et al., 1994; Oms et al., 2000; Sémah
et al., 2000; Voinchet et al., 2004; Falguères, 2003; Carbonell et al.,
2008). Although environmental data from vertebrate faunas are
available, data from archaeobotanical sources remain scarce. The first
dispersals out of Africa appear to have taken place between 1.9 and
1.5 million years ago (Bar-Yosef, 1994; Swisher et al., 1994; Gabunia
and Vekua, 1995; Arribas and Palmqvist, 1999; Bar-Yosef and BelferCohen, 2001). At this early stage in hominin evolution, populations
probably possessed the biocultural means to adapt their behaviour to
new environments (O'Connell et al., 1999; Antón et al., 2002).
However, as with most other mammalian taxa, early hominins were
likely to have been affected by African terminal Pliocene climatic and
environmental changes (Potts, 1998; Zeitoun, 2000; Holmes, 2007;
Bonnefille, 1995; DeMenocal and Bloemendal, 1995; Vrba, 1995; Bobe
et al., 2002; Bobe and Behrensmeyer, 2004), which also affected
⁎ Corresponding author. UMR 7194 CNRS Département de Préhistoire du Muséum
national d'Histoire naturelle, 1, rue René Panhard, 75013 Paris, France.
E-mail address: [email protected] (E. Messager).
0031-0182/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.palaeo.2010.01.020
Eurasia (Sémah, 1986; Combourieu-Nebout, 1990; Zagwijn, 1992;
Combourieu-Nebout, 1993; Suc et al., 1997; Aguirre and Carbonell,
2001).
The site of Dmanisi, situated in present-day Georgia, has yielded
abundant fossils of early hominids. It is now well-dated to the very
beginning of the Upper Matuyama Chron (1.77 Ma), making the site a
very good opportunity to undertake a palaeoenvironmental study.
Most organic bio-proxies (i.e. spores, pollen grains, fruits and
seeds) deposited in dry sedimentary environments and subjected to
weathering are not usually well-preserved. As a result, the identification of climatic changes in these types of terrestrial sequences is
often very difficult, if not impossible. On the other hand, the resistant
composition of phytoliths (hydrated silicon dioxide opal) has been
shown to be less affected by weathering processes. Phytoliths can be
recovered in large concentrations and have been used in reconstruction of palaeoclimate and palaeoenvironments for a variety of
sediments, which include loess (Tungsheng et al., 1996; Blinnikov
et al., 2002; Lu et al., 2007), lacustrine sediments (Thorn, 2004), sand
dunes (Horrocks et al., 2000), archaeological sediments (Albert et al.,
1999; Piperno et al., 2000), palaeosols (Fredlund and Tieszen, 1997a;
Delhon et al., 2003) and marine sediments (Abrantes, 2003). Their
characteristic morphology can be used as a diagnostic tool to identify
the plants that produced them. For instance, plants of the family
Poaceae are known to contain large quantities of phytoliths in their
tissues (Hodson et al., 2005), and these exhibit significant morphological variability among the several Poaceae subfamilies. Therefore,
2
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
this proxy is particularly useful for grassland studies (Fredlund and
Tieszen, 1997a,b; Blinnikov et al., 2002; Strömberg, 2002, 2004).
Other plant groups which tend to yield large concentrations of
phytoliths include the Cyperaceae (Le Cohu, 1973) and Arecaceae
(Runge, 1999) families from the Monocotyledonous group as well as
plants of Dicotyledonous taxa, such as certain tropical trees (Piperno,
1988; Hodson et al., 2005). As a result, the analysis of phytoliths is
frequently applied to reconstruct the histories of tropical ecosystems
(Alexandre et al., 1997; Barboni et al., 1999; Runge, 1999). In tropical
regions, this type of analysis provides a good tree cover proxy for the
reconstruction of paleoenvironments. Phytolith analysis has also been
successfully applied to reconstructing palaeoenvironments in temperate climatic contexts (Verdin et al., 2001; Delhon et al., 2003;
Strömberg et al., 2007).
In this study, we used phytolith assemblages to reconstruct the
environmental and climatic conditions which prevailed during the
deposition of the Dmanisi archaeological deposits. We propose that
the study of phytoliths should be used in combination with water
stress indices for palaeoenvironmental reconstructions, in addition to
other archaeobotanical data.
Fig. 2. Schematic stratigraphic section of Block 2. × vertical exaggeration.
Modified from Lordkipanidze et al., 2007.
2. Study area and present environmental conditions
3. Geology
Dmanisi (44° 21′E, 41° 19′N) is located at an elevation of 1015 m in
the Masavera River valley (South East Georgia), in the Lesser
Caucasus, 85 km southwest of Tbilisi (Fig. 1). The Dmanisi region
receives less precipitation than the Colchis region of western Georgia
where the proximity of the Black Sea supports the present-day
subtropical forests (Nakhutsrishvili, 1999; Denk et al., 2001; Volodicheva, 2002). In contrast to the Colchis lowlands, the Dmanisi region
does not constitute a favorable refuge for relict floras as a result of its
cooler and drier continental climate, which is accentuated by the
orographic effect of nearby mountain chains (Volodicheva, 2002). The
elevation of the site favors oak (Quercus iberica), and oak-hornbeam
forests (Q. iberica, Carpinus caucasica and Carpinus orientalis). In
humid contexts, however, Fagus orientalis (Beech) can sometimes
appear on hilltop forests. Human activities in the vicinity of the site
have significantly altered the vegetation of the surrounding area.
The archaeological site is situated on a promontory at the confluence of the Masavera and Pinezauri rivers, whose valleys were
eroded into the local Cretaceous volcaniclastic and marine rocks,
which form the hills surrounding the site. Latest Pliocene basaltic
eruptions west of Dmanisi resulted in a flood of lavas down the
Masavera Valley, which filled the valley, covered the lower part of the
promontory, and spilled for a short distance up the Pinezauri Valley,
creating a dam of that stream. The lavas cooled to form the Masavera
Basalt, dated by 40Ar/39Ar to 1.85±0.01 Ma (Gabunia et al., 2000a,b);
this basalt is conformably overlain by the hominin and artifactbearing Plio-Pleistocene volcaniclastic and colluvial deposits at the
site (Fig. 2).
Dmanisi's sediments have been divided into two main stratigraphic units; Stratum A directly overlies the Masavera Basalt, and has
been subdivided into four major substrata (A1–A4). A minor erosional
Fig. 1. Location of the Dmanisi site in Georgia.
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
disconformity separates Stratum A from Stratum B, which is subdivided into substrata B1–B5.
The most complete and thickest stratigraphic exposure is in the
recently excavated Unit M5, located 75 m west of Block 2 archaeological excavations (Fig. 3). In M5, the contacts between strata are
either conformable or marked by minor erosion (Ferring et al., 2008).
With the exception of colluvial clasts in the lower part of Stratum B2,
all of the deposits in M5 appear to be primary ash falls, separated by
varying but overall brief periods of pedogenesis and/or minor erosion.
The Block 2 excavation area has a complex sedimentary section that
is 2–3 m thick (Fig. 2). Located down slope from M5, the Block 2 deposits
were subjected to piping and gullying of Stratum A deposits at the
initiation Stratum B1 deposition (Fig. 2). Rapid burial by low energy
eolian and slope processes led to the superb stratification and
preservation of bones in the pipe-gully facies of Stratum B1, which
contain all of the hominin remains recovered thus far in the excavations
in Blocks 1 and 2.
The geochronology of the Dmanisi deposits has been established by
absolute dating, paleomagnetic analyses and biostratigraphic correlations. The Masavera Basalt and all of Stratum A deposits yield normal
geomagnetic polarity, and are dated to 1.85–1.77 Ma, corresponding to
the end of the Olduvai subchron. The 1.81 Ma 40Ar/39Ar age on Stratum
A1a ashes (Lumley et al., 2002) is consistent with this dating
3
interpretation. All of the Stratum B deposits at Dmanisi show reversed
geomagnetic polarity, and correspond to the Upper Matuyama Chron.
The minor disconformity at the A–B contact, as well as microfaunal
assemblages indicate that these deposits and the associated archaeological, hominin and megafaunal remains contained within these
levels accumulated very quickly after the Olduvai-Matuyama reversal
(Lordkipanidze et al., 2007). This interpretation is supported by the
1.76 Ma 40Ar/39Ar age on the Orozmani Basalt, which overlies the
Dmanisi sediments at a locality west of the site (Gabunia et al., 2000a).
All of the strata at Dmanisi exhibit pedogenic features, with the
exception of the rapidly deposited pipe-gully fills exposed in Blocks 1
and 2. In Stratum A sediments, soil development is minimal, with
pedogenic carbonate filaments and clay linings of pores. Soil
development in Strata B2–B5 was controlled by apparently reduced
rates of deposition, and is registered by cambic and/or argillic soil
horizons in Strata B2 and B5, and by significantly more pedogenic
carbonate than seen in Stratum A. Of additional significance is a
diagenetic laminar carbonate horizon (also known as “Kerki” of
Dzaparidze et al., 1989, 2002), which occurs across the site near the
A–B boundary, but clearly cross-cuts that contact. This indurated
horizon appears to have functioned as an aquiclude, helping to
preserve faunal and hominin remains which underlie this horizon in
Blocks 1 and 2.
Fig. 3. Synthetic stratigraphic sequences of M5 and Block 2 with location of samples indicated.
4
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
4. Archaeological data
The archaeological deposits at Dmanisi have yielded large amounts
of lithics and faunal remains associated with sediments from Stratum
B. Their provenience suggests that they accumulated during the
beginning of the Upper Matuyama Chron (Gabunia et al., 2000a;
Rightmire et al., 2006; Lordkipanidze et al., 2007). The first hominin
specimen was discovered in 1991 and since then, over 40 hominin
remains including mandibles, skulls and postcranial elements have
been recovered. The hominins from Dmanisi possess several craniofacial characteristics similar to those of Homo habilis, although this
population appears to be closer to the stem from which Homo erectus
may have evolved (Rightmire et al., 2006). These early Homo fossils
support hominin dispersal out of Africa at the beginning of the
Pleistocene. Furthermore, the hominin remains at Dmanisi are in
direct association with faunal remains (macro- and micromammals)
and numerous lithic artefacts. The mammal remains represent
Eurasian taxa (Lordkipanidze et al., 2007) and suggest that a mosaic
environment consisting of open steppe and gallery forests existed
during the occupation of Dmanisi (Gabunia et al., 2000b).
5. Materials and methods
5.1. Sample collecting
Sampling for phytolith analysis was done at two correlated
sections: Block 2 and M5. Because of their lower position on the
promontory slope, deposits in Block 2 reveal more erosion between
strata than seen in M5. In Block 2, eight samples were carefully
collected from each of the following strata: A1a, A2, A3, B1, B2, B3 and
B4. As mentioned previously, the M5 sequence has not been subject to
any major disturbance such as the piping-gullying observed in the
main excavation areas. Since the deposits there are horizontal and in
situ, warranting continuous sampling, and 27 samples were collected
from the base to 1.40 m below the present-day surface (Fig. 3).
5.2. Phytolith extraction, counting and classification
Phytoliths were extracted from sediment samples using a method
adapted from techniques described by Lentfer and Boyd (1998).
Carbonates were dissolved in an HCl bath and organic matter was
removed with a 15% H2O2 solution heated at 60 °C until reaction
ceased. A 180 µm mesh sieve was then used to recover fine particles.
For deflocculation and clay removal, sediment residue was shaken
with a 15% sodium hexametaphosphate solution for 15 min. Most of
the clay particles were removed through several decantations.
Densimetric separation of phytoliths (d < 2.3) from the quartz and
other mineral particles was achieved using a heavy liquid solution
(either ZnBr2 or Sodium polytungstate, both with d = 2.35). After
cleaning, the residue was suspended in a glycerine solution for
mounting on glass slides. The slides were observed under a “Zeiss
standard™” Microscope at 600× magnification. Each phytolith was
classified according to its morphology, following several systems
(Twiss et al., 1969; Mulholland, 1989; Fredlund and Tieszen, 1994)
and the International Code for Phytolith Nomenclature (ICPN
Working Group et al., 2005). Where possible, more than 300
individual phytoliths were counted per sample. The observed types
of phytoliths are classified into 13 different categories (Fig. 4; Table 1).
Since one phytolith type can be found in different plant taxa
(redundancy) and one plant taxon may produce several different
phytolith types (multiplicity) (Rovner, 1971; Rovner, 1983; Brown,
1984; Mulholland et al., 1988; Mulholland, 1989), it is difficult to
designate a specific taxonomical classification for each phytolith form.
However, recognisable morphotypes may be attributed to certain
Fig. 4. Photographs of phytolith morphotypes identified in Dmanisi sequences: a. elongate, b. acicular, c. bulliform, d. short acicular, e. bilobate, f. polylobate trapeziform, g. saddle, h.
rondel-trapeziform short cell, i. sinuate trapeziform, j. globular, k. point-hair, l. cylindric sulcate, and m. globular echinate.
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
5
Table 1
Phytolith morphotypes identified in Dmanisi sequences with their taxonomic attributions.
Morphotypes
Main taxonomic attribution
Bibliography
Elongate
Acicular
Short acicular
Bulliform, (cuneiform and parallepipedal)
Rondel and trapeziform short cell
Sinuate trapeziform
Bilobate
Polylobate trapeziform
Saddle
Globular (smooth and granulate)
Cylindric sulcate
Globular echinate
Point-hair
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae, Pooideae
Poaceae, Pooideae
Poaceae, Panicoideae
Poaceae
Poaceae, Chloridoideae
cf. Dicotyledonous
cf. Dicotyledonous
cf. Arecaceae
No taxonomic value
Twiss et al., 1969
Prat, 1932; H.-Y Lu et al., 2006
H.-Y Lu et al., 2006; Kaplan et al., 1992
Twiss et al., 1969; Twiss, 1992
Twiss et al., 1969; Fredlund and Tieszen, 1994; Mulholland, 1989; Rosen, 2001
Twiss et al., 1969; Fredlund and Tieszen, 1994; Brémond et al., 2004
Twiss et al., 1969; Brown, 1984; Mulholland, 1989; Fredlund and Tieszen, 1994; Lu and Liu, 2003
Twiss et al., 1969
Twiss et al., 1969; Mulholland, 1989; Fredlund and Tieszen, 1994
Bozarth, 1992; Alexandre et al., 1997; Albert et al., 1999; Runge, 1999; Delhon et al., 2003
Bozarth, 1992; Delhon, 2005; Borba-Roschel et al., 2006
Piperno, 1988; Runge, 1999
Piperno, 1985, 1989; Pearsall, 2000; Bozarth, 1992
grass and non-grass taxa (Table 1). The following 13 categories were
used: (1) elongate, (2) acicular, (3) short acicular and (4) bulliform
phytoliths are essentially formed in the epidermal long cells of grasses
(Twiss et al., 1969; Piperno, 1988; Mulholland, 1989; Fredlund and
Tieszen, 1994) but they can also be produced by other groups
(Piperno, 1988; Strömberg, 2002). The class short acicular could be
included into the common class acicular. However, significant
differences in both morphology and frequencies were observed and
led us to create a specific class, which could be useful in future
investigations. Further categories are: (5) rondel, (6) sinuate trapeziform, (7) bilobate, (8) polylobate and (9) saddle phytoliths, which are
produced exclusively in epidermal short cells of grasses. They can be
used to identify the main Poaceae subfamilies recorded in the
phytolith assemblage (Twiss et al., 1969; Fredlund and Tieszen,
1994). The rondel and sinuate morphotypes occur dominantly in the
subfamily Pooideae (Table 1). The bilobate morphotypes occur
dominantly in the subfamily Panicoideae. The saddle morphotype is
produced in high proportion by the Chloridoideae. (10) Globular
phytoliths correspond to the various circular and spheroid morphotypes already recognized and considered as characteristic of Dicotyledonous group (Bozarth, 1992; Alexandre et al., 1997; Albert et al.,
1999; Runge, 1999; Delhon et al., 2003). The last three categories are:
(11) cylindric sulcate which is usually assigned to herbaceous or
woody dicotyledonous but can also be produced in small amounts by
conifers and ferns (Piperno, 1988; Runge, 1999); (12) “point-hair”
includes atypical trichome phytoliths, but has no taxonomic value
because it can be produced by grasses and non-grasses; and finally,
(13) globular echinate morphologies could be assigned to the Palmae
(Piperno, 1988; Runge, 1999) although this form is not the typical
globular echinate associated with this group. The relative abundance
(%) of each morphotype for every sample is diagramatically represented to trace changes in assemblages through time. 95% confidence
intervals (using the total count of each sample) are indicated to
support the robustness of the phytolith assemblage interpretation
(Piperno, 2006; Barboni et al., 2007; Strömberg, 2009) especially for
morphotypes which in some cases have low proportions: bulliform,
bilobate, polylobate, saddle, and globular.
evapotranspiration (Brémond et al., 2005b; Delhon, 2005, 2007;
Strömberg et al., 2007). With the exception of grass short cells, most
silica opal precipitation occurring in plants tissues is linked to
evapotranspiration processes (Hutton and Norrish, 1974; Motumora
et al., 2000; Webb and Longstaffe, 2002). Water flowing from the
roots up to the plant leaves is induced by evaporation, which triggers
phytolith production in the tissues of the epidermis. Silicification of
long cells, like the bulliform cells, occurs as a result of evapotranspiration (Madella, 2002). Among the “long cells” class, bulliform cells
play an essential role in water regulation. In fact, when the moisture
decreases, they can modify their morphology, which permits the grass
to leaf-roll (Moulia, 1994). At a later stage of silica accumulation in the
plant, the bulliform cells become inefficient at regulating water flow
and allowing leaf movement to avoid the desiccation of grass leaves
(Parry and Smithson, 1958). The production of silicified bulliform cells
(phytoliths) seems to occur under conditions of strong transpiration.
Therefore, the abundance of bulliform cells in soils and sediments
may be used to indicate the water stress intensity as recorded by the
grasses. Their frequency in sediments is used to indicate past
ecological and climatic conditions as experienced by the vegetation
at a local scale. The frequency of bulliform phytoliths has been used to
investigate and evaluate plant water availability (Verdin et al., 2001;
Brémond et al., 2005b; Delhon, 2005, 2007; Borba-Roschel et al., 2006;
Strömberg et al., 2007; Barboni et al., 2007).
Three indices have previously been proposed to estimate the grass
water stress based on phytolith assemblages in different regions of the
world (Table 2). The Bi index was defined as a proxy of hydric stress for
Holocene sequences in the Mediterranean area (Delhon, 2005, 2007). It
was compared to calcareous horizon development in palaeosols to
evaluate the relationship with levels of evapotranspiration. The Fs index
was established on modern data and compared with evapotranspiration
intensity in West Africa (Brémond et al., 2005b). There, the Fs index fits
well with the aridity gradient of the diverse vegetation communities;
however locally wet areas in dry zones may also induce high production
of bulliform phytolithis. The Bull/GSSC index was proposed to evaluate
the water stress intensity in the Eastern Mediterranean during the
Tertiary Period. In these studies, the water stress index usually indicated
dry conditions, but showed that high rates of bulliform production could
5.3. Indices based on phytolith data
Some authors proposed phytolith ratios (also referred to as
indices) to reconstruct changes in past climate and vegetation. The
D/P (Dicotyledonous/Poaceae) ratio is a proxy of tree cover density
(Alexandre et al., 1997; Brémond et al., 2005a); the Iph index is an
indicator of arid conditions (Diester-Haass et al., 1973) and the Ic
climatic index is an indicator of climate conditions (Twiss, 1987,
1992). Ic and IpH indices are calculated with the frequencies of the
different “C4 grass” groups, and are generally used in tropical
environments where these grasses are well-represented. Other
indices have recently been proposed to estimate the intensity of
Table 2
Three water stress indices quotient already published (Brémond et al., 2005a,b; Delhon,
2005, 2007; Strömberg et al., 2007).
Delhon; 2005 BI =
Bulliform Index
Brémond et al:; 2005
Fan shaped
∑phytoliths of bulliform cells
∑phytoliths of elongate cells
Fs =
Strömberg et al:; 2007 =
∑phytoliths of bulliform cells
ð∑Poaceae phytoliths−∑phytoliths of elongate cellsÞ
∑phytoliths of bulliform cells
∑GSSC ð¼∑phytoliths of short cellsÞ
× 100
6
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
7
Fig. 6. Correspondence analysis of the M5 dataset (without elongate): each sample is represented with its stratigraphic position.
occur in locally saturated soils. In order to assess their suitability as
proxy indicators of moisture fluctuations, the different indices were
compared for the Dmanisi phytolith analyses.
6. Results
6.1. Phytolith analysis: main results
Phytoliths are very abundant in the overall stratigraphy of the
Dmanisi deposits. Grasses (Poaceae) are the dominant taxon in all
phytolith samples (Table 1). Pooideae and Panicoideae appear to be the
best represented subfamilies in the assemblages. The Dicotyledonous
group was also identified, although this group is relatively scarce. The
globular echinate morphotype could be interpreted as a phytolith from
the Arecaceae family, but its very low frequency (0.3% in sample M5 15)
and its dispersion capacity does not permit confirmation.
In the M5 section (Fig. 5), the lower part of the deposits (layers
A1a, A1b and A2) yielded phytolith assemblages that are dominated
by the following forms: elongate (ranging from 44.3% to 60.5%),
rondel (5.3%–27.5%), sinuate (3%–10.4%), acicular (5.3%–18.4%) and
short acicular (0.2%–13.8%) classes, associated with bilobate phytoliths (0.3%–4.3%). In Strata A3–A4, short acicular (0.3%–2%) and
bilobate (0.3%–1.8%) frequencies are lower, while the trapeziform
sinuate (8%–15.5%) and bulliform (0.9%–9.1%) frequencies increase
significantly. In B1 and B2 deposits, acicular (17.8%–22.7%) phytoliths
are very frequent, while the rondel category (6%–12.3%) shows lower
frequencies than in Stratum A. Bilobate phytoliths are absent above
Stratum B1. Strata B1 and B2 yielded the highest values of bulliform
(4.4%–8.6%) and trapeziform sinuate (9.8%–26.1%) morphotypes.
Except for the highest sample (MV 27), the strata B3 and B4 samples
are characterised by decreased frequencies of bulliform, trapeziform
sinuate and acicular forms, and increased frequencies of the rondel
category. Throughout the whole sequence, Dicotyledonous phytoliths
remain rare, although they seem to be better represented in both the
lowest and the highest layers of the deposits. Phytolith assemblages
highlight a shift commencing in the uppermost A deposits that
reaches its maximum in the B2 deposits. To test this hypothesis a
statistical correspondence analysis was performed with the Past,
v.1.88 (Hammer et al., 2001) on the M5 dataset without elongate type.
The first three co-ordinate axes represent 80.65% of the total variation
in the assemblage dataset and axis 1 and axis 2 explain 46% and 21% of
the total variation respectively (Fig. 6). These first two co-ordinate
axis indicate two main groups in the dataset. The first axis tends to
separate samples coming from the lower deposits (A1a, ,A1b A2) with
positive values, and samples coming from the middle–upper deposits
(A4, B1, B2) with negative values. Samples from A3, which represent
the transition between these both groups, possess intermediate
values. This analysis demonstrates a clear separation between the
assemblages from the lower part of A deposits and the assemblages
from the end of A through the B1–B2 deposits.
Rondel and sinuate forms are essentially produced by the Pooideae
subfamily (Twiss et al., 1969; Mulholland, 1989; Fredlund and
Tieszen, 1994; Brémond et al., 2004, 2008). They are dominant in
this sequence, and their presence suggests temperate climatic
conditions. In the Dmanisi sequences, the bilobate morphotype
appears to be principally produced by the Panicoideae subfamily
(Fredlund and Tieszen, 1994; Lu and Liu, 2003; Messager, 2006). In
modern flora, the Panicoideae subfamily is mostly composed of tall
Fig. 5. Phytolith diagram from the Dmanisi M5 profile, the “95% confidence interval” (1.96⁎standard error of the %) was mentioned for the most used morphotypes (bulliform,
bilobate, polylobate, saddle, and globular) which sometimes show low proportions.
8
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
grasses that grow in warm and humid conditions (Tieszen et al., 1979;
Watson and Dallwitz, 1992; Fredlund and Tieszen, 1994, 1997a,b;
Scott, 2002). Although the bilobate morphotype does not show
significant proportions in the lower deposits, as indicated by the error
bar, their abrupt and complete disappearance in the B deposits could
suggest a shift in the edaphic or environmental conditions during this
period. Furthermore, the phytolith analysis carried out in Block 2
shows the same trend (Fig. 7), with a progressive decrease through
the stratigraphy of the bilobate morphotype, and its total disappearance in Stratum B. In the Block 2 section, short acicular frequencies
tend to decline progressively. Sinuate and bulliform rates reach their
maximum values in the B deposits (15.6% in B3 and 11.9% in B2). The
phytolith assemblages of the both sequences (Block 2 and M5)
demonstrate modifications in the local vegetation probably reflecting
regional climatic shifts.
6.2. Phytolith analysis: water stress indices
Three water stress indices calculated with bulliform phytolith
frequencies have been applied to both sequences (Figs. 8 and 9),
revealing similar patterns. A large disparity in water stress indices is
found between the lower (A1a–A3) and the upper part (A4–B3) of the
stratigraphy. The difference of the water stress index between these
two groups was tested using t-tests as suggested by Strömberg, 2009.
A Welsh test for unequal variances for the Bi index confirmed a
significant difference (p = 0.0003) in Bi values between the lower
(A1a–A3) and the upper part (A4–B3) of the M5 sequence.
Comparison of the three indices in both profiles reveals similar
overall trends (Figs. 8 and 9). In the M5 profile, the indices exhibit
small oscillations in the lower part of the section with a major increase
in values in the upper part (upper Stratum A through Stratum B).
Three small peaks in the indices are recorded in A1b, A2 and A3, but
the highest values are recorded in A4, B1 and B2. In Block 2, a similar
overall increase in the three indices can be observed in B1, B2 and B3.
In the upper part of the M5 profile, the Bi index differs slightly from
the Fs and Bull/GSSC indices. In fact, the highest Bi (sample 22)
corresponds to a small decrease in the two other indices (Fig. 8). This
difference stems from the decline of elongate class frequencies in this
sample which results in higher Bi and lower Fs and Bull/GSSC indices.
Sample 27 shows a decrease of short cells and a higher value of
elongated cells (the opposite case). With the exception of these two
minor differences which correspond to the mode of calculation, this
study demonstrates that the three all point to a significant vegetation
shift within the sequence.
6.3. Phytolith analysis: D/P index
The D/P index (Dicotyledonous/Poaceae), a proxy that can be
related to tree cover density, was calculated for each sample
(Alexandre et al., 1997; Brémond et al., 2005a). In this study,
phytoliths assigned to the Dicotyledonous group are composed of
cylindric sulcate and globular forms (Alexandre et al., 1997; Albert et
al., 1999; Borba-Roschel et al., 2006; Barboni et al., 2007). The results
show that the values of this index are very low in both the M5 and
Block 2 sequences (Figs. 8 and 9), with the highest D/P ratios (ranging
from 0.10 to 0.17) in Strata A1a and B4. In the M5 section, A1b, A2, B1,
B2 and B3 samples show low and stable D/P values, ranging from 0.02
to 0.08. A welsh test did not reveal significant differences (p = 0.04) in
D/P values between Stratum A1a and the samples from A1b, A2, B1, B2
and B3. However, in both M5 and Block 2, the phase with high water
stress indices is characterised by low Dicotyledonous frequencies
(Figs. 8 and 9).
7. Discussion
Abundant and well-preserved phytoliths are preserved in the
Dmanisi sediments. Changes in phytolith assemblage compositions
correlate well between the M5 and Block 2 sections, following the
Fig. 7. Phytolith diagram from Dmanisi Block 2 sequence, the “95% confidence interval” (1.96*standard error of the %) was mentioned for the most used morphotypes (bulliform,
bilobate, polylobate, saddle, and globular) which sometimes show low proportions.
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
9
Fig. 8. Water stress indices and D/P index for the M5 profile.
lithostratigraphy. In both sections, the bilobate class is restricted to
Stratum A (with highest frequencies in A1a) and the highest water
stress indices are from Strata B1–B2 samples. The bilobate morphotype is also fairly abundant in the lower part of the Orozmani section,
which is situated only a few kilometres away from Dmanisi and is
chronologically and stratigraphically correlated with Stratum A1a
deposits at Dmanisi (Gabunia et al., 2000a; Messager, 2006). These
taxa disappear abruptly at the onset of Stratum B deposition at
Dmanisi. This phytolith signal could be interpreted as a decrease in
moisture in the Dmanisi area at the time of the deposition of the upper
part of the sequence.
A notable increase in water stress indices, beginning in Stratum A4,
is recorded with the help of indices calculated using bulliform
phytolith frequencies. Bulliform phytoliths appear to be a very good
proxy for water stress among grasses. The carbonate accumulation
identified in the middle part of the stratigraphy (Fig. 3) seems to
correspond to the phase of highest water stress values (Figs. 8 and 9).
The calcareous horizons are the result of a secondary precipitation
following the A–B stratigraphic transition (Lordkipanidze et al., 2007).
A high water stress index has already been related to CaCO3 horizons
in palaeosol sequences in Mediterranean area (Delhon, 2007). This
process of carbonate precipitation in soils depends on various
environmental factors, including evapotranspiration (Becze-Deak
et al., 1997). Further investigations are needed to better understand
the timing and factors leading to CaCO3 precipitation. It could be
closely linked to elevated evapotranspiration suggested by the water
stress indices in the Dmanisi sequence.
Grass water stress indices derived from bulliform frequencies can
be used to assess climate characteristics during deposition. These new
results using the Dmanisi data clearly suggest an increase in
evapotranspiration took place at the end of Stratum A (especially in
A4) and increased during Stratum B (especially B1 and B2). The
marked change in phytolith derived indices in both profiles is very
likely related to an environmental shift in the Dmanisi area. The
changes in water stress indices can be compared to the different
phytolith morphotypes and other proxies coming from the deposit, as
suggested by Strömberg et al. (2007). In the upper deposits of
Dmanisi, only two diatoms were identified and there is no
sedimentary signature of wetland environments. High evapotranspiration levels due to wet local conditions are therefore excluded.
10
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
Fig. 9. Water stress indices and D/P index for the Block 2 sequence.
Instead, the high water stress indices may rather be explained by a
significant reduction in precipitation in the Dmanisi region towards
the end of the Olduvai subchron. The lack of the bilobate morphotype
(assigned to Panicoideae) in the assemblages of the upper deposits
could be related to such a climatic shift. These subtropical grasses
(which grow preferentially in wet conditions) disappeared in the
lowest B deposits (B1). Then, the modification in bilobate frequencies,
revealing a change in vegetation, occurred after the beginning of the
aridity event indicated at the end of A deposits (A4). These results
suggest that there is a delay in the plants' reaction to unfavourable
environmental conditions. In the B deposits, the abundance of sinuate
morphotypes, a good tracer of Pooideae temperate C3-grasses
(Barboni et al., 2007), corresponds with the apparent disappearance
of the subtropical grasses. A lower precipitation rate, recorded by
water stress indices, could have been associated with a cooling
climate as indicated by high values of the sinuate morphotype.
Consistently low frequencies of Dicotyledonous phytoliths characterise these fossil assemblages (Figs. 8 and 9) Analysis of modern
phytolith samples from open and forested plant communities in
Transcaucasia do not reveal significant differences in D/P ratios
(Messager, 2006; unpublished). In temperate ecosystems (Brémond
et al., 2004; Strömberg, 2004) as in Afromontane forests (Barboni
et al., 2007), dicotyledonous, and especially woody taxa produce very
few or no globular phytoliths, leading to an over-representation of
grasses versus woody trees and shrubs. While this index will not be
used to reconstruct tree cover in Dmanisi, it should be noted that the
lowest D/P values correspond to the phase of high water stress indices
(Figs. 8 and 9). Increased aridity could have been a limiting factor for
the expansion of some dicotyledonous plants in the Dmanisi vicinity.
Palynological analysis previously undertaken on the Dmanisi
sediments (Kvavadze and Vekua, 1993; Kvavadze, 1997; Messager,
2006) indicates that the pollen spectra from B sediments suggest the
presence of a steppe-forest, dominated by grasses (Poaceae) with
steppic elements. Palynological evidence for forested ecosystems
show that these were present in the regional environment (Kvavadze
and Vekua, 1993; Messager, 2006), but the predominance of Poaceae
indicates open, grassy environments. Phytolith assemblages recovered in stratum B are evidence of the local herbaceous ecosystem.
Fruit analyses at Dmanisi (Messager et al., 2008) showed that almost
all carpological remains recovered in the B sediments belong to
xerophilous taxa, which are adapted to open and dry environments.
Archaeobotanical data provided by carpo-remains and pollen correlate well with phytolith data and water stress indices indicating
decreased precipitation in the Dmanisi sequence. Macro- and micromammal taxa identified from Dmanisi support a reconstruction of
open, relatively dry environments, with regional forested areas
(Gabunia et al., 2000b; Lordkipanidze et al., 2007). According to all
these climatic and ecological proxies (palaeobotanical and palaeontological data), a reduction in precipitation took place in the Dmanisi
area at the end of the Olduvai subchron, which corresponds to the
period of hominin occupation (Fig. 10). Therefore, we suggest that the
Dmanisi hominins occupied a relatively open environment of steppeforest, characterised by a rather dry climate. Between 1.8 and 1.6 Ma
large-scale faunal turnovers and climatic instabilities have been
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
11
References
Fig. 10. Chrono-ecological data synthesis.
documented (Aguirre and Carbonell, 2001). Due to the intensification
of glacial–interglacial cycles, this period is characterised by climatic
and environmental changes in Eurasia (Sémah, 1986; CombourieuNebout, 1993; Suc et al., 1997; Aguirre and Carbonell, 2001), in
particular at the end of Olduvai subchron (Combourieu-Nebout, 1990;
Zagwijn, 1992). Palaeoecological studies at Eurasian sites bearing on
the first hominin dispersals from Africa suggest that early migrations
occurred in several waves, following open, grassland-dominated
ecosystems (Tchernov, 1992; Bar-Yosef, 1994; Sahnouni and de
Heinzelin, 1998; Sahnouni et al., 2002; Dennell, 2003; Strauss and
Bar-Yosef, 2001).
8. Conclusion
Phytolith analyses of the Dmanisi stratigraphic sequence have
significantly improved our understanding of the palaeoenvironment
and palaeoclimate in which hominins lived in the Caucasus during the
early Pleistocene. The good state of phytolith preservation allowed us
to apply water stress indices, which in turn provide a significant
climatic indicator. This signal supports the conclusion that the
environmental and climatic conditions changed in this region during
the period represented by the Dmanisi deposits. Phytoliths and other
archaeobotanical data point to an ecological shift towards increasing
aridity that was contemporaneous with human presence in this
region.
Acknowledgements
This study was prepared in collaboration with the Georgian National
Museum and the French Museum of Natural History (MNHN). We are
grateful to C. Falguères, V. Lebreton, L. Marquer, A. Mgeladze, B.
Patenaude, J. Renault-Miskovsky, S. Thiébault and P. Voinchet for their
help and to Sally Reynolds for English revision. This work was partly
supported by the Fyssen Foundation and the French Foreign Office. We
thank Doris Barboni and an anonymous reviewer for their constructive
suggestions, which greatly improved the manuscript.
Abrantes, F., 2003. A 340,000 year continental climate record from tropical Africa —
news from opal phytoliths from the equatorial Atlantic. Earth and Planetary Science
Letters 209 (1–2), 165–179.
Aguirre, E., Carbonell, E., 2001. Early human expansions into Eurasia: the Atapuerca
evidence. Quaternary International 75, 11–18.
Albert, R.M., Lavi, O., Estroff, L., Weiner, S., Tsatskin, A., Ronen, A., Lev-Yadun, S., 1999.
Mode of occupation of Tabun Cave, Mt Carmel, Israel during the Mousterian period:
a study of the sediments and phytoliths. Journal of Archaeological Science 26 (10),
1249–1260.
Alexandre, A., Meunier, J.-D., Lézine, A.-M., Vincens, A., Schartz, D., 1997. Phytoliths:
indicators of grassland dynamics during the late Holocene in intertropical Africa.
Paleogeography, Paleoclimatology, Paleoecology 136, 213–229.
Antón, S., Leonard, W.R., Robertson, M.L., 2002. An ecomorphological model of initial
hominid dispersal from Africa. Journal of Human Evolution 43, 773–785.
Arribas, A., Palmqvist, P., 1999. On the ecological connection between sabre-tooths and
hominids: faunal dispersal events in the lower Pleistocene and a review of the
evidence for the first human arrival in Europe. Journal of Archaeological Science 26,
571–585.
Barboni, D., Bonnefille, R., Alexandre, A., Meunier, J.-D., 1999. Phytoliths as
paleoenvironmental indicator in the Middle Awash hominid site, Ethiopia.
Paleogeography, Paleoclimatology, Paleoecology 152, 87–100.
Barboni, D., Brémond, L., Bonnefille, R., 2007. Comparative study of modern phytolith
assemblages from inter-tropical Africa. Palaeogeography, Palaeoclimatology, Palaeoecology 246, 454–470.
Bar-Yosef, O., 1994. The Lower Palaeolithic of the Near East. Journal of World Prehistory
8, 211–264.
Bar-Yosef, O., Belfer-Cohen, A., 2001. From Africa to Eurasia—early dispersals.
Quaternary International 75 (1), 19–28.
Becze-Deak, J., Langohr, R., Verrechia, E.P., 1997. Small scale secondary CaCO3 accumulations in selected sections of the European loess belt. Morphological forms and
potential for palaeoenvironmental reconstruction. Geoderma 76, 221–257.
Blinnikov, M.S., Busacca, A., Whitlock, C., 2002. Reconstruction of the Late Pleistocene
grassland of the Columbia basin, Washington, USA, based on phytolith records in
loess. Palaeogeography, Palaeoclimatology, Palaeoecology 177 (1–2), 77–101.
Bobe, R., Behrensmeyer, A.K., Chapman, R.E., 2002. Faunal change, environmental
variability and late Pliocene hominin evolution. Journal of Human Evolution 42 (4),
475–497.
Bobe, R., Behrensmeyer, A.K., 2004. Expansion of grassland ecosystems in Africa in
relation to mammalian evolution and the origin of the genus Homo. Palaeogeography, Palaeoclimatology, Palaeoecology 207, 399–420.
Bonnefille, R., 1995. A reassessement of the Plio-Pleistocene pollen record of East Africa.
In: Vrba, E.S., Denton, G.H., Partridge, T.C., Burckle, L.H. (Eds.), Paleoclimate and
Evolution with Emphasis on Human Origins. Yale University Press, New Haven, CT,
pp. 300–310.
Borba-Roschel, M., Alexandre, A., Drummond Chicarino Varajao, A.F., Meunier, J.D.,
Chicarino Varajao, C.A., Colin, F., 2006. Phytoliths as indicators of pedogenesis and
paleoenvironmental changes in the Brazilian cerrado. Journal of Geochemical
Exploration 88, 172–176.
Bozarth, S., 1992. Classification of opal phytoliths formed in selected Dicotyledons
native to the Great Plains. In: Rapp, G., Mulholland, S. (Eds.), Phytoliths Systematics,
Emerging Issues. Plenum Press New York, pp. 193–214.
Brémond, L., Alexandre, A., Véla, E., Guiot, J., 2004. Advantages and disadvantages of
phytolithes analysis for the reconstruction of Mediterranean vegetation: an
assessment based on modern phytolith, pollen and botanical data (Luberon,
France). Review of Palaeobotany and Palynology 129 (4), 213–228.
Brémond, L., Alexandre, A., Hély, C., Guiot, J., 2005a. A phytolith index as a proxy of tree
cover density in tropical areas: calibration with Leaf Area Index along a forest-savanna
transect in southeastern Cameroon. Global and Planetary Change 45, 277–293.
Brémond, L., Alexandre, A., Peyron, O., Guiot, J., 2005b. Grass water stress estimated
from phytoliths in West Africa. Journal of Biogeography 32, 311–327.
Brémond, L., Alexandre, A., Wooller, M.J., Hely, C., Williamson, D., Schafer, P.A., Majule,
A., Guiot, J., 2008. Phytolith indices as proxies of grass subfamilies on East African
tropical mountains. Global and Planetary Change 61, 209–224.
Brown, D., 1984. Prospects and limits of a phytolith key for grasses in the central United
States. Journal of Archaeological Science 11 (4), 345–368.
Carbonell, E., Bermúdez de Castro, J.M., Parés, J.M., Pérez-González, A., Cuenca-Bescós,
G., Ollé, A., Mosquera, M., Huguet, R., van der Made, J., Rosas, A., Sala, R., Vallverdú,
J., García, N., Granger, D.E., Martinón-Torres, M., Rodríguez, X.P., Stock, G.M., Vergès,
J.M., Allué, E., Burjachs, F., Cáceres, I., Canals, A., Benito, A., Díez, C., Lozano, M.,
Mateos, A., Navazo, M., Rodríguez, J., Rosell, J., Arsuaga, J.L., 2008. The first hominin
of Europe. Nature 452, 465–469.
Combourieu-Nebout, N., 1990. Les cycles glaciaire-interglaciaire en région méditerranéenne de − 2, 4 ∼ − 1, 1 Ma: analyse pollinique de la série de Crotone (Italie
Méridionale). Paléobiologie Continentale 17, 35–39.
Combourieu-Nebout, N., 1993. Vegetation response to Upper Pliocene glacial/interglacial cyclicity in the Central Mediterranean. Quaternary Research 40, 228–236.
Delhon, C., 2005. Anthropisation et paléoclimats du Tardiglaciaire à l'Holocène en
moyenne vallée du Rhône: études pluridisciplinaires des spectres phytolithiques et
pédo-anthracologiques de séquences naturelles et de sites archéologiques. Ph.D.
Thesis, Université Paris I, 2 vol. 262, 583 pp.
Delhon, C., 2007. Phytolith and pedoanthracology analysis of “off-site” Holocene
sequences from Mondragon (Middle Rhone Valley, south of France). In: Zurro, D.
(Ed.), Madella, M. Recent Studies in Phytolith Analysis, Oxbow Books, Plants People
and Places, pp. 175–188.
12
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
Delhon, C., Alexandre, A., Berger, J.-F., Thiébault, S., Brochier, J.-L., Meunier, J.-D., 2003.
Phytolith assemblages as a promising tool for reconstructing Mediterranean
Holocene vegetation. Quaternary Research 59, 48–60.
DeMenocal, P.B., Bloemendal, J., 1995. Plio–Pleistocene subtropical African climate
variability and the paleoenvironment of hominid evolution: a combined data–
model approach. In: Vrba, E., Denton, G., Burckle, L., Partridge, T. (Eds.), Paleoclimate
and Evolution with Emphasis on Human Origins. Yale University Press, New Haven,
pp. 262–288.
Denk, T., Frotzler, N., et Davitashvili, N., 2001. Vegetational patterns and distribution of
relict taxa in humid temperate forests and wetlands of Georgia (Transcaucasia).
Biological Journal of the Linnean Society 72, 287–332.
Dennell, R., 2003. Dispersal and colonisation, long and short chronologies: how
continuous is the Early Pleistocene record for hominids outside East Africa? Journal
of Human Evolution 45 (6), 421–440.
Diester-Haass, L., Schrader, H.J., Thiede, J., 1973. Sedimentological and paleoclimatological investigations of two pelagic ooze cores off Cape Barbas, North-West Africa.
Meteor Forshungergebnisse 16, 19–66.
Dzaparidze, V., Bosinski, G., Bugianisvili, T., Gabunia, L., Justus, A., Klopotovskaja, N.,
Kvavadze, E., Lordkipanidze, D., Maisuradze, G., Mgeladze, N., Nioradze, M.,
Pavlenishvili, E., Schmincke, H.-U., Sologasvili, D., Tusabramisvili, D., Tvalcrelidze, M.,
Vekua, A., 1989. Der alfipaläolithische Fundplatz Dmanisi in Georgian (Kaukasus).
Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 36, 67–116.
Falguères, C., 2003. ESR dating and the human evolution: contribution to the chronology
of the earliest humans in Europe. Quaternary Science Reviews 22, 1345–1351.
Ferring, C.R., Lordkipanidze, D., Berna, F., Ohms, O., 2008. Geology and Formation
Processes at Dmanisi in the Georgian Caucasus. Abstracts of the 73rd Annual
Meeting, Society for American Archaeology, p. 195.
Fredlund, G., Tieszen, L., 1994. Modern phytolith assemblages from the North American
Great Plains. Journal of Biogeography 21, 321–335.
Fredlund, G., Tieszen, L., 1997a. Phytoliths and carbon isotope evidence for late
quaternary vegetation and climate change in the southern Black Hills, South
Dakota. Quaternary research 47, 206–217.
Fredlund, G., Tieszen, L., 1997b. Calibrating grass phytolith in climatic terms:
application to late Pleistocene assemblages from Kansas and Nebraska. Palaeogeography, Palaeoclimatology, Palaeoecology 136 (1–4), 199–211.
Gabunia, L., Vekua, A., Lordkipanidze, D., Swisher III, C.C., Ferring, R., Justus, A., Nioradze, M.,
Tvalcrelidze, M., Anton, S.C., Bosinski, G., Jöris, O., Lumley de, M.A., Majsuradze, G.,
Mouskhelishvili, A., 2000a. Earliest Pleistocene hominid cranial remains from Dmanisi,
Republic of Georgia: taxonomy, geological setting, and age. Science 288, 1019–1025.
Gabunia, L., Vekua, A., Lordkipanidze, D., 2000b. The environmental contexts of early
human occupation of Georgia (Transcaucasia). Journal of Human Evolution 38,
785–802.
Gabunia, L., Vekua, A., 1995. A Plio-Pleistocene hominid from Dmanisi, East Georgia,
Caucasus. Nature 373, 509–512.
Hammer, Ø., Harper, D.A.T.P., Ryan, D., 2001. PAST: paleontological statistics software
package for education and data analysis. Palaeontologia Electronica 4 (1) 9 pp.
Hodson, M.J., White, P.J., Mead, A., Broadley, M.R., 2005. Phylogenetic variations in the
silicon composition of plants. Annals of Botany 96, 1027–1046.
Holmes, K.M., 2007. Using Pliocene palaeoclimatic data to postulate dispersal pathways
of early hominins. Palaeogeography, Palaeoclimatology, Palaeoecology 248 (1–2),
96–108.
Horrocks, M., Deng, Y., Ogden, J., Sutton, D.G., 2000. A reconstruction of the history of a
Holocene sand dune on Great Barrier Island, northern New Zealand, using pollen
and phytolith analyses. Journal of Biogeography 27, 1269–1277.
Hutton, J., Norrish, K., 1974. Silicon content of wheat husks in relation to water
transpired. Australian Journal of Agricultural Research 25, 203–212.
ICPN Working Group, Madella, M., Alexandre, A., Ball, T., 2005. International code for
phytolith nomenclature 1.0. Annals of Botany 96 (2), 253–260.
Kaplan, L., Smith, M., Sneddon, L., 1992. Cereal grain phytoliths of Southwest Asia and
Europe. In: Rapp, J., Mulholland, S. (Eds.), Phytolith Systematics Emerging Issues.
Plenum Press, New York, pp. 149–174.
Kvavadze, E., 1997. The Perspective of the Coprolotical Study of Mammals by Pollen
Analysis for Solution of Paleoecological and Ethological Problems. Abstract
Biochrom '97 Congress, Montpellier, pp. 121–128.
Kvavadze, E., Vekua, A., 1993. Vegetation and climate of the Dmanisi man period (East
Georgia) from palynological data. Acta Palaeobotanica 33 (2), 343–355.
Le Cohu, M.-C., 1973. Examen au Microscope Electronique à Balayage des cônes de silice chez
les Cypéracées. Compte rendu de l'académie des sciences de Paris 277, 130–1303.
Lentfer, C.J., Boyd, W.E., 1998. A comparison of three methods for the extraction of
phytoliths from sediments. Journal of Archaeological Science 25, 1159–1183.
Lordkipanidze, D., Jashashvili, T., Vekua, A., Ponce de León, M.S., Zollikofer, C.E.,
Rightmire, G.P., Pontzer, H., Ferring, R., Oms, O., Tappen, M., Bukhsianidze, M.,
Agusti, J., Kahlke, R., Kiladze, G., Martinez-Navarro, B., Mouskhelishvili, A., Nioradze,
M., Rook, L., 2007. Postcranial evidence from early Homo from Dmanisi, Georgia.
Nature 449, 305–310.
Lu, H., Liu, K.B., 2003. Morphological variations of lobate phytoliths from grasses in
China and the south-eastern United States. Diversity and Distributions 9, 73–87.
Lu, H.-Y., Wu, N.-Q., Yang, X.-D., Jiang, H., Liu, K.-B., Liu, T.-S., 2006. Phytoliths as quantitative
indicators for the reconstruction of past environmental conditions in China. I: phytolithbased transfer functions. Quaternary Science Reviews 25 (9–10), 945–959.
Lu, H.-Y., Wu, N.-Q., Liu, K.B., Jiang, H., Liu, T.S., 2007. Phytoliths as quantitative
indicators for the reconstruction of past environmental conditions in China II:
palaeoenvironmental reconstruction in the Loess Plateau. Quaternary Science
Review 26 (5–6), 759–772.
Lumley de, H., Lordkipanidze, D., Féraud, G., Garcia, T., Perrenoud, C., Falguères, C.,
Gagnepain, J., Saos, T., Voinchet, P., 2002. Datation par la méthode 40Ar / 39Ar de la
couche de cendres volcaniques (couche VI) de Dmanissi (Géorgie) qui a livré des
restes d'hominidés fossiles de 1, 81 Ma. Comptes Rendus Palevol 1, 181–189.
Madella, M., 2002. Phytolith Deposition and Taphonomy in a Loess–Palaeosol Sequence
of the Italian Pre-Alps. The Val Sorda (Veneto region). Oral communication, 4th
IMPR, 28th–31st August 2002.
Messager, E., 2006. Apports des études paléobotaniques à la reconstitution paleoenvironnementale du site de Dmanissi et de sa région (Géorgie). Ph.D. Thesis, Museum
National d'Histoire Naturelle, Paris, France. 243 pp.
Messager, E., Lordkipanidze, D., Ferring, C.R., Deniaux, B., 2008. Fossil fruit identification
by SEM investigations, a tool for palaeoenvironmental reconstruction of Dmanisi
site. Georgia Journal of Archaeological Science 35 (10), 2715–2725.
Motumora, H., Fujii, T., Suzuki, M., 2000. Distribution of silicified cells in the leaf blades of
Pleioblastus chino (Franchet et Savatier) Makino (Bambusoideae). Annals of Botany
85, 751–757.
Moulia, B., 1994. The biomechanics of leaf rolling. Biomimetics 2, 267–281.
Mulholland, S., 1989. Grass Opal Phytolith Production: A Basis for Archaeological
Interpretation in the Northern Plains. Archaeobotany Through Phytolith Analysis
Symposium, Annual Meeting of the Soc. Amer. Archaeology, Abstracts, The Phytolitarian
Newsletter, 6 (1), p. 4.
Mulholland, S., Rapp, G., Ollendorf, A., 1988. Variation in phytoliths from corn leaves.
Canadian Journal of Botany 66, 2001–2008.
Nakhutsrishvili, G.S., 1999. The vegetation of Georgia (Caucasus). Braun-Blanquetia 15,
1–68.
O'Connell, J.F., Hawkes, K., Blurton Jones, N.G., 1999. Grandmothering and the evolution
of Homo erectus. Journal of Human Evolution 36, 461–486.
Oms, O., Parés, J.M., Martínez-Navarro, B., Agustí, J., Toro, I., Martínez-Fernández, G.,
Turq, A., 2000. Early Human Occupation of Western Europe: Paleomagnetic Dates
for Two Palaeolithic Sites in Spain. Proceedings of the National Academy of Sciences
of the United States of America 97, 10,666–10,670.
Parry, D.W., Smithson, F., 1958. Silicification of bulliform cells in grasses. Nature 181,
1549–1550.
Pearsall, D., 2000. Palaeoethnobotany, a Hand-book of Procedure. Academic Press, New
York. 700 pp.
Piperno, D., 1985. Phytolith analysis and tropical paleo-ecology: production and
taxonomic significance of siliceous forms in the new world plant domesticates and
wild species. Review of Palaeobotany and Palynology 45, 185–228.
Piperno, D., 1988. Phytolith Analysis: An Archaeological and Geological Perspective.
Academic Press, San Diego. 280 pp.
Piperno, D., 1989. The occurrence of phytoliths in the reproductive structures of
selected tropical angiosperms and their significance in tropical palaeoecology,
palaeoethnobotany and systematics. Review of Palaeobotany and Palynology 61,
147–173.
Piperno, D., 2006. Phytoliths: A Comprehensive Guide for Archaeologists and
Paleoecologists. AltaMira Press, New York. 304 pp.
Piperno, D., Andres, T., Stothert, K., 2000. Phytoliths in Cucurbita and other Neotropical
Cucurbitaceae and their occurrence in Early Archaeological sites from the Lowland
American Tropics. Journal of Archaeological Science 27 (3), 193–208.
Prat, H., 1932. L'épiderme des graminées. Etude anatomique et systématique. Annales
des Sciences Naturelles, Botanique, 10ème série, pp. 118–324.
Potts, R., 1998. Environmental hypotheses of hominin evolution. Yearbook of Physical
Anthropology 41, 93–136.
Rightmire, G.P., Lordkipanidze, D., Vekua, A., 2006. Anatomical descriptions, comparative studies and evolutionary significance of the hominin skulls from Dmanisi,
Republic of Georgia. Journal of Human Evolution 50 (2), 115–141.
Rosen, A.M., 2001. Phytolith evidence for agro-pastoral economies in the Scythian
period of southern Kazakhstan. In: Meunier, J.-D., Colin, F. (Eds.), Phytoliths:
Applications in Earth Sciences and Human History. A.A. Balkema, Lisse, pp. 183–198.
Rovner, I., 1971. Potential of opal phytoliths for use in paleoecological reconstructions.
Quaternary Research 1, 343–359.
Rovner, I., 1983. Plant opal phytolith analysis: major advances in archaeobotanical
research. In: Schiffer, M. (Ed.), Advances in Archaeological Method and Theory.
Academic Press, New York, pp. 225–266.
Runge, F., 1999. The opal phytolith inventory of soils in central Africa quantities,
shapes, classification and spectra. Review of Palaeobotany and Palynology 107,
23–53.
Sahnouni, M., de Heinzelin, J., 1998. The site of Ain Hanech revisited: new investigations
at this Lower Pleistocene site in northern Algeria. Journal of Archaeological Science
25, 1083–1101.
Sahnouni, M., Hadjouis, D., van der Made, J., Derradji, A., Canals, A., Medig, M.,
Behahrech, H., Harichane, Z., Rabhi, M., 2002. Further research at the Oldowan site
of ain Hanech, North-eastern Algeria. Journal of Human Evolution 43, 925–937.
Scott, L., 2002. Grassland development under glacial and interglacial conditions in
southern Africa: review of pollen, phytolith and isotope evidence. Palaeogeography, Palaeoclimatology, Palaeoecology 177 (1–2), 47–57.
Sémah, A.M., 1986. Le Milieu naturel lors du premier peuplement de Java: Résultats de
l'analyse pollinique. Ph.D. Thesis, Université d'Aix-Marseille, 3 vol. pp. 188-118-16.
Sémah, F., Saleki, H., Falguères, C., 2000. Did early man reach Java during the Late
Pliocene? Journal of Archaeological Science 27, 763–769.
Strauss, L.G., Bar-Yosef, O., 2001. Out of Africa in the Pleistocene: an introduction.
Quaternary International 75 (1), 1–3.
Strömberg, C.A.E., 2002. The origin and spread of grass-dominated ecosystems in the
late Tertiary of North America: preliminary results concerning the evolution of
hypsodonty. Palaeogeography, Palaeoclimatology, Palaeoecology 177 (1–2), 59–75.
Strömberg, C.A.E., 2004. Using phytolith assemblages to reconstruct the origin and
spread of grass-dominated habitats in the Great Plains during the late Eocene to
early Miocene. Palaeogeography, Palaeoclimatology, Palaeoecology 207, 239–275.
E. Messager et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 288 (2010) 1–13
Strömberg, C.A.E., Werdelin, L., Friis, E.M., Saraç, G., 2007. The spread of grass-dominated
habitats in Turkey and surrounding areas during the Cenozoic: phytolith evidence.
Palaeogeography, Palaeoclimatology, Palaeoecology 250, 18–49.
Strömberg, C.A.E., 2009. Methodological concerns for analysis of phytolith assemblages:
does count size matter? Quaternary International 193, 124–140.
Suc, J.-P., Bertini, A., Leroy, S.A.G., Suballyova, D., 1997. Towards the lowering of
the Pliocene/Pleistocene boundary to the Gauss-Matuyama reversal. Quaternary
International 40, 37–42.
Swisher III, C.C., Curtis, G.H., Jacob, T., Getty, A.G., Suprijo, A., Widiasmoro, 1994. Age of
the earliest known hominids in Java, Indonesia. Science 263, 1118–1121.
Tchernov, E., 1992. Eurasian–African biotic exchanges through the Levantine corridor
during the Neogene and Quaternary. In: von Koenigswald, W., Werdelin, L. (Eds.),
Mammalian Migration and Dispersal Events in the European Quaternary, Courier
Forschungsinstitut Senckenberg, Frankfurt, 153, pp. 103–123.
Thorn, V.C., 2004. Phytolith evidence for C4-dominated grassland since the early
Holocene at Long Pocket, northeast Queensland. Australia, Quaternary Research 61,
168–180.
Tieszen, L., Senyimba, M., Imbamba, S., Troughton, J., 1979. The distribution of C3 and C4
grasses and carbon isotope discrimination along an altitudinal and moisture
gradient in Kenya. Oecologia 37, 337–350.
Tungsheng, Liu, Zhengtang, Guo, Naiqin, Wu., Houyuan, Lu., 1996. Prehistoric
vegetation on the Loess Plateau: steppe or forest? Journal of Southeast Asian
Earth Sciences 13 (3–5), 341–346.
Twiss, P.C., 1987. Grass-opal phytoliths as climatic indicators of the Great Plains
Pleistocene. In: Johnson, W.C. (Ed.), Quaternary Environments of Kansas.
Guidebook Series 5, Kansas Geological Survey, pp. 179–188.
Twiss, P.C., 1992. Predicted world distribution of C3 and C4 grass phytoliths. In: Rapp, G.,
Mulholland, S. (Eds.), Phytolith Systematics Emerging Issues. Plenum Press, New York,
pp. 113–128.
13
Twiss, P.C., Suess, E., Smith, R., 1969. Morphology Classification of Grass Phytoliths.
Proceedings of the Soil Science Society of America 33, 109–115.
Verdin, P., Berger, J.-F., Lopez-Saez, J.-A., 2001. Contribution of phytolith analysis to the
understanding of historical agrosystems in the Rhône mid-valley (Southern France).
In: Meunier, J.-D., Colin, F. (Eds.), Phytoliths: Applications in Earth Sciences and
Human History. A.A. Balkema, Lisse, pp. 155–172.
Voinchet, P., Bahain, J.J., Falguères, C., Laurent, M., Dolo, J.M., Despriee, J., Gageonnet, R.,
2004. ESR dating of quartz extracted from quaternary sediments. Application to
fluvial terraces system of northern France. Quaternaire 15 (1–2), 135–141.
Volodicheva, N., 2002. Caucasus Mountains. In: Shahgedanova, M. (Ed.), The Physical
Geography of Northern Eurasia, pp. 350–376.
Vrba, E.S., 1995. On the connections between paleoclimate and evolution. In: Vrba, E.S.,
Denton, G.H., Partridge, T.C., Burckle, L.H. (Eds.), Paleoclimate and Evolution with
Emphasis on Human Origins. Yale University Press, New Haven, pp. 24–47.
Watson, L., Dallwitz, M.J., 1992. Grass Genera of the World. Descriptions, Illustrations,
Identification, and Information Retrieval; including Synonyms, Morphology,
Anatomy, Physiology, Phytochemistry, Cytology, Classification, Pathogens, World
and Local Distribution, and References. 1081 pp.
Webb, E., Longstaffe, F., 2002. Climatic influences on the oxygen isotopic composition of
biogenic silica in prairie grass. Geochimica et Cosmochimica Acta 66 (11), 1891–1904.
Zagwijn, W.H., 1992. The beginning of the Ice Age in Europe and its major subdivisions.
Quaternary Science Reviews 11, 583–591.
Zeitoun, V., 2000. Adequation Between Environmental Shifts and Human Speciations
during the Plio-Pleistocene. Comptes Rendus de l'Académie de Sciences - Serie IIa:
Sciences de la Terre et des Planètes 330 (2), 161–166.