Content uploaded by John F Hoffecker
Author content
All content in this area was uploaded by John F Hoffecker on Apr 05, 2020
Content may be subject to copyright.
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
1
Geoarchaeological and bioarchaeological studies at Mira,
an early Upper Paleolithic site in the Lower Dnepr Valley, Ukraine
_____________________________________________________________________________
John F. Hoffecker1*, Vance T. Holliday2, Vadim N. Stepanchuk3, Alexis Brugère4,
Steven L. Forman5, Paul Goldberg6, Oleg Tubolzev7, and Igor Pisarev8
Abstract. New geoarchaeological and bioarcheological research was undertaken at the open-air
site of Mira, which is buried in deposits of the Second Terrace of the Dnepr River, roughly 15
km downstream from the city of Zaporozhiye in Ukraine. Previous excavation of the site
revealed two occupation layers dating to ~32,000 cal BP. The lower layer (II/2) yielded
bladelets similar to those of the early Gravettian, while the upper layer (I) contained traces of an
artificial shelter and hundreds of bones and teeth of horse (Equus latipes). Mira represents the
only firmly dated early Upper Paleolithic (EUP) site in the Dnepr Basin, and occupies a unique
topographic setting for the EUP near the center of the broad floodplain of the Dnepr River. The
site was visited during a period of floodplain stability, characterized by overbank deposition and
weak soil formation under cool climate conditions. Mira was used as a long-term camp, but also
was the locus of large-mammal carcass-processing associated with a nearby kill of a group of
horses (Layer I).
______________________________________________________________________________
1Institute of Arctic and Alpine Research, University of Colorado at Boulder, 1560 30th Street, Boulder, Colorado
80309-0450 USA
2Departments of Anthropology and Geosciences, University of Arizona, PO Box 210030, Tucson, Arizona 85721-
0030 USA
3Stone Age Department, Institute of Archaeology, Ukrainian Academy of Sciences, Heroes of Stalingrad Avenue 12,
Kiev 04210, Ukraine
4Maison de l’Archéologie et de l’Ethnologie, CNRS UMR 7041 “Archéologies environmentales” 21 allée de
l’Université. 92023 Nanterre Cedex - France
5Luminescence Dating Research Laboratory, Department of Earth and Environmental Sciences, University of
Illinois, Chicago, Illinois 80607-7059 USA
6Department of Archaeology, Boston University, Boston, Massachusetts 02215 USA
7Novaya Arkheologicheskaya Shkola, Pr. Metallurov 22/2, Zaporozhye 69006 Ukraine
838, 2a Gagarin Street, Zaporozhye 69057 Ukraine
Geoarchaeology: An International Journal 29: 61–77 (2014)
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
2
INTRODUCTION
Although there are isolated traces of the Upper Paleolithic on the East European Plain that
antedate 40,000 cal BP (e.g., Hoffecker et al., 2008), most sites classified as “early Upper
Paleolithic” (EUP) date to between 40,000 and 30,000 cal BP. They are contemporaneous to
classic EUP industries in Western Europe, such as the Aurignacian. The density of EUP sites on
the East European Plain is significantly lower than that in Western Europe, however, probably
owing to the scarcity of natural shelters on the former, especially the central plain. All or almost
all known EUP occupations of the East European Plain are open-air sites.
Despite comparatively low density, these sites yield a wealth of information on their occupants
that suggest both similarities and differences with their EUP contemporaries in Western Europe.
Lithic assemblages assigned to the Aurignacian are known but rare (Noiret, 2005) and an early
Gravettian industry is present (e.g., Kostenki 8, Layer II [Anikovich Anisyutkin, & Vishnyatsky,
2007]), while East European archaeologists have defined several local EUP industries that are
unknown in Western Europe (e.g., Gorodtsovan [Rogachev & Anikovich, 1984]).
Representational art is less common than it is in Western Europe, but the EUP of the East
European Plain contains spectacular burials (i.e., Sungir’ [Bader & Bader, 2000]). It also reveals
traces of artificial shelters and sites at which groups of large mammals were killed and butchered
(e.g., Kostenki 15), both of which probably reflect the dominance of open-air localities
(Hoffecker et al., 2010).
The discovery and investigation of the Mira site in the Lower Dnepr Valley of south-central
Ukraine added a new dimension to the EUP of Eastern Europe. Mira remains the only firmly
dated EUP site in the Dnepr Basin (Stepanchuk et al., 2009). Its geomorphic context—buried 10
meters below the surface of the Second Terrace under a mass of fluvial and eolian deposits—
helps explain why EUP sites are so scarce in the region. Mira was situated near the center of the
wide Dnepr River floodplain, which is a unique setting for the EUP. Nevertheless, it reveals a
familiar pattern for EUP sites, including evidence for the butchering of a large group of
mammals and traces of an artificial shelter. It also contains artifacts of the early Gravettian, as
well as an assemblage similar to those assigned to the local Gorodtsovan industry (Stepanchuk,
2005).
THE MIRA SITE: LOCATION AND RESEARCH HISTORY
The Mira site is located on the Lower Dnepr River in south-central Ukraine at 47°40’ North 34°50’
East (Figure 1). The site is found on the west bank of the river near the village of Kanevskoye,
roughly 15 km south of the city of Zaporozhye (which represents the dividing point between the
middle and lower segments of the Dnepr River). It rests on a low terrace that is ~30 meters above
the river, and is approximately 40 meters above mean sea level (Figure 2).
The Mira site was discovered in 1995 by I. B. Pisarev, who initially investigated the site in 1995–
1996. In 1997, a test trench was excavated into the deposits, revealing the stratified layers of
Paleolithic artifacts (Stepanchuk et al., 1998, 2004). Under the direction of V. N. Stepanchuk,
excavations were conducted in 2000, and some additional research on the stratigraphy was
performed in 2001, exposing a total of 60 m2 for the years 1997–2001. Some small-scale
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
3
excavations were undertaken in 2004–2005 and 2008–2009, exposing an additional 10 m2. A
stratigraphic profile was recorded by P. Haesaerts and N. Gerasimenko (Stepanchuk et al., 2004;
Stepanchuk, 2005: 25, fig. 2).
Figure 1. Map of Eastern Europe, showing the location of Mira
and other major EUP sites in the region.
New field and laboratory research was performed at Mira during 2012. A stratigraphic profile over
11 m deep was exposed at the site, from the top of the second terrace to below the occupation zone.
Approximately 5 m2 of deposits containing the Upper Paleolithic occupation layers were excavated.
Sediment samples were collected and analyzed for OSL dating and soil micromorphology; charcoal
fragments were collected for new radiocarbon dating. A section in a ravine (Glubokii Yar) incised
into the second terrace 500 m west of the Mira section (see Figure 2) also was examined and
recorded (but not sampled) for stratigraphic comparison. Many of the faunal remains recovered
during earlier years were examined at the Institute of Archaeology, Ukrainian Academy of Sciences
in Kiev. The results of the new research are presented below (following the description of the
occupation layers excavated during 1997–2009).
Upper Paleolithic Occupation Layers
Excavations at Mira during 1997–2009 revealed two occupation layers that are dated to the later
phase of the EUP (~32,000 cal BP). The upper level (Layer I) produced more than 50,000 artifacts
and a substantial quantity of associated faunal debris, along with traces of hearths and a former
dwelling structure. The lower occupation level (Layer II/2) yielded a comparatively small quantity
of lithics and faunal debris. An intermediate level (Layer II/1) was archaeologically sterile, but
yielded fragments of charred wood. Although isolated lithics were found above Layer I, they may
represent items displaced upward from the main cultural layer (potentially by frost action).
A human tooth fragment was recovered from Layer I in association with traces of a former structure
(see description of features below). The fragment represents a portion of the crown of either the
first or second upper molar, possibly of a young adult, and is assigned to Homo sapiens
(identification by C. G. Turner) (Stepanchuk, 2005: 28).
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
4
Figure 2. Satellite image of the Mira site and immediate
setting. The level of the Dnepr River has been elevated
roughly 1 meter by a reservoir and currently is commensurate
with the first terrace scarp.
Layer I
The artifacts recovered from Layer I are largely composed of dark-colored chert imported from
the Eastern Carpathian Mountains over a distance of > 500 km (Stepanchuk, 2005: 28). The
lithic assemblage comprises almost 53,000 items (mostly small flakes or “chips”). Only 2 cores
(and 5 chert fragments) are reported; 94% of the unretouched blanks are flakes (n = 579), but
some blades and bladelets also are present (Stepanchuk, 2005: 29–30).
The most common retouched tools (1.4%) include non-geometric microliths (n = 138), end-
scrapers (n = 36), scaled pieces (n = 20), points (n = 18), side-scrapers (n = 17), points (n = 18),
Dufour bladelets (n = 15), combination tools (n = 15), and Mousterian points (n = 13). The side-
scrapers include simple, double, canted, convergent, and double convergent forms. The Dufour
bladelets are characterized as “atypical” (Stepanchuk, 2005: 34). Also common are retouched
flakes (n = 81), retouched bladelets (n = 50), and retouched blades (n = 45). There are >400
“atypical bladelets,” which include twisted and curved forms. Less common items include
borers (n = 8), micro-points (n = 7), and burins (n = 5). Among bifacial implements are bifaces
(complete = 4), leaf-shaped points (including tip fragments), and others (Stepanchuk, 2005: 30–
36). The combined presence of Middle and Upper Paleolithic technology and tool types is not
uncommon in the later EUP of the East European Plain, especially among assemblages assigned
to the Gorodtsovan Culture (e.g., Rogachev & Anikovich, 1984: 183–186).
Among the non-stone artifacts are two small point tips (awls?), one of which is of antler (?), a
possible mid-section fragment of a needle (bone), roughly 40 retouchers (bone), 10 perforated
carnivore teeth or ornaments, several fragments of amber, and several other items (Stepanchuk,
2005: 36–37). Features on the Layer I occupation floor include 4 former hearths, 6 small pits,
and an arrangement of 12 paired post-holes that apparently delineate the boundary of a former
artificial shelter (sub-rounded in plan). Traces of the post-holes, which range 3–16 cm in
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
5
diameter, extend into the underlying level (Layer II/1). The dwelling is estimated to have
covered an area of 14.4 m2 (Stepanchuk, 2005: 28) (Figure 3).
Figure 3. Occupation floor of EUP Layer I at Mira, illustrating the distribution of
artifacts, faunal remains, and other debris excavated in 1997–2001 (from V.N.
Stepanchuk). The stratigraphic position of both EUP occupation layers is indicated by
“6” in Figure 4.
Layer II/2
Although excavated over a comparable area to that of Layer I (~60 m2), the lower occupation
layer yielded only ~200 stone artifacts. The source area for the raw material lies in western
Ukraine, 300–350 km from Mira (Stepanchuk, 2005: 27). Retouched pieces are confined to five
complete and several fragmentary backed bladelets, an end-scraper, and two fragments of flake
tools (Stepanchuk, 2005: 28).
GEOARCHAEOLOGY
Stratigraphy
The Mira site is buried in alluvium of the Second Terrace of the Dnepr River along the
uppermost segment of the Lower Dnepr Valley. At the location of the site (~15 km downstream
from Zaporozhye), the Second Terrace is ~30 meters above the river level, which is elevated by
~1 meter as a result of a downstream dam and reservoir. Although the Second Terrace was
assigned to the Middle Pleistocene by Goretskii (1970) and others, it is clear that the alluvial
facies accumulated during the Late Pleistocene (Matoshko, Gozhik & Ivchenko, 2002: 345–349).
The uppermost portion of the terrace is composed of primary or reworked eolian deposits that
apparently date to the Last Glacial Maximum and Late Glacial (MIS2 age equivalent) (Figure 4).
The stratigraphic profile section exposed in 2012 consisted of ~11 meters of alluvium dominated
by sand and silt along with a number of buried soils (see below). The profile is unique for the
EUP of Eastern Europe, and a detailed description is presented in Table I. The complete absence
of sediments >2 mm in diameter (i.e., larger than sand size) indicates that the site was on the
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
6
floodplain throughout aggradation, from before the time of Upper Paleolithic occupation until
after the incision that left the surface abandoned as a terrace.
Figure 4. The Second Terrace of the Dnepr River at the location of the Mira
site. Numbers refer to portions of the stratigraphic sequence depicted in specific
figures (5A = see Figure 5A; 5B = see Figure 5B; 6 = see Figure 6) (photograph
by JFH, August 2012). The vertical distance of the line marked “5A” is 6
meters.
Most of the section, from ~240 cm below the surface to the base of the exposure (at a depth of
1100 cm) is composed of sand, mostly well-sorted fine sand. This texture suggests prolonged
sedimentation not affected by extreme perturbations in discharge or flow regime. Evidence of
iron oxidation and reduction is common and indicates that the water table was high, probably
immediately below the surface, during much of the aggradation of the lower part of the section
(below ~600 cm). The presence of secondary calcium carbonate in soils above ~600 cm reveals
a deeper water table in the upper portion of the profile, probably due to rapid accumulation of
sediment. From 0 to 240 cm below the modern surface, more silt is mixed with the sandy
sediment. Whether this represents primary airfall loess (“sandy loess”) or sediment redeposited
from sandy loess on the uplands is unclear (Figure 5).
Below the modern soil formed in primary or reworked eolian sediments (0–166 cm), buried soils
and pedocomplexes (PK) were recorded in the section at depths of: (1) 50-68 cm (b1 soil); 120-
166 cm (b2 soil), 240–325 cm (multiple A-Bt and A-Bw profiles of PKb3); (2) 430–475 cm
(multiple Bk horizons of PKb4); (3) 575–598 cm (multiple A-C profiles of PKb5); (4) 605–622
cm (b6 soil); and (5) ~958 to 1005 cm (Agb7 and Agb8, formed under high water-table
conditions). They represent periods of floodplain stability, probably of variable duration based
on degree of expression, perhaps because the stream channel shifted away from the area of the
section.
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
7
The Upper Paleolithic occupation layers (at ~958 to ~1005 cm below the surface) are buried
within two green/gray clay-rich zones that exhibit traces of soil development (Agb soil horizons;
see soil micromorphology description below). At the time of occupation, these zones probably
were damp, relatively stable surfaces with abundant vegetation that trapped fines (dust?). The
greenish (gleyed) color suggests water-logging, probably due to the presence of a water table
immediately below the surface (Figure 6).
The ravine exposure located west of the site (Glubokii Yar) revealed a profile similar to that
described at Mira, and suggests that the latter represents a regional alluvial record. The
combined profiles provide a regional picture of alluvial history in the later Pleistocene. Further,
the location of the section, several km from either valley wall or the Third Terrace, shows that
the Upper Paleolithic occupants of Mira were making use of a very broad floodplain (Figure 7).
Radiocarbon Dating
A total of 14 radiocarbon dates were obtained on samples collected from Mira during earlier
excavations (1997–2001), and the results already have been published (Stepanchuk, 2005: 26,
table 1). The dates indicate that, although divided by a sterile layer (Layer II/1), the two Upper
Paleolithic layers (Layers I and II/2) were occupied at roughly the same time—temporally
separated by a few centuries or decades. Both occupations date to ca. 32,000–31,000 cal BP and
may be correlated with the Greenland Stadial 5 (GS 5) cool-climate period in the Greenland ice-
core record (e.g., Weninger and Jöris, 2008: 773–776).
In 2012, new samples (wood charcoal) were collected for radiocarbon dating, and the results are
presented in Table II. Sample preparation was performed at the INSTAAR Radiocarbon
Laboratory (University of Colorado at Boulder) and the AMS ages were obtained from the
University of California at Irvine (USA) accelerator. Although the date on archaeologically
Figure 5A. Stratigraphic profile at
the Mira site: upper portion of
sequence to a depth of 650 cm
below modern surface. Numbers
refer to depth in centimeters.
Letters refer to soil horizons (e.g.,
Ab) and Pedocomplex (PK). OSL
ages are given for the upper portion
of the sequence on the left (photo-
graph by JFH, August 2012).
Figure 5B. Stratigraphic profile at
the Mira site: lower portion of
sequence between 650 and 950 cm
below modern surface. Numbers
refer to depth in centimeters.
Letters refer to soil horizons (e.g.,
Ab) and Pedocomplex (PK) (photo-
graph by JFH, August 2012).
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
8
sterile Layer II/1 is anomalously young and presumably reflects sample contamination, the dates
on the two occupation layers conform well to the dates obtained earlier from other laboratories,
and further strengthen the chronology of the Mira site.
Figure 6. The Upper Paleolithic occupation layers at Mira. Layer numbers refer
to archaeological, not lithological, units. The locations sampled for micro-
morphology are indicated by “TS” (photograph by VNS, August 2012). This area
of the site was excavated in 2012 (e.g., unit 31Г) and is located at the west end of
the previously excavated area shown in Figure 3.
OSL Dating
During 2012, sediment samples were collected from the exposed stratigraphic profile for Optical
Stimulated Luminescence (OSL) dating. Samples were taken from both the occupation layers and
at several depths in the overlying deposits of the Second Terrace. The units overlying the
occupation layers were dated by OSL because materials suitable for radiocarbon dating are absent in
these sediments, while OSL dating of the occupation layers was undertaken to supplement the
radiocarbon dating and provide an additional check on the OSL dates from the younger levels. OSL
samples were processed at the Luminescence Dating Research Laboratory, Department of Earth
and Environmental Sciences, University of Illinois at Chicago (USA).
From the sediment samples, 150–250 micron quartz grains were extracted. Small aliquots (200–500
grains/aliquots) were used for dating. OSL ages were determined by the single aliquot regeneration
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
9
(SAR) method from analyses of 30 aliquots, and most aliquots were used in the final age
calculation. The data showed high precision with overdispersion (OD) values </= 20% (at two
sigma errors), which indicates a single population of grain ages, best modeled as a log-normal
distribution. One sigma errors are 7–9%, reflecting the error in the moisture content and the number
of aliquots used in the age calculation. (If 60 or 90 aliquots are measured, the error may drop by 2–
3%; the statistically lowest errors possible for OSL ages are about 4–5%.) The environmental dose
rate for the samples was low, ~ 1.6, 1.0 and 0.70 mgrays/year, reflecting low K, U, and Th content.
The cosmic dose is an appreciable component (10–25%) of this quantity, and was calculated from
the latitude/longitude, elevation, and depth of the sample.
Figure 7. The Third Terrace of the Dnepr River (west bank), photographed at a location ~15 km southwest
of Mira on the modern surface of the Second Terrace, illustrates the width of the floodplain at the time of
occupation (~32,000 cal BP) (photograph by JFH, August 2012).
The OSL dating results are presented in Table III. The two dates from the upper portion of the
stratigraphic profile (collected at 2.7 m and 4.7 m below the modern surface of the Second Terrace,
respectively) may indicate relatively rapid aggradation of the Dnepr River floodplain during early
and middle MIS 2-age equivalent (Last Glacial Maximum), followed by slower accumulation of
sediment after ~20,000 years ago. The dates from the lower portion of the profile were obtained on
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
10
samples from the EUP occupation layers, both of which are ~10 m below the surface. The dates
(21,990±1925 yr; 27,665±2430 yr) were run with differing estimates of sediment moisture content
and overlap at the two-sigma level (Figure 8).
Figure 8. Luminescence data of quartz aliquots for sample UIC3338: (a)
regenerative dose response curve with inset figures show representative photon
decay curves for the natural with exposure to blue light from diodes (470 +/- 20
nm); (b) radial plot of equivalent dose data and lines indicates 2 sigma limits.
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
11
Soil Micromorphology
In order to better understand their depositional setting and immediate environment, the Agb8 soil
with Upper Paleolithic occupation layers (including intermediate Layer II/1) were sampled for
soil micromorphology analysis (e.g., Courty et al., 1989). Thin sections (50 x 75 mm) were
prepared at a commercial laboratory and analyzed at the Department of Archaeology, Boston
University (USA). The thin sections were scanned on a flat-bed scanner and then examined with
binocular and petrographic microscopes at magnifications ranging from 8x to 15x (binocular
microscope), and from 20x to 200x with the petrographic microscope. The following
observations were made on samples from each layer (nomenclature follows that of Courty et al.
[1989] and Stoops [2003]):
Layer I (upper occupation level): The sample consists of compact sandy silty clay with
fragments of charcoal in the lower quarter of the sample. The charcoal tends to occur as
individual, unconnected mm-sized pieces, or as finer, silt-sized remains that are well integrated
into the matrix, suggesting that it has been incorporated into the matrix by small scale
bioturbation (size of earthworms) or weak cryoturbation. There are localized circular to
elongated domains comprising well-sorted, clean quartz sand with no fine interstitial matrix.
Similarly, rounded aggregates of finer silty clay appear to grade into bands of silty clay that at
the mesoscale (i.e., ~10x magnification) are reminiscent of ice-lensing features (see van Vliet-
Lanoe, 1985).
The most interesting aspects of this sample are the presence of charcoal and the sandy domains.
The charcoal tends to occur as isolated distinct mm-sized pieces, as well as silt-sized charcoal,
that are well integrated into the finer silty clay matrix. Both occurrences show that the charcoal
is not intact, but likely has been reworked biologically or by cold phenomena. The latter is
suggested by weak ice lensing, as well as the capping on an elongated fragment of charcoal. The
origin of the sandy domains could also be related to frost affected soils and appear to be fissures
that were later filled with the sand (Van Vliet-Lanoë, 2010). Finally, the generally ‘tight’
porosity in all of the samples (a bit less in this sample, which exhibits some vughy porosity)
resembles that found in fragic horizons. This observation also is consistent with evidence of cold
climate conditions (Figure 9).
Layer II/1 (archaeologically sterile): This sample is massive and richer in the finer silty clay
component than the overlying layer. The sand grains occur with porphyric-related distribution
within the matrix, and as above, exhibit mm- to cm-size domains of clear sand. In addition, mm-
size pores display ferruginous hypocoatings, indicating some gleying/hydromorphism. Along
with a greater abundance of fine material, we can also observe some textural pedofeatures.
These are expressed ~5-μm-thick pale yellow limpid clay coatings around individual sand grains,
as well as slightly thicker void coatings formed within the silty clay matrix. This observation is
significant because it shows that clay illuviation took place after the suggested freeze-thaw
events that produced both the weak banded fabric of the fine fraction and the movement of
quartz sand, which (as described for the sample from Layer I) seems to be genetically associated
with cold soil phenomena. Unfortunately, there is no way to determine the horizon from which
these clay coatings were derived, only that such a zone is above this sample. The overlying
sample did not, however, contain any translocation features, or any effects of gleying. The fact
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
12
that the gleying is found also in the underlying sample (Layer II/2), points to subsurface,
groundwater, gleying.
Figure 9. Soil micromorphology of the sediments containing Upper
Paleolithic occupation Layer I at Mira: band of charcoal with silty clay
capping, an ice-related feature (PPL) (photograph by PG, December
2012).
Layer II/2 (lower occupation level): The proportion of quartz sand to the finer silty clay matrix
is similar to that in the above sample. On the other hand, the segregation of the quartz fraction
from the silty clay matrix is striking in this sample, and is exhibited by a near vertical, roughly
cm-wide tongue of clean quartz sand in the center of the thin section. In addition, secondary iron
staining is somewhat less abundant than in the above sample, although it is not present as
hypocoatings around pores, but as impregnations associated with the remains of charcoal or
organic matter. Moreover, translocated clay is more abundant in this sample than above, and it is
somewhat limited to the finer fraction and less so as coatings around coarser quartz grains. This
increased degree of translocation suggests that this sample is more within what was a weak Bt
soil horizon. The clearly defined textural tongue containing clean quartz sand reflects the
position of this sample possibly close to a former surface and is part of a number of ‘bleached’
areas observed in the field at this position. In any case, it is reasonable to infer that the formation
of the tongues is related to frost-affected soils, whereas the translocation of the limpid clay—as
coatings around quartz grains and void coatings within the matrix—is tied to more temperate
conditions. The presence of ferruginous impregnations is indicative of groundwater effects that
might precede the formation of the tongues, as there is no secondary iron staining in the tongues.
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
13
BIOARCHAEOLOGY
Both plant and animal remains were recovered from the Upper Paleolithic occupation layers at
Mira, and their analysis contributes significantly to interpretation of the site. Although sediment
samples collected during the earlier excavations yielded small numbers of pollen grains and
spores, several layers contain plant macro-fossils, including relatively large fragments of burned
wood, which is unusual for Paleolithic sites. The occupation layers, especially Layer I, produced
a large quantity of well-preserved mammalian remains, including a sample of several hundred
bones and teeth of horse; the latter were another focus of the 2012 research, and their analysis is
described below.
Plant Macro-Fossils
During the earlier excavations (1997–2001), charcoal fragments (identified as Pinus sp. by F.
Damblon) were recovered from Layer II/1 without evidence of human occupation. The macro-
fossils in this level evidently reflect the effects of a wildfire.
Burned wood fragments also were collected from Upper Paleolithic occupation Layer I,
concentrated in two locations of the 5-m2 area excavated during 2012. The fragments included 5
large pieces, the largest of which is approximately 30 cm in diameter, and more than 15 smaller
pieces (each roughly 1–2 cm in diameter and several mm in thickness). The macro-fossils have
not been identified. The two concentrations of wood fragments were found several cm below the
level of bone fragments mapped in these units, and may or may not be associated with human
occupation (Figure 10A).
Figure 10A. Wood fragments recovered from Layer I (photograph by VNS,
August 2012).
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
14
Mammal Remains
Several thousand bones and teeth of mammals were recovered from the occupation layers at
Mira, and a list (presence/absence) of species is presented in Table IV. Most remains were found
in Layer I, while less than 40 fragmented bones were found in Layer II/2. The bones and teeth
were identified by O. P. Zhuravlev, P. V. Puchkov, and L. I. Rekovets.
Among medium and large mammals, horse (Equus latipes) predominates, representing 72% of
NISP (number of identified specimens) in Layer I, and at least some of the bone fragments in
Layer II/2. Most of the remaining bones and teeth (22% of NISP) in Layer I are assigned to
arctic fox (Alopex lagopus). Less common taxa include hare (Lepus europaeus), steppe marmot
(Marmota bobac), mammoth (Mammuthus primigenius), red deer (Cervus elaphus), reindeer
(Rangifer tarandus), giant deer (Megaloceros giganteus), and steppe bison (Bison priscus). In
addition to horse, some of the bone fragments in Layer II/2 were identified as bison.
Analysis of Horse Remains from Layer I
Among large mammals, only the horse remains from Layer I constitute a sufficiently large
sample for analysis of skeletal-part representation and mortality profile. An analysis of the
taphonomy of the horse remains from Layer I was undertaken by one of us (A. Brugère) as part
of a separate study, and some of the results, supplemented by observations of the senior author,
are presented below.
Weathering, Breakage, and Surficial Damage. The bones are moderately weathered. In terms of
color, a random sample (n = 48) was primarily “brownish yellow (10YR6/6), “very pale brown”
(10YR7/4), and “light yellowish brown” (10YR6/4). A combined sample of femur and humerus
shaft bones (n = 55) were classified according to weathering stages defined by Behrensmeyer
(1978) and Johnson (1985): Stage 1 (0%), Stage 1/2 (49%), Stage 2 (31%), Stage 2/3 (15%), and
Stage 3 (5%).
There is evidence of substantial breakage of bones in a fresh or green condition. A combined
sample of femur and humerus fragments (epiphyses and shafts) (n = 81) contained examples of
green (52%), dry (25%), and undetermined (35%) fractures (some bones exhibited more than one
type of break). Green breakage types observed include sawtooth, spiral, V-shaped, cone fracture,
and others.
The bones exhibited little evidence of root etching or carnivore gnawing—only isolated
examples of each were observed. Probable tool cut-marks and percussion marks were noted on a
number of bones (n = 16), including fragments of the ribs, humerus, radius, pelvis, femur, tibia,
and metapodials (Figure 10B).
Skeletal-Part Representation. Most skeletal parts are represented, including a large quantity of
isolated teeth. The distribution of parts (excluding teeth) is shown in Table V, in which raw
counts of identified bone fragments (NISP) have been converted to minimum number of animal
units (MAU), and the latter have been normalized (by dividing each MAU value by the greatest
MAU value in the assemblage) (e.g., Lyman, 1994: 104–110). Total NISP counts for some parts
were higher than those presented in Table V, because some fragments (e.g., radius shaft) could
not be assigned to a specific part in the table. The best represented parts of the skeleton are the
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
15
proximal femur and distal tibia, while the least well represented parts are some of the lower limb
elements such as the astragalus and metatarsal.
Figure 10B. Horse remains from Layer I: probable stone tool cut
marks on a tibia shaft fragment (photograph by JFH, November
2012).
For comparative purposes, the skeletal-part distribution for the same taxon in the broadly
contemporaneous occupation at Kostenki 14, Layer II (Hoffecker et al., 2010: 1078, table 2) also
is shown in Table V. Most of the same parts found at Mira are present at Kostenki 14, but there
are significant differences in the proportional representation of specific elements. Statistical
comparison of the two distributions (appendicular parts only) yields a Kolmogorov-Smirnov
value of 1.73, which is significant at the 0.01 level (e.g., Klein and Cruz-Uribe, 1984: 73–74).
Mira yielded relatively few bones of the lower extremities, and the contrast with Kostenki 14 is
especially pronounced for the metapodials, larger tarsals, and phalanges. Because these elements
are dense (see Lam et al., 1999: 348–353) and easy to identify (even when broken), degree of
weathering or fragmentation is unlikely to account for the difference. As a food utility index
(FUI) for horse (Outram & Rowley-Conwy, 1998: 845, table 6) indicates, however, they possess
low food value in comparison to other elements. The poor representation of these parts at Mira
probably reflects selective retrieval of “meaty” portions of the carcass by the occupants of the
site (Stepanchuk, 2005: 26).
Demographic Data. Variations in the presence/absence and eruption and wear of teeth provide
information of the age and sex of the horses represented in Layer I. These data indicate that both
adults and juveniles were present and that females probably were predominant among the adults.
This age/sex distribution is consistent with that of a mare-band among living horses (e.g., Berger,
1986; Niven, 2007: 373).
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
16
In a separate study, one of us (A. Brugère) grouped the isolated cheek-teeth into age sets.
Although results varied by tooth, they all indicated a group composed of several adults and
young individuals. For example, the left mandibular teeth yielded the following age distribution:
0–1 years (n = 4), 1–3 years (n = 3), 3–5 years (n = 4), 5–10 years (n = 3), >10 years (n = 2).
Based on data from Zhuravlev and Puchkov, Stepanchuk (2005: 28) reported only two canines
among the horse teeth, suggesting most of the adult horses were females (e.g., Turner, 2002:
204).
Articulated Skeletal Elements. At several locations within the areas excavated earlier (2000 and
2008), Stepanchuk (2005: 26) encountered two or more horse bones in anatomically correct
order (i.e., not disarticulated). These include the following: first and second phalanx (unit 25ж);
astragalus, tarsals, and others (unit 26Е), metacarpal and 10 others (unit 25E); and vertebrae
(unit?).
The presence of articulated skeletal elements is significant and indicates that portions of horse
carcasses were transported intact to the occupation area. This, in turn, suggests that the location
of the kill was not remote, but probably within a few thousand meters from the site. Articulated
horse bone sequences (vertebrae and foot bones) were encountered at Kostenki 14, Layer II
(Rogachev, 1957: 78), as well as other contemporaneous EUP sites in Europe, including
Kostenki 15 and the Aurignacian units at Solutré (eastern France) (Rogachev & Sinitsyn, 1982:
163; Olsen, 1989: 305–314). Sequences of articulated bone also are common in sites interpreted
as kill-butchery or large-mammal carcass-processing locations in North America (e.g., Frison,
1974: 64–66; Johnson, 1987: 124; Todd, 1987: 140–150).
Conclusions. The taphonomic characteristics of the horse remains in Layer I probably indicate
the butchering of a group of horses near the site (Stepanchuk, 2005). This conclusion is based on
the following observations: a) large concentration of bones and teeth representing more than a
dozen individual horses of varying age and sex; (b) virtually all skeletal parts represented and
multiple groups of articulated bones, including vertebrae and lower limb elements; (c) traces of
carnivore damage almost entirely absent; and (d) high proportion of bones fractured when fresh
and some percussion and cut marks (cut marks sometimes observed in anatomically significant
locations), apparently reflecting multiple phases of a butchering process. The demographic
profile of the horses probably reflects a mare-band comprising an adult male, several adult
females, and multiple juveniles (Berger, 1986; Olsen, 1995). The pattern is similar to that
observed at two other late EUP sites on the East European Plain: Kostenki 14, Layer II and
Kostenki 15 (Hoffecker et al., 2010: 1078–1081).
Although there is no direct evidence of hunting (e.g., point embedded in horse bone) as opposed
to scavenging carcasses from a natural catastrophe, the former seems most likely, given the
almost complete absence of carnivore damage to the bones, as well as the lack of features in the
area that are associated with catastrophic death (e.g., box canyon subject to flash floods). In fact,
the topographic setting of the site (i.e., center of wide floodplain) raises questions about how its
occupants were able to trap and kill a large group of horses without some form of barrier or
enclosure.
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
17
MIRA AND THE EUP OF EASTERN EUROPE
Mira remains the only dated EUP site in the immense Dnepr Basin, and fills a major void in the
EUP record of the East European Plain. The geomorphic context of the site suggests that the
scarcity of EUP sites in this region—and more generally on the East European Plain—is due not
to sparse settlement, but to deep burial in sediments that are largely inaccessible and rarely
subject to accidental exposure and discovery. There probably are many more sites similar to
Mira buried in the alluvium of the Second Terrace.
The site dates to a late phase of the EUP (35,000–30,000 cal BP) and is broadly
contemporaneous with several other major EUP sites/layers on the East European Plain,
including Molodova V, Layer 10; Kostenki 14, Layer II; Kostenki 15; and Sungir’ (Sinitsyn &
Hoffecker, 2006; Haesaerts et al., 2003: 166–167; Hoffecker et al., 2010: 1078; Marom et al.,
2012). Radiocarbon dates on the two occupation layers at Mira cluster tightly around ~32,000
cal BP and correlate with a cold oscillation (GS 5) in the Greenland ice-core record. The
sediments containing the occupation debris exhibit some evidence of cold climate, and mammal
remains associated with the upper layer include arctic fox.
Mira is found in a paleo-topographic setting unique to the EUP of Eastern Europe. The position
of the site in relation to the Third Terrace indicates that it was located near the center of the
broad Dnepr River floodplain—not on a low terrace above the floodplain or along a side-valley
ravine. The site was occupied during an interval when the floodplain was relatively stable—
subject to periodic overbank deposition—and weak soil formation under saturated conditions
was occurring on the surface. No evidence of spring activity was detected in the analysis of the
soil micromorphology (although a modern spring outlet is found ~1 km west of the site in a
Holocene ravine) and it remains unclear what attracted people to this specific area of the
floodplain on at least two occasions.
During the second occupation (which took place at least a few decades or possibly centuries after
the first occupation), a group of horses—probably a mare-band—was killed near the site by its
inhabitants. Close proximity to the kill location is suggested by the relatively complete
representation of skeletal elements, and the presence of several articulated bone sequences. The
hunting of mare-bands seems to have been widespread during the later EUP, not only on the East
European Plain (i.e., at Kostenki), but also in Western Europe (at Solutré in eastern France)
(Olsen, 1989; Hoffecker et al., 2010). In contrast to the horse bone assemblage at Kostenki 14,
Layer II, Mira contains few lower limb elements. These bones are comparatively low in food
value, and apparently were abandoned at the kill location—probably indicating that the later was
not as close to the site as at Kostenki 14. Also in contrast to other EUP sites that yield evidence
for the hunting of mare-bands, there was no natural trap or cul-de-sac in the vicinity of Mira,
which lends support to the suggestion (raised at other sites [e.g., Olsen, 1995: 73]), that some
form of artificial barrier or enclosure was constructed to trap the horses. The large wood
fragments recovered from Layer I and II/1 indicate that materials were available to construct a
fence or barrier.
Although technically a stratified site, the cultural levels at Mira were occupied in relatively rapid
succession (within a few centuries or less), but the relationship between the two artifact
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
18
assemblages remains unclear. Bladelets recovered from the lower occupation level (Layer II/2)
are diagnostic of the early Gravettian (Stepanchuk, 2005: 27–28), which also is represented at
the roughly contemporaneous occupation at Kostenki 8, Layer II (Anikovich, Anisyutkin &
Vishnyatsky, 2007: 233–236). New radiocarbon dates from Buran-Kaya III in Crimea suggest
that the Gravettian in Eastern Europe may be almost as early as 40,000 cal BP (Prat et al., 2011).
This, in turn, suggests that it is historically connected to the Ahmarian bladelet industry, now
documented in the northern and southern Caucasus (Adler et al., 2006; Golovanova et al., 2010),
and apparently a proxy for modern human movement directly from the Levant to Eastern Europe
via the Caucasus (Hoffecker, 2012).
The spatial and temporal range of the sites suggests that the early Gravettian industry occupies a
major place in the EUP of Eastern Europe. Its visibility probably is reduced, however, by two
factors that reflect the character of the archaeological record of the East European Plain (i.e.,
scarcity of natural shelters): (a) comparatively few known sites due to low archaeological
visibility; and (b) predominance of artifacts (e.g., flake scrapers) related to large-mammal
butchery and carcass-processing that are not diagnostic of specific cultural entities. In other
words, the people who produced the bladelets in Layer II/2 at Mira may have been part of a
widespread cultural phenomenon in Eastern Europe 40,000–30,000 cal BP, but without leaving
an archaeological record comparable to the contemporaneous Aurignacian of Western Europe.
The upper layer at Mira (Layer I) also contains numerous bladelets, but they are not particularly
diagnostic of a specific EUP industry (see Stepanchuk, 2005: 34–34); they might be connected to
several contemporaneous sites/layers in Eastern Europe that contain elements of the Aurignacian
technocomplex (e.g., Kostenki 1, Layer III). The assemblage from Layer I has been widely
assigned to a local East European industry (Gorodtsovan) that exhibits a combination of typical
Middle and Upper Paleolithic technology and tool forms (e.g., Cohen & Stepanchuk, 1999: 298–
301; Anikovich, Anisyutkin & Vishnyatsky, 2007: 262). An alternative explanation is that the
“Middle Paleolithic” forms, which include side-scrapers and bifaces, represent expedient tools
related to the processing of large-mammal carcasses, and it should be noted that other
“Gorodtsovan” sites (e.g., Kostenki 14, Layer II; Kostenki 15) also yield evidence of horse
carcass-processing (Hoffecker, 2011).
ACKNOWLEDGMENTS
The 2012 Mira project was supported by a general grant from the L.S.B. Leakey Foundation,
administered by the Illinois State Museum. The authors are grateful to Chad Wolak and Patrick
Cappa at the AMS Radiocarbon Prep Laboratory at the Institute of Arctic and Alpine Research
(INSTAAR), University of Colorado at Boulder, and also acknowledge the editors of
Geoarchaeology and several anonymous reviewers for their comments on the draft, which
improved the final version of the paper.
REFERENCES
Adler, D.S., Belfer-Cohen, A., & Bar-Yosef, O. (2006). Between a rock and a hard place: Neandertal-
modern human interactions in the southern Caucasus. In N. J. Conard (Ed.) Neandertals and
Modern Humans Meet (pp. 165–187). Tübingen: Krems-Verlag.
Anikovich, M.V., Anisyutkin, N.K., & Vishnyatsky, L.B. (2007). Uzlovye problemy perekhoda k
verkhnemu paleolitu v Evrazii.Saint Petersburg: Nestor-Istoriya.
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
19
Bader, O. N. & Bader, N. O. (2000). Verkhnepaleoliticheskoe poselenie Sungir’. In T. I. Alekseeva & N.
O. Bader (Eds.) Homo sungirensis verkhnepaleoliticheskii chelovek: ekologicheskie i
evolyutsionnye aspekty issledovaniya (pp. 21 – 27). Moscow: Nauchnyi Mir.
Behrensmeyer, A.K. (1978). Taphonomic and ecologic information from bone weathering. Paleobiology,
4, 150–162.
Berger, J. (1986). Wild horses of the Great Basin. Chicago: University of Chicago Press.
Cohen, V.Yu. & Stepanchuk, V.N. (1999). Late Middle and early Upper Paleolithic evidence from the
East European Plain and Caucasus: a new look at variability, interactions, and transitions. Journal
of World Prehistory, 13(3), 265–319.
Courty, M.- A., Goldberg, P. & Macphail, R. I. (1989). Soils and Micromorphology in Archaeology. New
York: Cambridge University Press.
Frison, G.C. (1974). The Casper Site: A Hell Gap bison kill on the High Plains. New York: Academic
Press.
Galbraith, R.F., Roberts, R.G., Laslett, G.M., Yoshida, H., & Olley, J.M. (1999). Optical dating of single
and multiple grains of quartz from Jinmium rock shelter, northern Australia, part 1, Experimental
design and statistical models. Archaeometry, 41, 339–364.
Goldberg, P., & Macphail, R.I. (2006). Practical and theoretical geoarchaeology. Boston: Blackwell.
Golovanova, L.V., Doronichev, V.B., Cleghorn, N.E., Kulkova, M.A., Sapelko, T.V. & Shackley, M.S.
(2010). Significance of ecological factors in the Middle to Upper Paleolithic transition. Current
Anthropology, 51, 655–691.
Goretskii, G.I., 1970. Allyuvialnaya letopis’ Velikogo Pra-Dnepra. Moscow: Nauka.
Haesaerts, P., Borziak, I., Chirica, V., Damblon, F., Koulakovska, L. & van der Plicht, J. (2003). The east
Carpathian loess record: a reference for the middle and late Pleniglacial stratigraphy in central
Europe. Quaternaire, 14,163–188.
Hoffecker, J. F. (2011). The early Upper Paleolithic of eastern Europe reconsidered. Evolutionary
Anthropology, 20(1), 24–39.
Hoffecker, J. F. (2012). A new framework for the Upper Paleolithic of Eastern Europe. Proceedings of the
European Society for the Study of Human Evolution, 1, 101.
Hoffecker, J.F., Holliday, V.T., Anikovich, M.V., Sinitsyn, A.A., Popov, V.V., Levkovskaya, G.M.,
Pospelova, G.A., Forman, S.L., Lisitsyn, S.N. & Giaccio, B. (2008). From the Bay of Naples to
the River Don: the Campanian Ignimbrite eruption and the Middle to Upper Paleolithic transition
in Eastern Europe. Journal of Human Evolution, 55, 858–870.
Hoffecker, J.F., Kuz’mina, I.E., Syromyatnikova, E.V., Anikovich, M.V., Sinitsyn, A.A., Popov, V.V. &
Holliday, V.T. (2010). Evidence for kill-butchery events of early Upper Paleolithic age at
Kostenki, Russia. Journal of Archaeological Science, 37, 1073–1089.
Holliday, V.T. (2004). Soils in archaeological research. New York: Oxford University Press.
Holliday, V.T., Hoffecker, J.F., Goldberg, P., Macphail, R.I., Forman, S.L., Anikovich, M. & Sinitsyn, A.
(2007). Geoarchaeology of the Kostenki-Borshchevo sites, Don River, Russia. Geoarchaeology:
An International Journal, 22(2), 183–230.
Holliday, V.T. & Mandel, R.E. (2006). Geoarcheology of the Plains, Southwest, and Great Lakes. In D.
H. Ubelaker (Ed.), Handbook of North American Indians: Environment, origins, population, vol.
3. (pp. 23–46). Washington: Smithsonian Institution Press.
Johnson, E. (1985). Current developments in bone technology. In M. B. Schiffer (Ed.), Advances in
Archaeological Method and Theory, vol. 8. (pp. 157-235). New York: Academic Press.
Johnson, E. (1987). Cultural activities and interactions. In E. Johnson (Ed.), Lubbock Lake: Late
Quaternary studies on the Southern High Plains (pp. 120–158). College Station: Texas A&M
University Press.
Klein, R.G. & Cruz-Uribe, K. (1984). The analysis of animal bones from archeological sites. Chicago:
University of Chicago Press.
Kozlowski, J.K. (2007). The significance of blade technologies in the period 50-35 kya BP for the
Middle-Upper Palaeolithic transition in central and Eastern Europe. In P. Mellars, K. Boyle, O.
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
20
Bar-Yosef & C. Stringer (Eds.), Rethinking the human revolution: new behavioural and
biological perspectives on the origin and dispersal of modern humans (pp. 317–328). Cambridge:
McDonald Institute Monographs, Cambridge.
Lam, Y.M., Chen, X. & Pearson, O.M. (1999). Intertaxonomic variability in patterns of bone density and
the differential representation of bovid, cervid, and equid elements in the archaeological record.
American Antiquity, 64, 343–362.
Lyman, R.L. (1994). Vertebrate taphonomy. Cambridge: Cambridge University Press.
Marom, A., McCullagh, Higham, T.F.G., Sinitsyn, A.A. & Hedges, R.E.M. (2012). Single amino acid
radiocarbon dating of Upper Paleolithic modern humans. Proceedings of the National Academy
of Sciences doi/10.1037/pnas.1116328109
Matoshko, A.V., Gozhik, P.F. & Ivchenko, A.S. (2002). The fluvial archive of the Middle and Lower
Dnieper (a review). Netherlands Journal of Geosciences, 81, 339–355.
Montet-White, A. (2002). Les outillages des chasseurs de Solutré. In: Combier, J., Montet-White, A.
(Eds.) Solutré 1968-1998, Société Préhistorique Française Mémoire, 30, 225–241.
Murray, A. S. & Wintle, A. G. (2003). The single aliquot regenerative dose protocol: potential for
improvements in reliability. Radiation Measurements, 37, 377–381.
Niven, L. (2007). From carcass to cave: Large mammal exploitation during the Aurignacian at Vogelherd,
Germany. Journal of Human Evolution, 53, 362–382.
Noiret, P. (2005). The Aurignacian in Eastern Europe. Anatolia, 29, 39–56.
Olsen, S.L. (1989). Solutré: a theoretical approach to the reconstruction of Upper Palaeolithic hunting
strategies. Journal of Human Evolution, 18, 295–327.
Olsen, S.L. (1995). Pleistocene horse-hunting at Solutré: why bison jump analogies fail. In: Johnson, E.
(Ed.) Ancient Peoples and Landscapes (pp. 65–75). Lubbock: Museum of Texas Tech University.
Outram, A.K. & Rowley-Conwy, P. (1998). Meat and marrow utility indices for horse (Equus). Journal of
Archaeological Science, 25, 839–849.
Prat, S., Péan, S.C., Crépin, L., Drucker, D.G., Puaud, S.J., Valladas, H., Lázničková-Galetová, M., van
der Plicht, J. & Yanevich, A. (2011). The oldest anatomically modern humans from far southeast
Europe: direct dating, culture, and behavior. PLoS ONE, 6, e20834.
Prescott, J.R. & Hutton, J.T. (1994). Cosmic ray contributions to dose rates for luminescence and ESR
dating: large depths and long-term time variations. Radiation Measurements, 23, 497–500.
Richter, D., Tostevin, G. & Škrdla, P. (2008). Bohunician technology and thermoluminescence dating of
the type locality of Brno-Bohunice (Czech Republic). Journal of Human Evolution, 55, 871–885.
Rogachev, A.N. (1957). Mnogosloinye stoyanki Kostenkovsko-Borshevskogo raiona na Donu i problema
razvitiya kul’tury v epokhu verkhnego paleolita na Russkoi Ravnine. Materialy i issledovaniya po
arkheologii SSSR, 59, 9–134.
Rogachev, A.N. & Anikovich, M.V. (1984). Pozdnii paleolit Russkoi Ravniny i Krima. In P. I.
Boriskovskii (Ed.), Paleolit SSSR (pp. 162–271). Moscow: Nauka.
Rogachev, A.N. & Sinitsyn, A.A. (1982). Kostenki 15 (Gorodtsovskaya stoyanka). In N. D. Praslov & A.
N. Rogachev (Eds.), Paleolit Kostenkovsko-Borshchevskogo raiona na Donu 1879–1979 (pp.
162–171). Leningrad: Nauka.
Sinitsyn, A.A. & Hoffecker, J.F. (2006). Radiocarbon dating and chronology of the early Upper
Paleolithic at Kostenki (Russia). Quaternary International, 152-153, 175–185.
Stepanchuk, V.N. (2005). The archaic to true Upper Paleolithic interface: the case of Mira in the Middle
Dnieper area. Eurasian Prehistory, 3(1), 23–41.
Stepanchuk, V.N., Cohen, V. Yu. & Pisaryev, I. B. (1998). Mira, a new Late Pleistocene site in the
Middle Dnieper, Ukraine (preliminary report). Pyrenae, 29, 195–204.
Stepanchuk, V.N., Cohen, V.Yu., Gerasimenko, N.P., Damblon, F., Haesaerts, P., Zhuravlev, O.P.,
Kovalyukh, N.N., Petrougne, V.F., van der Plicht, J., Putshkov, P.V., Rekovets, L.I. & Turner,
C.G. (2004). Bogatosharova stoyanka Mira na Sredn’omu Dnipri: osnoni rezul’tati razkopok
2000 roku (pp. 62–98). Kiev: Kam’yana doba Ukraini.
Geoarchaeological and bioarchaeological studies at Mira, early UP site in Lower Dnepr Valley, Ukraine
21
Stepanchuk, V.N., Sapozhnikov, I.V., Gladkikh, M.I. & Ryzhov, S.N. (2009). Ukrainian Upper
Paleolithic between 40/10.000 BP: Current insights into environmental-climatic change and
cultural development. In F. Djindjian, J. Kozlowski, & N. Bicho (Eds.), Le Concept de Territories
dans le Paléolithique Supérieur Européen (pp. 63–74). BAR International Series 1938. Oxford:
British Archaeological Reports.
Stoops, G. (2003). Guidelines for analysis and description of soil and regolith thin sections. Madison: Soil
Science Society of America.
Todd, L.C. (1987). Taphonomy of the Horner II bone bed. In G. C. Frison & L. C. Todd (Eds.), The
Horner site: The type site of the Cody Cultural Complex (pp. 107–198). Orlando: Academic
Press.
Turner, E. (2002). Analyse archéozoologique des restes de faune magdaléniens de Solutré (Secteurs I11 et
P16). In J. Combier & A. Montet-White (Eds.), Solutré 1968-1998, Société Préhistorique
Française Mémoire, 30, 197–217.
van Vliet-Lanoë, B. (1985). Frost effects in soils. In J. Boardman (Ed.), Soils and Quaternary landscape
evolution (pp. 117–158). Chichester: John Wiley & Sons.
van Vliet-Lanoë, B. (2010). Frost action. In G. Stoops, V. Marcelino & F. Mees (Eds.), Interpretation of
micromorphological features of soils and regoliths (pp. 81–108). Amsterdam: Elsevier.
Weninger, B. & Jöris, O. (2008). A 14C calibration curve for the last 60 ka: the Greenland-Hulu U/Th
timescale and its impact on understanding the Middle to Upper Paleolithic transition in western
Eurasia. Journal of Human Evolution, 55, 772–781.
1
Table I. Stratigraphic profile for Mira, recorded in August 2012 (V. T. Holliday)
Depth, cm
Bench
Soil
Horizon
PH
strat
Description/comments
0-15
1
A
1
Fine sandy loam; 10YR 4/3 slightly moist; very weak subangular
blocky; clear
15-50
1
C
2
Fine sandy loam; 10YR 4/4 slightly moist, very weak subangular
blocky; very irregular, mixed boundary (due to bioturbation)
50-68
1
Ab1
3
Fine sandy loam; 10YR 3/2 slightly moist; very weak subangular
blocky; krotovinas common; very irregular, mixed boundary
68~120
1 -
91cm
2
Cb1
4
Fine sandy loam; 10YR 4/4 & 3/2 slightly moist; very weak
subangular blocky; mixed; common krotovinas below ~95 cm
~120~148
2
A1b2
5
Fine sandy loam; 10YR 3/1 slightly moist; very weak subangular
blocky; mixed; common krotovinas; very irregular boundary
~148-166
2
A2b2
5
Fine sandy loam; 10YR 4/3 slightly moist; very weakly subangular
blocky; some krotovinas; single 1-2mm discontinuous clay band;
irregular boundary
166-240
2 -
185cm
3
Cb2
6,7
Fine sandy loam; 10YR 4/3 slightly moist, 4/4 dry; massive; wavy
1-2mm continuous clay band ~200cm; wavy, distinct 2-3mm clay
band ~220cm; abrupt, wavy boundary.
Note: PH has lower stratigraphic break at ~200cm and very wavy;
perhaps a clay band?
240-325
3 -
290cm
4
Pedo
complex
b3
8
Four distinct buried soils: A/Bt (240-247); Bt (261-266, 277-
281, 285-295); Bw (310-325); A = 10YR 5/2.5 dry, sandy clay
loam; Bt, Bw = 5/4 dry, sandy clay loam; C = 10YR 6/3 dry, fine
sandy loam; Bt = fine prismatic & fine subangular blocky; thin
clay films on ped faces; krotovinas common; all boundaries =
clear, wavy. OSL12-4 (UIC3344) 268cmbs
325-350
4
11?
Laminated silt; 10YR 7/3, 6/3, 6/4 slightly moist; abrupt
2
boundary
350~385
4 -
352cm
5
13
Massive fine sand; 10YR 6/4 slightly moist; few krotovinas below
350 cm
~385-430
5
13?
As above, but large, common krotovinas
430-475
5
Pedo
Complex
b4
14,
15?
Zones of calcareous & non-calcareous fine sandy loam (dipping to
the east): Bk (430-437 cm, 445-452 cm, 465-475 cm) = 10YR 7/2
dry; weak subangular blocky; common, fine threads & bodies
carbonate; non-calcareous fine sandy loam, 10YR 7/3 dry; clear
boundary OSL12-3 (UIC3342) 472cmbs
475-535
5
15?
Non-calcareous fine sandy loam, 10YR 7/3 dry; thin, weak
carbonate bodies at 503, 509 cm; 2-3 distinct zones of carbonate
on west wall of bench; clear boundary
535-575
5 -
553cm
6
16?
Massive, non-calcareous fine sand; 10YR 5.5/3 slightly moist;
clear, smooth boundary
575-598
6
Pedo
Complex
b5
16
Multiple (~12) weak A-C soils; A = 10YR 4.5/3 slightly moist; C
= 10YR 5.5/3 slightly moist; abrupt, irregular erosional contact
598-605
6 -
606cm
?
Massive fine sand; 10YR 7/2 slightly moist; irregular, mixed
boundary
605-622
7
Ab6
17
Fine sand, 10YR 6.5/1 slightly moist; very weak subangular
blocky; irregular, mixed boundary; this soil and overlying white
sand cut out to the east; probably same unconformity noted at
598 cm in main section
622-651
7
Massive fine sand; 10YR 7/2 slightly moist; clear, wavy boundary
651-706
7
Upper Fe-ox zone in olive fine sandy loam matrix; Fe-ox
concretions are vertical, dense, up to 2 cm wide, and up to 5 cm
long; each more or less tapering with depth; clear boundary
706-720
7
Massive fine sand, 10YR 7/2 slightly moist; clear, wavy boundary
3
~720~770
7 -
731cm
8
Lower Fe-ox zone (same color and morphology as upper Fe-ox zone)
in olive gray fine sandy loam matrix
~770~785
8
Massive fine sand, 10YR 7/2 slightly moist; clear wavy boundary
~785-881
8 -
791cm
9
Laminated sand, 10YR 8/2 & 7/2 dry; clear, horizontal boundary
881-891
9-
891cm
Laminated sand, 2.5Y 7/4 dry; upper boundary cross cuts Fe-ox;
abrupt, horizontal boundary
891-896
10
Two laminae of “green clay” loamy fine sand (fine sand with
silt), 2.5Y 6/2d 4/2 slightly moist; common Fe-ox stains
(following roots?); massive; separated by clean sand, 10YR 8/2 &
7/2 dry; bands are horizontal but somewhat wavy with abrupt,
irregular (bioturbated?) boundaries
896-925
10
Bedded, somewhat wavy fine sand 10YR 8/2 & 7/2 dry, with rare
vertical Fe-ox stains; abrupt, irregular (bioturbated?)
boundary.
925-941
10
Bedded fine sand 10YR 8/2 & 7/2d with six lenses of loamy fine
sand “green clay” 2.5Y 6/2d 4/2 slightly moist, each 3-5 cm
thick; bands are horizontal, but somewhat wavy with abrupt,
irregular (bioturbated?) boundaries, locally bifurcating and
rejoining
945-956
10
Laminated loamy fine sand “green clay” 2.5Y 6/2 dry 4/2 slightly
moist with a few sand lenses; bands are horizontal but somewhat
wavy with abrupt, irregular (bioturbated?) boundaries
956~958
10-
958cm
Fine sand, 10YR 7.5/2 dry; discontinuous; thins irregularly to
the northernmost excavation block; locally missing; abrupt,
irregular boundaries (bioturbation?)
~958~973
Agb7
Uppermost “gray/green clay” (fine sandy clay); varies 15-18 cm
thick; massive; common irregular, tabular bodies of fine sand
~7.5YR 7/2 dry, with very irregular (mixed?) boundaries
throughout upper half; upper 6-8 cm light gray 2.5Y 6/2 dry;
4
upper few cm locally darker (mixing from gray clay above? weak A
horizon?); below is a distinct Fe-ox zone ~8 cm thick, but
locally as thin as 3 cm, 7.5YR 4/6, 5/6 dry, most pervasive and
distinctive in upper 1-2 cm, most of the rest of the Fe-ox zone
is not as pervasive with gray showing through and mostly 10YR
6/8, 5/8 dry, fading with depth; in plan view, in the south
excavation block exposing the top of the Fe-ox zone, the
distinct 7.5YR zones are localized as isolated bodies or lining
pockets of white sand; most of the Fe-ox is 10YR; in the
northernmost excavation block, the Fe-ox zone is more pervasive,
2.5Y 6/8 moist, with bodies 7.5YR 5/6 4/6 moist; the lower part
of this “gray clay” is lighter gray 10YR 6/1 dry, 5/1 moist;
very irregular, abrupt lower boundary where sand is present
below; otherwise clear, smooth, generally horizontal boundary
with lower clay
Note: isolated flakes in this unit
~973~975
Fine sand, 10YR 7.5/2; up to 8 cm thick (in south excavation
area), thins to north and appears as localized lenses; very
irregular, abrupt upper and lower boundaries (bioturbation?)
~975~1005
Agb8
Lower “gray/green clay” (fine sandy clay); mostly 10YR 5/2, 5/3
slightly moist; massive; upper 10 cm has discontinuous, roughly
horizontal Fe-ox zones; common, discontinuous pockets of sand
below occupation Layer II/1, along with some of the Fe-ox;
clear, roughly horizontal, but irregular boundary
Note: Upper Paleolithic occupation Layer I and Layer II/2 in top
and bottom of this lower “gray clay zone,” respectively.
TS 12-1 & OSL 12-1 (UIC3338) at 980cmbs in CLI
TS 12-1 at 990cmbs
TS 12-3 & OSL 12-2 (UIC3339) at 1000cmbs in CLII/2
~1005~1023
Pale tan medium sand, 10YR 7/3 slightly moist; massive; gradual
boundary ~1020~1023 cm into underlying sand
~1023~1025
White fine Sand, 10YR 7/2 slightly moist; massive; mixed
boundary
5
~1025~1030
Bioturbated olive sand, 2.5Y 6/3 slightly moist; mixed boundary
~1030~1048
White fine sand, 10YR 7/2 slightly moist with faint olive zone
2.5Y 6/2 slightly moist, ~1038~1045 cm
~1048~1065
White fine sand with distinct Fe-ox zones; horizontally bedded
~1065~1075
Fine sand with pervasive Fe-ox coloration; pale tan sand lenses
become more common with depth; horizontally bedded
~1075~1086
Medium sand with some Fe-ox following bedding and common
irregular Fe-ox bodies; horizontal bedding; gradational boundary
~1085~1110
White medium sand, massive
~1110~1155
White medium sand; massive with few Fe-ox bodies and rare Mn-ox
bodies
Notes:
Depth following a Bench number is the depth of that bench below surface.
Bench 10 is top of narrow excavation block running N-S along W wall of North Block excavations.
PH strat = stratigraphic units defined by Paul Haesaerts and N. Gerasimenko (see Stepanchuk, 2005, fig. 2)
Fe-ox = iron oxides; Mn-ox = manganese oxides.
6
Table II. Radiocarbon dates from Mira: 2012 (INSTAAR Radiocarbon Laboratory)
_________________________________________________________________________________________________________
Lab No. Layer Material graphite ∂13 C wrt PDB fraction 14C age calibrated age1
used modern
_________________________________________________________________________________________________________
CURL-15810 Layer I charcoal 0.56 mg -23.2‰ 0.0379±.001 26,290 ± 220 yrs BP 31,127 ± 367 calBP
CURL-15800 Layer I charcoal 0.63 mg -24.7‰ 0.0324±.001 27,540 ± 260 yrs BP 32,167 ± 280 calBP
CURL-15808 Layer II/1 charcoal 0.37 mg -23.6‰ 0.1791±.0011 13,815 ± 50 yrs BP 17,011 ± 155 calBP
CURL-15795 Layer II/2 charcoal 0.65 mg -24.1‰ 0.033±.001 27,400 ± 260 yrs BP 32,041 ± 231 calBP
CURL-15789 Layer II/2 charcoal 0.28 mg -19.1‰ 0.0567±.0012 23,050 ± 180 yrs BP 27,606 ± 453 calBP
__________________________
1CalPal online quickcal2007 ver 1.5
7
Table III. Optically stimulated luminescence ages on quartz grains (150–250 micron) for sediments from Mira (S. L. Forman)
_______________________________________________________________________________________________________________
Lab No. DEPTH Aliquotsa Over- equivalent U Th K H2O Cosmic Dose rate OSL age
Dispersionb dose (Gray)c (ppm)d (ppm)d (%)d (%) dose (yr)f
(%) (mGray/yr)e (mGray/yr)
_____________________________________________________________________________________________________________________________________
UIC3344 268 cm 28/30 16±2 18.60±0.83 1.2±0.1 4.6±0.1 1.11±0.01 15±5 0.19±0.02 1.59±0.08 11,710±1000
UIC3342 472 cm 28/30 25±4 14.10±0.82 0.5±0.1 1.8±0.1 0.44±0.01 15±5 0.15±0.02 0.70±0.04 20,120±1945
UIC3338 980 cm 29/30 19±3 19.16±0.97 0.8±0.1 3.2±0.1 0.71±0.01 15±5 0.08±0.01 0.88±0.04 21,990±1925
UIC3339 1000 cm 30/30 17±2 23.31±1.05 0.7±0.1 2.6±0.1 0.56±0.01 30±5 0.10±0.01 0.84±0.04 27,665±2430
________________________________________________
aAliquots used in equivalent dose calculations versus original aliquots measured.
bValue reflects precision beyond instrumental errors; value of ≤ 20% (at 2 sigma limits) indicate low dispersion in equivalent dose values and an unimodal
distribution.
cEquivalent dose calculated on a pure quartz fraction with about 200-500 grains/aliquot (~2 mm plate area) and analyzed under blue-light excitation.
(470 ± 20 nm) by single aliquot regeneration protocols (Murray & Wintle, 2003). Equivalent dose calculated using the Central Age Model (Galbraith et al, 1999).
dU, Th, and K content analyzed by inductively-coupled plasma-mass spectrometry analyzed by Activation Laboratory LTD, Ontario, Canada.
eCosmic dose rate calculated from parameters in Prescott and Hutton (1994).
fSystematic and random errors are included and reported errors are at one sigma; reference year for ages is AD 2000.
8
Table IV. Vertebrate Remains from Mira (O. P. Zhuravlev, P. V. Puchkov, and L. I. Rekovets)
_________________________________________________________________________
Species Layer I Layer II/2 Layer II/2
_________________________________________________________________________
Lepus cf. europaeus + - -
Ochotona cf. pusilla + - -
Marmota bobac + - -
Myospalax sp. + - -
Lagurus lagurus + - -
Eolagurus luteus + - -
Clethrionomys sp. + - -
Microtus gregalis - - +
Microtus cf. socialis + - -
Microtus oeconomus + - -
Microtus arvalis-socialis + - -
Alopex lagopus + - -
Vulpes vulpes + - -
Vulpes corsac + - -
Meles meles + - -
Mammuthus primigenius + - -
Equus latipes + ? +
Cervus elaphus + - -
Megaloceros giganteus + - -
Rangifer tarandus + - -
Bison priscus + - +
9
Table V. Comparative Representation of Skeletal Parts for Equus latipes (broad-toed horse) at Mira,
Layer I and Kostenki 14, Layer II (Hoffecker et al., 2010: 1078, table 2).
___________________________________________________________________________________________
Mira-Layer I Kostenki 14-Layer II
Skeletal Part NISP MAU %MAU NISP MAU %MAU FUI
cranium 14 ? ? ? ? 8.0
maxilla 12 ? ? ---
mandible 44 2.0 36 44 6.5 46.4 3.3
hyoid 1 0.5 9 1 1.0 0.7 ---
atlas 2 2.0 36 ? ---
axis 0 0.0 0 ? ---
vertebrae 7 ? ? ? ? ---
ribs 37 ? ? ~250 ? ---
lumbar vertebrae 0 0.0 0 ? ? ---
caudal vertebrae 2 1.0 18 ? ? ---
scapula 7 2.5 45.5 51 11.5 82.1 6.7
humerus-proximal 5 2.5 45.5 12 3.0 21.4 6.7
humerus-distal 2 1.0 18 27 11.0 78.6 6.3
radius-proximal 8 4.0 72.7 22 8.5 60.7 3.9
radius-distal 2 1.0 18 21 10.0 71.4 2.7
ulna 4 2.0 36 16 7.5 53.6 3.9
carpals 14 1.5 27 111 -- -- 1.4
metacarpal-proximal 2 1.0 18 18 8.5 60.7 0.7
metacarpal-distal 4 2.0 36 17 8.5 60.7 0.3
innominate 8 1.0 18 14 2.5 17.9 23.7
femur-proximal 11 5.5 100 12 4.5 32.1 20.3
femur-distal 6 3.0 54.5 11 5.0 35.7 20.3
patella 3 1.5 27 21 10.5 75.0 ---
tibia-proximal 3 1.5 27 5 2.5 17.9 11.3
tibia-distal 11 5.5 100 28 14.0 100.0 6.8
calcaneus 1 0.5 9 20 10.0 71.4 3.4
astragalus 1 0.5 9 20 10.0 71.4 3.4
tarsals 9 0.5 9 87 -- -- 3.4
metatarsal-proximal 1 0.5 9 26 10.0 71.4 1.7
metatarsal-distal 2 1.0 18 17 8.5 60.7 0.8
phalanx 1 6 1.25 22.7 39 9.8 70.0 0.4
phalanx 2 7 1.5 27 44 11.0 78.6 0.4
phalanx 3 4 1.0 18 37 9.3 66.4 0.4
sesamoid 1 -- -- ? ---
NISP = Number of Identified Specimens; MAU = Minimum Animal Units
%MAU = percentage of largest MAU value in assemblage; FUI = Food Utility Index (see Outram & Rowley-Conwy, 1998: 845, table 6)
