Palaeoenvironmental Context of Coprolites and Plant
Microfossils from Unit II. Azokh 1
Louis Scott, Lloyd Rossouw, Carlos Cordova, and Jan Risberg
Abstract Poor pollen preservation in cave deposits is due to
oxidation and increasing scarcity of pollen with distance
from the cave entrance. After an attempt to obtain pollen
grains from the sediments in Azokh 1 (Lesser Caucasus)
failed, two coprolites from Unit II were investigated for their
microfossil contents. They contained few diatoms (including
the rare Pliocaenicus), even less pollen but numerous
phytoliths that were compared with those in selected levels
of cave deposits and modern soil from outside. Grass silica
short cell phytoliths give evidence of vegetation typical of a
temperate climate for Unit II, which included C
Not only the coprolites from Azokh are useful but the whole
sequence of deposits has good potential for palaeoclimatic
reconstruction based on for phytolith studies. The diatoms
observed indicate feeding from a relatively moist terrestrial
environment and availability of lake and/or running water.
Резюме Для изучения экологической ситуации в
процессе возникновения отложений в пещере Азох 1
(Малый Кавказ)химическому анализу были подвергнуты
два образца копролитов.Исследование было предпринято
после попытки получения пыльцы из мелкозернистого
седимента,которая окончилась неудачей по причине
продолжительной оксидации и разложения в условиях
постоянного изменения влажности в пещере,а также
возрастающей нехватки переносимой по воздуху пыльцы
от входа в глубь пещеры.В качестве альтернативного
источника пыльцы и других микроископаемых элементов
были исследованы два копролита,обнаруженных в
подразделении II.Они содержали редко встречающиеся
виды диатомеи,включая Pliocaenicus sp., немного
пыльцы и большое количество фитолитов.Фитолиты в
копролитах были сопоставлены с образцами,отобран-
ными из нескольких слоев отложений внутри и из
современной почвы за пределами пещеры.Различные
типы фитолитов рода Poaceae (силицированные короткие
клетки травы)в пределах подразделения II указывают на
типичную для умеренного климата растительность,
которая включает C
травы и несколько отличается от
современной смешанной флоры.Плотность лесного
покрова не может быть определена без дальнейшего
изучения нетравяных фитолитов в копролитах и
седименте.Последние указывают на то,что мелкозернис-
тая седиментная последовательность в Азох 1имеет
одинаково хороший потенциал для анализа фитолитов в
копролитах и,следовательно,для палеоэкологической
реконструкции всей последовательности отложений,в
том числе и для более обширного региона.Обнаруженные
диатомовые водоросли свидетельствуют об относительно
влажной почве и наличии озерной или речной воды в
качестве источника питания.
Keywords Fossil scats Phytoliths Diatoms MIS 5
L. Scott (&)
Department of Plant Sciences, University of the Free State,
PO Box 339,Bloemfontein 9300, South Africa
Department of Archaeology, National Museum,
PO Box 266,Bloemfontein 9300, South Africa
Department of Geography, Oklahoma State University,
Stillwater, OK 74078, USA
Department of Physical Geography, Stockholm University,
106 91 Stockholm, Sweden
©Springer Science+Business Media Dordrecht 2016
Yolanda Fernández-Jalvo et al. (eds.), Azokh Cave and the Transcaucasian Corridor,
Vertebrate Paleobiology and Paleoanthropology, DOI 10.1007/978-3-319-24924-7_13
The primary way of reconstructing past vegetation and
environments is usually by means of pollen analysis of lakes
and swamps but this method is not regularly used in caves.
Fine-grained cave deposits often provide little or no infor-
mation about past climates because concentrations of aeri-
ally introduced material like pollen, which is introduced as
dust in the cave by air currents or other means, is usually not
high. The inﬂux of transported microscopic particles decli-
nes progressively deeper into a cave and beyond 20 to 30 m
it is very low (Coles and Gilberstone 1994; Navarro et al.
2001; Hunt and Rushworth 2005). Conditions for preser-
vation are often not ideal on cave ﬂoors, and considering
these constraints both for pollen and phytoliths in
ﬁne-grained cave sediments, richer alternative sources may
Our ﬁrst attempt at Azokh 1 in the Lesser Caucasus to
extract pollen from sediments of Units I and V, at depths of
77 cm and lower in the Azokh cave system, was unsuc-
cessful. After exposure for more than 20 years sediments
near the cave opening became dry, crumbly, cracked, bio-
turbated and oxidized (Fernández-Jalvo et al. 2010) and
therefore not suitable for pollen analysis. Other plant
microfossil research may still be feasible, and initial inves-
tigations indicated that sediments potentially contain phy-
toliths and starch inclusions (Fernández-Jalvo et al. 2010).
A further possible reason for the lack of pollen may be the 40
m distance of the present excavation area to cave entrance,
which is removed from aerial pollen sources. Although air
currents are relatively active today (Y. Fernandez-Jalvo,
personal communication 2006), we do not have evidence that
this was the case in the past. Preservation qualities may not be
ideal and pollen could also have been destroyed by a com-
bination of highly oxidizing conditions and microbial action
in acidic bat guano rich in phosphates, which is present
throughout the cave sequence, and by wet-dry cycles in the
cave such as recognized in Unit II (Marin-Monfort et al.
2016). Fresh bat guano is rich in pollen, but in fossil layers it
could have been decomposed over time (Carrion et al. 2006).
Extensive carbonate cementation occurs in some parts of the
Azokh 1 excavation, mainly in levels closer to the limestone
cave walls, as result of seasonal and drip-water ﬂows, and
this could also have played a role in destroying pollen grains.
It has been reported that damp areas near cave walls have
poorer pollen preservation (Navarro et al. 2001; Carrion et al.
In view of the paucity of pollen in the Azokh deposits and
in order to obtain additional dietary or environmental data,
we turned our attention to the coprolites to search for pollen
and phytoliths in them for comparison with phytoliths in the
surrounding deposits. Coprolittes are biogenic inclusions
that trap plants from outside the cave (Thompson et al. 1980;
O’Rourke and Mead 1985; Scott 1987), and they can be
useful alternatives as sources for micro-plant remains
because their inclusions are sealed off and protected more
effectively from adverse sedimentary conditions such as
dampness and oxidation. In the long run these conditions, if
severe enough can destroy pollen anywhere, also in copro-
lites, but the chances are that they will survive longer inside
a coprolite than in unconsolidated deposits (Navarro et al.
2001; Scott et al. 2003). Coprolites in caves can therefore
shed light on prehistoric conditions not only because their
shape, size and structure represent prehistoric fauna, but also
because their microscopic contents provide clues about past
vegetation and climate. Apart from research on hyrax dung
deposits in Africa and the Middle East, previous studies of
coprolites in Africa and Europe were often based on hyena
coprolites (Scott 1987; Carrión et al. 2007) because these
coprolites are more frequently found in caves than those of
other animals, for example badgers, which are less frequent
(Carrion et al. 2005).
Plant microfossils in a coprolite can be derived from an
animal’s diet, it’s drinking water, ingested dust, or that
which became attached to the dung via air currents soon
after defecation. Dung usually traps a representative
assemblage of pollen and organic and siliceous dust derived
from wide surroundings where the animals were roaming.
Dung pellets fossilize in caves to become solid coprolites
that preserve their micro-contents under more stable condi-
tions than those in the surrounding ﬁne, looser deposits,
which experience local variations of humidity and temper-
ature. As long as coprolites do not disintegrate, their slightly
acidic conditions are not necessarily harmful to microscopic
inclusions. Coprolites therefore prevent decomposition and
destruction of organics, but this can be temporary because
the microscopic contents can be lost in the long run if local
conditions deteriorate (Scott et al. 2003).
Because signiﬁcant differences occur in the morphology
of microscopic phytolith types produced by the main Poa-
ceae subfamilies (or grass silica short-cell types (GSSC))
(Twiss et al. 1969; Brown 1984; Mulholland 1989; Fredlund
and Tieszen 1994), these microfossils in coprolites promised
to be an informative tool. Despite the prevalence of ‘multi-
plicity’and ‘redundancy’in GSSC assemblages (Rovner
1971,1983), i.e. the occurrence of a variety of types in one
grass taxon as well as the occurrence of the same type in
different taxa, ﬂuctuations in the frequencies of certain types
can be still be used to distinguish between the grass sub-
families (Fredlund and Tieszen 1994; Rossouw 2009).
Coprolite fragments of unidentiﬁed origin have been
found at Unit II and Vu of Azokh 1. Unit II also yielded two
complete coprolites (no’s 5153 and 5246) of which 5246,
and some obviously derived stone artifacts together with
fossils typical of Unit II, were apparently displaced in Unit I
288 L. Scott et al.
through modern bioturbation (Murray et al. 2016;
Marin-Monfort et al. 2016). During extraction of palaeob-
otanical remains, a bone fragment was recovered indicating a
carnivorous (or omnivorous) diet of the animal that produced
the coprolites. Although several other coprolite fragments
are available in different layers, only these two were com-
plete and undamaged and were used for plant microfossil
This paper deals with the extraction of microfossils from
the two coprolites in Azokh Cave. With the aim of shedding
light on possible environmental conditions that existed
during their formation, we investigate the potential of
microfossils in the coprolites and discuss them in the context
of other fossils and the reconstruction of faunal paleoecology
and charcoal that have been found in the deposits from
where the coprolites came (Andrews et al. 2016).
Environment Around the Cave
The cave at Azokh (39° 37′9.17″N, 46° 59′18.59″E) is in
the Lesser Caucasus at 962 m elevation and the environment
is described in this volume (Andrews et al. 2016; Fernán-
dez-Jalvo et al. 2016). The rainfall is approximately
600 mm/year, falling mainly in May–June and September–
October, while the driest month is January (Republic of
Armenia 1999). The faunal contents in the sequence of
300 kyr in the Azokh Cave sedimentary sequence show
some variations but are typical of steppe, arid conditions or
deciduous woodlands (Andrews et al. 2016). Evidence of the
surrounding vegetation in the past can be derived from
charcoal in Unit II and Unit Vu, consisting mainly of Prunus
(80%) that was probably the most abundant tree species and
could have been gathered by humans as ﬁrewood while
fruits were probably dispersed around the cave (Allué2016;
Andrews et al. 2016).
According to descriptions of present vegetation and plant
communities in the Caucasus region it can broadly be divi-
ded into three zones: foothill grassland, lower-mountain
mixed hardwood forest, and mountain subalpine grassland
(Sharrow 2007). According to global grass distribution
maps, this part of the Caucasus consists mainly of species of
the Pooideae (c. 300 species), which dominate over other
groups like Chloridoideae (17), Paniceae (13), Andro-
pogoneae (6) and Arundinoideae (6). The subfamily Pooi-
deae is the premier group of grasses occupying cool
temperate and boreal regions (Cross 1980; Clayton and
Azokh Cave falls in the lower mountain mixed hardwood
forest which is generally found at 600–1,100 m elevation
(Gulisashvili et al. 1975; Sharrow 2007). At present, most
land suitable for farming has been ploughed, and areas
suitable for grazing have been grazed. Moderate slopes have
often been cleared for use as crop or hay ﬁelds, forming
large openings in the forest, but areas of forest still exist on
steep slopes (Sharrow 2007). The vegetation on the slopes in
the vicinity of the cave are currently grassy woodland veg-
etated by Carpinus,Quercus (probably Q. iberica), Tilia and
Fraxinus with an understory of Prunus,Cornus,Corylus,
Crataegus and Paliurus spina-christi (Andrews et al. 2016).
In the general surroundings Paliurus and Ziziphus is com-
mon in “shibliak”(i.e. secondary woodland that develop
after forest clearing) (Gabrielian and Fragman-Sapir 2008).
The grasslands of lower elevations once occupied the
generally eastern facing foothills and lower slopes of the
mountains at about 300–600 m elevation with an annual
precipitation of approximately 250–400 mm (Sharrow
2007). Further, cool-season grasses occur with several types
of woody species and herbaceous sagebrush in the more
xeric areas while shrubs such as buckthorn, hawthorn, and
black-wood are found in the more mesic areas.
Materials and Methods
Pollen, Phytolith and Diatom Extraction
The two coprolites (No.’s 5153 = AZ1’08 II-I50#12 and
5246 = AZ1’08 I-H49#4) (Fig. 13.1), which measured 50 ×
49 ×33 and 48 ×47 ×30 mm respectively, were sawed in
half. One half of each was saved and the other processed for
plant microfossil extraction. The studied halves were cleaned
by removing the outside 1 to 2 mm layers, which were also
saved together with the dust obtained from sawing. They
were cleaned further by water to remove dust and then
treated in 10% HCl, and cleaned by centrifuging several
times using water. Mineral separation was then performed
by ﬂoating the silica and organic fraction on sodium poly-
tungstate solution (S.G. 2.3) and washing in a centrifuge.
Microscope slides were mounted in glycerine jelly and
investigated under light microscope, using up to 100×oil
Fig. 13.1 The two studied coprolites 5153 (a) and 5246 (b) from
Azokh I, Unit II
13 Coprolites and Plant Microfossils 289
immersion magniﬁcation. Five sediment samples for com-
parison were also chemically processed in the same way as
the coprolites for phytolith investigation (Table 13.1).
Criteria for the Identification
of Phytolith Types
Nine different grass silica short cell (GSSC) phytoliths were
classiﬁed morphologically following the International Code
for Phytolith Nomenclature (ICPN Working Group et al.
2005). Four different morphological variations of the bilo-
bate morphotype within the Poaceae are recognized in this
study (Rossouw 2009). Variant 1 possesses orbicular lobes
that are symmetrical in planar view, and it has a central
portion or neck equal or greater than one third of total length
of body. Variant 2 has a comparatively short central portion
with orbicular to ovate lobes that are symmetrical in planar
view and with the length of its central portion equal to or less
than one third of total length of body. Variant 3 is always
asymmetrical in planar view with the length of its central
portion less than one third of total length of body. This type
is comparable to the irregular complex dumbbell types rec-
ognized by Twiss et al. (1969) and the ‘Other lobate’cate-
gory in Fredlund and Tieszen (1994). The fourth variant, or
Stipa-type bilobate, is a predominantly pooid morphotype,
which appear trapezoidal or tabular in side view with gen-
erally ovate to scutiform lobes (Mulholland 1989; Fredlund
and Tieszen 1994; Rossouw 2009).
Other GSSC-phytoliths that were identiﬁed include
polylobates, commonly produced by the C
subfamily and C
Panicoid species as well as cross and
saddle morphotypes, which are primarily produced in the
Panicoideae and C
respectively. Trapezoidal, rondel and oblong morphotypes
are largely produced by the C
Pooideae and Dan-
thionoideae subfamilies. Trapezoids are six-sided, square or
rectangular silica bodies with few sides parallel (Rossouw
2009). Planar margins are angular and not medially con-
stricted. The trapezoid category is equivalent to types 1b,
1d and 1f in Twiss et al. (1969), the rondel types described
by Mulholland (1989), the conical and pyramidal types in
Fredlund and Tieszen (1994) and rondel types g, h and i in
Thorn (2004). The rondel is cylindrical or semi-cylindrical,
tapers distally, and resembles a truncated cone (Mulholland
1989). It is circular, elliptical or acicular in planar view
(Rossouw 2009). This morphotype compares to type 1a in
Twiss et al. (1969) and the conical type in Fredlund and
Tieszen (1994). The oblong morphotype includes six-sided
silica bodies that are at least twice as long as broad with
parallel or nearly parallel sides. Oblong-shaped phytoliths
are deﬁned as having smooth, sinuous or crenate planar
edges and trapezoidal cross-sections (Rossouw 2009). It
corresponds with types 1c, 1g and 1h in Twiss et al.
(1969), and the “longer forms with more polygonal
cross-sections”in Mulholland (1989, p. 495). There are
also elongated long-cell morphotypes as well as acicular
(trichome) and bulliform morphotypes. All other unidenti-
ﬁed morphotypes, which represent a variety of plant fam-
ilies that include dicotyledons or gymnosperms, were
classiﬁed as “indeterminate”. In this provisional study we
did not attempt to identify this group because we did not
have reference material from the local plants.
Less than one diatom per gram was obtained from only one
of the coprolites and a single diatom was found in 5 g of
deposits of Unit I. The following species were among the
diatoms recorded: Hantzschia amphioxys (Fig. 13.2a), Pin-
nularia borealis (Fig. 13.2b) and Nitzschia sp. (Fig. 13.2c).
Both H. amphioxys and P. borealis are aerophilic taxa
(preferring shallow or running water) and therefore indica-
tive of moist conditions e.g., lake shores, ground water
springs, wetlands or wet soils (Denham et al. 2009).
Table 13.1 Samples of Azokh Cave coprolites and deposits
Unit Depth cm Type Age MIS
5360 Modern soil
<200 yrs 1
233 Coprolite #
290 L. Scott et al.
A fourth diatom type represents Pliocaenicus (Fig. 13.2d),
a genus which has been recorded from a Tertiary freshwater
environment in China (Stachura-Suchoples and Jahn 2009).
The diatom specimen from the Unit 1 is obscured by sediment
but could tentatively be Achnanthes sp., a genus which may
be found in a large variety of environments and therefore least
Both the coprolites and studied cave deposits were very rich
in siliceous phytoliths ranging from well preserved to bro-
ken, etched and damaged. Phytoliths in the coprolite 5246
were more corroded than 5153, while those deposits of
Unit II (5302) from which the latter was derived were better
preserved in comparison with other levels. Examples of the
recorded phytoliths are shown in Fig. 13.3a–f. Counts show
that GSSC-phytoliths (grass silica short cell phytoliths)
make up more than 60% of the total number of phytoliths
counted in coprolite 5153, while 5246 has a lower percent-
age (Fig. 13.4a). The rest consist of indeterminate phytoliths
of various plants which could include a variety of uniden-
tiﬁed dicotyledonous and gymnosperm species and include
hair bases (Fig. 13.3c), trichomes and phytoliths of leaves,
branches or fruits. In view of this high proportion of
unidentiﬁed forms we cannot make an accurate reconstruc-
tion of the vegetation without further research on these
types. The GSSC-phytoliths are composed of several types
(Fig. 13.4b) that include comparatively few bilobate mor-
photypes (Fig. 13.3e) representing less than 2% of the
GSSC-component. Polylobate morphotypes (Fig. 13.3f)
account for less than 3% of the total number of
GSSC-phytoliths counted. The highest frequencies of epi-
dermal short cells in the coprolites are represented by
trapezoidal, rondel and oblong morphotypes (totalling c.
90%) (Fig. 13.3 a, b and d).
Pollen and Other Microfossils
The coprolites were poor in pollen and there was not enough
for determining past vegetation composition. Only two
pollen grains and some possible spores were found in
coprolite 5153. One is Asteraceae, most probably belonging
to Artemisia (not illustrated) of which the morphology is
Fig. 13.2 Diatoms recorded in the two Azokh Cave coprolites,
Hantzschia amphioxys (44 µm) (a), Pinnularia borealis (31 µm) (b),
Nitzschia sp. (54 µm) (c) and Pliocaenicus (44 µm) (d)
Fig. 13.3 Some phytolith types from the Azokh coprolites: Trapezoid
(14 µm) (a); Oblong (32 µm) (b); Basal view of trichome (42 µm) (c);
Rondel (20 µm) (d); Bilobate Variant 2 (24 µm) (e); Polylobate
(26 µm) (f)
13 Coprolites and Plant Microfossils 291
well known, but Echinops or Centaurea cannot be excluded.
The second pollen type (Fig. 13.5a) is as yet unidentiﬁed.
Possible fern or bryophyte spores or algae also occur. The
preservation of the two pollen grains was reasonable enough
to reveal their morphological characteristics suggesting that
their low concentration is possibly not a result of destructive
processes but due to low pollen availability. Their brownish
color suggests that they are indeed fossil and not modern
Long silica structures (longer than 200 microns) were
observed in coprolite 5153. They resemble sponge spicules
(Fig. 13.5b) and are indicative of ﬂooding or could have
been derived from drinking water. In soils outside caves, this
is indicative of ﬂooding from a stream. Azokh 1 is seasonally
wet (Murray et al. 2016; Marin-Monfort et al. 2016), and
they may have formed at this time, but when the cave is dry,
they may have entered the animal’s alimentary tract through
drinking water. Also recorded in coprolite 5153 are com-
paratively large radial leaf trichomes of c. 174 and 200
micron diameters (Fig. 13.5b, c). Microscopic charcoal
occurred in small numbers as brown and black woody plant
remains (Fig. 13.5d) derived from occasional natural ﬁres or
ﬁne aerial dust ingested accidentally inside or outside the
cave from hearths. A spiny “corpuscle”(Fig. 13.5e) in
coprolite 5153 is similar to structures that have been seen in
hyena coprolites from sites in South Africa (L. Scott
unpublished data), and in a picture of an unknown structure
derived from a stool (http://www.dpd.cdc.gov/dpdx/html/
Fig. 13.4 Diagrams of phytoliths from Azok Cave surface soil, coprolites and deposits showing the ratio between GSSC (grass silica short cell)
phytoliths and indeterminate phytoliths (a), and the percentages of different types of GSSC phytolith types (b)
Fig. 13.5 Other microfossil types in the Azokh coprolites, unidenti-
ﬁed pollen (22 µm) (a), trichome (174 µm) (left) and sponge spicule
(218 µm) (right) (b), trichome (c. 200 µm) (c), woody ﬁbre (38 µm)
(d) and unidentiﬁed corpuscule (44 µm) (e)
292 L. Scott et al.
Unit II dates from 100 to 184 kyr according to ESR dating
(see Appendix and Murray et al. 2016) covering Marine
Isotope Stages 5 and 6. According to the available dates the
studied coprolites are from the very top part of the unit
(dated in 100 ±7 kyr) and belong to Stage 5 (Table 13.1)
but the exact age cannot be determined more precisely. They
are therefore more likely to represent a warm or a stadial
phase than a glacial period.
The present-day river near the area, the terraces of which
indicate that it was active in the Pleistocene (Murray et al.
2016), could be a potential source of the diatoms in the one
coprolite. Phytoliths in the coprolite 5246 were more cor-
roded than those in 5153 and its associated Unit II deposits,
which contained better preserved silica than other deposits.
The difference in preservation quality is difﬁcult to explain,
except for the indication of guano deposits during and after
burial. Corrosion might have resulted from harsh conditions
in the surroundings before the phytoliths were accidentally
ingested by the animal (as dust), or it might have occurred
later under ﬂuctuating water tables or dampness that affected
the silica inside the coprolite in the cave. It is known that at
present such ﬂuctuations do occur, and it is likely that they
also occurred in the past. Corrosion could have been
enhanced further by the corrosive qualities of bat guano.
Indications of guano and damp/dry ﬂuctuations at the cave
interior is indicated by secondary mineral formation, such as
tinsleyite, sepiolite, gypsum, ardelite or brushite (Magela da
Costa and Rúbia Ribeiro 2001; Marincea et al. 2002; White
and Culver 2012) detected by X-ray ﬂuorescence (XRF) and
X-ray diffraction (XRD) in both the sediment and the fossils
(Marin-Monfort et al. 2016).
Habitat structure inferred through a comparison of the
contribution from GSSC phytoliths versus non-grass phy-
toliths (e.g., Alexandre et al. 1997; Bremond et al. 2005)
points to grassy conditions in the region at the time when the
coprolites were formed, although the density of woody
components cannot be determined. However, the two
coprolites differ in content, and the more non-grass inclusions
in coprolite 5246 could be related to seasonal factors or could
simply be due the possibility that the coprolites represent
different habitats in which animals roamed (Fig. 13.4). As is
indicated by the non-grass silica like epidermal cells or other
round “blocky”phytoliths, several different unidentiﬁed
plant types could be included in the coprolite assemblages.
As can be inferred from the charcoal evidence (Allué2016)
woody species must have occurred locally, especially Pru-
nus. Phytoliths of this genus, which are not produced in
fruits, leaves and inﬂorescences of some species (Kealhofer
and Piperno 1998), were not identiﬁed, partly because their
morphologies are not known (Rovner 1971).
The proportion of GSSC–phytoliths versus indeterminate
silica bodies in the coprolite-bearing deposits is similar to
that of the Holocene deposits, but the recent soil shows a
lower proportion of grasses, which is typical of an over-
grazed area like that around the cave at present.
Some are taxonomically and ecologically signiﬁcant. The
underrepresentation of saddle and bilobate phytoliths and
comparatively high frequencies of trapezoidal and oblong
morphotypes recorded in Unit II clearly suggests that C
grassy conditions prevailed at the time when the coprolites
were formed and the place where they were ingested outside
the cave. This is supported by the presence of polylobates
recorded throughout Unit II. Polylobates are recorded in at
least 25% (n = 31) of modern Pooid species (Rossouw
2009). The phytoliths also indicate the presence of other
plants which can at present not be identiﬁed.
The surrounding area at the time of the coprolite pro-
duction could also have undergone dry summers resembling
that of alpine meadows of the Crimean Mountains or that of
the cold-dry steppe (winter-rain) of South Jordan (Cordova
2011). Because of the presence of Stipa-type in an area
where Paniceae and Danthonioideae are rare, the occurrence
of grasses of the Stipeae tribe, most of which reﬂect cold and
dry continental climates, is suggested.
The coprolite phytolith assemblages only give a reﬂection
of what is available in the environment and not necessarily
of the actual proportions of plant types. Potential bias in ratio
towards more grasses in the GSSC in relation to unidentiﬁed
silica in the coprolite samples is plausible in view of possible
selective consumption of grasses by carnivores as is recor-
ded in ecological studies worldwide (Skinner 1976;de
Arruda Bueno et al. 2002). However, comparison with the
phytoliths in the surrounding deposits of Unit II does not
suggest any marked bias. The cave deposit samples from the
Unit II sediment may be a more unbiased reﬂection of the
vegetation in the immediate surroundings than the coprolite
because they do not favor behavioral selection from a wider
The oblong/trapezoid phytolith ratios between the
coprolites and surrounding deposits differ slightly with more
oblong types in the latter. This could be from widely
roaming animals trapping phytoliths in their dung and not
from the local slopes next to the cave (as represented by the
cave deposits). In comparison to present conditions as
reﬂected by the modern sample outside the cave, oblong
types are more prominent but it is not possible to say if this
is due to climate a different climate or modern grazing dis-
turbance. The assemblages that occur in Unit I deposits
during the Holocene compare well with those in Unit II,
suggesting that climates did not differ markedly.
On the basis of other evidence the habitat varied
(Andrews et al. 2016). The large mammals and charcoal
indicate deciduous woodland while small mammals,
13 Coprolites and Plant Microfossils 293
amphibians and reptiles indicate open steppe environments.
The taphonomy of the latter group suggests that they were
probably brought to the cave from a distance by predators in
a setting similar to the present, where woodland occurs in the
vicinity of the cave and steppe not too far away. Therefore it
is not impossible that woodland existed similar to the veg-
etation that can potentially develop in the area today under
current climatic conditions and no agricultural disturbance.
1. Pollen was extremely rare in the two carnivore coprolites
investigated, and none was found in the sediments. The
lack of pollen is probably due to environmental condi-
tions and the location of the excavation 40 m into the
Azokh 1 passageway.
2. Phytoliths were abundant in the coprolites and in the
deposits of Unit II. Nine different grass silica short cell
(GSSC) phytolith types were identiﬁed, and these indi-
cate that the vegetation type was most likely a temperate
-grass steppe mosaic.
3. Phytoliths other than those of grasses were recorded and
they could have been derived from local woodland.
Caution is needed with the interpretation of the openness
of the vegetation in view of the unknown degree of
possible selection of phytoliths by the carnivore and due
to the limitation that a large number of the phytoliths
were not identiﬁed.
4. The few diatoms recovered suggest the availability of
5. Long silica structures (longer than 200 microns) were
observed in one of the coprolites. They resemble sponge
spicules and indicate wet conditions.
6. The discovery of numerous phytoliths show that the
Azokh deposits have great potential for a phytolith study
and interpreting environmental conditions throughout the
whole Azokh sequence. A more detailed analysis can
therefore be undertaken beyond the scope of this study.
The potential is demonstrated in deposits at the older
Dmanisi site in the Georgian Caucasus that contain
comparable phytolith assemblages indicating marked
changes in water stress in the region (Messager et al.
Acknowledgments We thank Yolanda Fernández-Jalvo for providing
the coprolites, initiating the study and providing relevant information.
We are also grateful to the authorities of Nagorno-Karabakh for the
support and permissions to work on these specimens. We are grateful to
Tania King and diggers for careful work collecting these fossils, as well
as ﬁeld assistants for modern soil sampling on the slope of the cave.
Thanks are extended to Karen Hardy for collecting sediment samples
from the section of Azokh.
Alexandre, A., Meunier, J. D., Lezine, A. M., Vincens, A., & Schwarz,
D. (1997). Phytoliths: Indicators of grassland dynamics during the
late Holocene in intertropical Africa. Palaeogeography, Paleocli-
matology, Palaeoecology, 136, 213–229.
Allué, E. (2016). Charcoal remains from Azokh 1: Preliminary results.
In Y. Fernández-Jalvo, T. King, L. Yepiskoposyan & P. Andrews
(Eds.), Azokh Cave and the Transcaucasian Corridor (pp. 297–
304). Dordrecht: Springer.
Andrews, P., Hixson Andrews, S., King, T., Fernández-Jalvo, Y., &
Nieto-Díaz, M. (2016). Paleoecology of Azokh 1. In Y. Fernández--
Jalvo, T. King, L. Yepiskoposyan & P. Andrews (Eds.), Azokh Cave
and the Transcaucasian Corridor (pp. 305–320). Dordrecht: Springer.
Appendix: Fernández-Jalvo, Y., Ditchﬁeld, P., Grün, R., Lees, W.,
Aubert, M., Torres, T., et al. (2016). Dating methods applied to
Azokh cave sites. In Y. Fernández-Jalvo, T. King, L. Yepisko-
posyan & P. Andrews (Eds.), Azokh Cave and the Transcaucasian
Corridor (pp. 321–339). Dordrecht: Springer.
Bremond, L., Alexandre, A., Peyron, O., & Guiot, J. (2005). Grass
water stress estimated from phytoliths in West Africa. Journal of
Biogeography, 32, 311–327.
Brown, D. A. (1984). Prospects and limits of a phytolith key for grasses
in the central United States. Journal of Archaeological Science, 11,
Carrión, J. S., Gil, G., Rodríguez, E., Fuentes, N., García-Antón, M., &
Arribas, A. (2005). Palynology of badger coprolites from central
Spain. Palaeogeography, Palaeoclimatology, Palaeoecology, 226,
Carrión, J. S., Scott, L., & Marais, E. (2006). Environmental
implications of pollen spectra in bat droppings from south-eastern
Spain and potential for palaeoenvironmental reconstructions.
Review of Palaeobotany and Palynology, 140, 175–186.
Carrión, J. S., Scott, L., Arribas, A., Fuentes, N., Gil, G., & Montoya,
E. (2007). Pleistocene landscapes in Central Iberia inferred from
pollen analysis of hyena coprolites. Journal of Quaternary Science,
Clayton, W. D., & Renvoize, S. A. (1986). Genera Graminum. Grasses
of the World. London: Her Majesty’s Stationary Ofﬁce.
Coles, G. M., & Gilbertson, D. D. (1994). The airfall-pollen budget of
archaeologically important sites: Creswell Crags, England. Journal
of Archaeological Science, 21, 735–755.
Cordova, C. (2011). The Stipa-type short cell. What does it mean
taxonomically, climatically and ecologically? Proceedings of the
8th International Meeting on Phytolith Research. Estes Park,
Cross, R. A. (1980). Distribution of sub-families of Gramineae in the
old world. Kew Bulletin, 35, 279–289.
de Arruda Bueno, A., da Silva, C., Belentani, S., & Carlos
Motta-Junior, J. (2002). Feeding ecology of the maned wolf,
Chrysocyon brachyurus (Illiger, 1815) (Mammalia: Canidae), in the
ecological station of Itirapina, Sao Paulo State. Brazil. Biota
Neotropica, 2(2), 1–9.
Denham, T., Sniderman, K., Saunders, K. M., Winsborough, B., &
Pierret, A. (2009). Contiguous multi-proxy analyses
(X-radiography, diatom, pollen, and microcharcoal) of Holocene
294 L. Scott et al.
archaeological features at Kuk Swamp, Upper Wahgi Valley, Papua
New Guinea. Geoarchaeology: An.International Journal, 24, 715–
Fernández-Jalvo, Y., King, T., Andrews, P., & Yepiskoposyan, L.
(2016). Introduction: Azokh Cave and the Transcaucasian Corridor.
In Y. Fernández-Jalvo, T. King, L. Yepiskoposyan & P. Andrews
(Eds.), Azokh Cave and the Transcaucasian Corridor (pp. 1–26).
Fernández-Jalvo, Y., King, T., Andrews, P., Yepiskoposyan, L.,
Moloney, N., Murray, J., et al. (2010). The Azokh Cave complex:
Middle Pleistocene to Holocene human occupation in the Caucasus.
Journal of Human Evolution, 58, 103–109.
Fredlund, G. G., & Tieszen, L. T. (1994). Modern phytolith assem-
blages from the North American Great Plains. Journal of Biogeog-
raphy, 21, 321–335.
Gabrielian, E., & Fragman-Sapir, O. (2008). Flowers of Transcaucasus
and adjacent areas: Including Armenia, Eastern Turkey, Southern
Georgia, Azerbaijan and Northern Iran. Ruggell: Gantner Verlag.
Gulisashvili, V. Z., Makhatadze, L. B., & Prilipko, L. I. (1975).
Rastitel’nost′Kavkaza. Moskva: Nauka. The vegetation of the
Caucasus (Trans: Russian). http://www.rusnature.info/reg/15_6.htm.
Hunt, C. O., & Rushworth, G. (2005). Airfall sedimentation and pollen
taphonomy in the West mouth of the Great Cave, Niah. Journal of
Archaeological Science, 32, 465–473.
ICPN Working Group, Madella, M., Alexandre, A., & Ball, T. (2005).
International Code for Phytolith Nomenclature 1.0. Annals of
Botany,96 (2), 253–260.
Kealhofer, L. & Piperno, D. R. (1998). Opal Phytoliths in Southeast
Asian Flora. Smithsonian Contributions to Botany 88, Washington,
D.C. Smithsonian Institution Press.
Magela da Costa, G., & Rúbia Ribeiro, V. (2001). The occurrence of
tinsleyite in the archaeological site of Santana do Riacho, Brazil.
Mineralogical Society of America, 86, 1053–1056.
Marincea, S., Dumitras, D., & Gibert, R. (2002). Tinsleyite in the “dry”
Cioclovina Cave (Sureanu Mountains, Romania): The second
occurrence. European Journal of Mineralogy, 14, 157–164.
Marin-Monfort, M. D., Cáceres, I., Andrews, P., Pinto, A. C., &
Fernández-Jalvo, Y. (2016). Taphonomy and Site Formation of
Azokh 1. In Y. Fernández-Jalvo, T. King, L. Yepiskoposyan &
P. Andrews (Eds.), Azokh Cave and the Transcaucasian Corridor
(pp. 211–249). Dordrecht: Springer.
Messager, E., Lordkipanidze, D., Delhon, C., & Ferring, C. R. (2010).
Palaeoecological implications of the Lower Pleistocene phytolith
record from the Dmanisi Site (Georgia). Palaeogeography, Palaeo-
climatology, Palaeoecology, 288,1–13.
Mulholland, S. C. (1989). Phytolith shape frequencies in North Dakota
grasses: A comparison to general patterns. Journal of Archaeolog-
ical Science, 16, 489–511.
Murray, J., Lynch, E. P., Domínguez-Alonso, P., & Barham, M.
(2016). Stratigraphy and sedimentology of Azokh Caves, South
Caucasus. In Y. Fernández-Jalvo, T. King, L. Yepiskoposyan &
P. Andrews (Eds.), Azokh Cave and the Transcaucasian Corridor
(pp. 27–54). Dordrecht: Springer.
Navarro, C., Carrión, J. S., Munuera, M., & Prieto, A. R. (2001).
A palynological study of karstic cave sediments on the basis of their
potential for palaeoecological reconstruction. Review of Palaeob-
otany and Palynology,117, 245–265.
O’Rourke, M. K., & Mead, J. (1985). Late Pleistocene and Holocene
pollen records from two caves in the Grand Canyon of Arizona,
USA. American Association of Stratigrapphic Palynologists, 16,
Republic of Armenia. (1999). A country study on the biodiversity of
Armenia. First National Report to the Convention on Biological
Diversity. Yerevan: Ministry of Nature Protection.
Rossouw, L. (2009). The application of fossil grass-phytolith analysis
in the reconstruction of Cainozoic environments in the South
African interior, PhD dissertation. University of the Free State,
Rovner, I. (1971). Potential of opal phytoliths for use in palaoecological
reconstruction. Quaternary Research, 1, 343–359.
Rovner, I. (1983). Plant phytolith analysis: Major advances in
archaeobotanical research. Advances in Archaeological Method
and Theory, 6, 225–266.
Scott, L. (1987). Pollen analysis of hyena coprolites and sediments
from Equus Cave, Taung, Southern Kalahari (S. Africa). Quater-
nary Research, 28, 144–156.
Scott, L., Fernández-Jalvo, Y., Carrión, J. S., & Brink, J. S. (2003).
Preservation and interpretation of pollen in hyena corprolites:
Taphonomical observations from Spain and Southern Africa.
Palaeontologia Africana, 39,83–91.
Sharrow, S. H. (2007). Natural resource management on the other side
of the world: The Nagorno Karabakh Republic. Society for Range
Skinner, J. D. (1976). Ecology of the brown hyena Hyaena brunnea in
the Transvaal with a distribution map for southern Africa. South
African Journal of Science, 72, 262–269.
Stachura-Suchoples, K., & Jahn, R. (2009). Middle Miocene record of
Plioaenicus changbaiense sp nov. from Changbai (Jilin Province,
China). Acta Botanica Croatica, 68, 211–220.
Thompson, R. S., Van Devender, T. S., Martin, P. S., Foppe, T., &
Long, A. (1980). Shasta ground sloth (Nothrotheriops shastense
Hoffstetter) at Shelter Cave, New Mexico: Environment, diet, and
extinction. Quaternaty Research, 14, 360–376.
Thorn, V. C. (2004). Phytolith evidence for C
since the early Holocene at Long Pocket, northeast Queensland,
Australia. Quaternary Research, 61, 168–180.
Twiss, P. C., Suess, E., & Smith, R. M. (1969). Morphological
classiﬁcation of grass phytoliths. Proceedings of the Soil Science
Society of America, 33, 109–115.
White, W. B., & Culver, D. C. (2012). Encyclopedia of Caves.
13 Coprolites and Plant Microfossils 295