Plant-food preparation area on an Upper Paleolithic
brush hut floor at Ohalo II, Israel
Ehud Weiss a,b,*, Mordechai E. Kislev c
, Orit Simchoni c
, Dani Nadel d
, Hartmut Tschauner e
a
Martin (Szusz) Department of Land of Israel Studies and Archaeology, The Institute of Archaeology, Bar-Ilan University, Ramat Gan 52900, Israel
b
Kimmel Center for Archaeological Sciences, Weizmann Institute of Science, Rehovot 76100, Israel
c
Faculty of Life Sciences, Bar-Ilan University, Ramat Gan 52900, Israel
d
Zinman Institute of Archaeology, University of Haifa, Haifa 31905, Israel
e
Department of Archaeology, Seoul National University, Gwanak-gu, Sillimdong 56-1, Seoul 151-742, Republic of Korea
a r t i c l e i n f o
Article history:
Received 19 November 2007
Received in revised form 16 March 2008
Accepted 17 March 2008
Keywords:
Archaeobotany
Spatial analysis
Food preparation
Ohalo II
Upper palaeolithic
Division of labor
a b s t r a c t
While a division of domestic space into separate sectors dedicated to different activities has been
suggested for a number of Upper Paleolithic hunter-gatherer sites, it has never been demonstrated based
on plant remains from this period. Moreover, due to the usual scarcity of plant macrofossils in archaeological
deposits, only animal food preparation activities associated with hearths have been reported in
the literature on Near Eastern prehistory. Ohalo II (Israel) is the first Upper Paleolithic site where such
a patterned use of interior space and plant processing are evidenced by the distribution of plant remains
on a sealed floor of a brush hut. This paper describes and interprets the distribution of almost 60,000
identified seeds and other plant remains on that floor, proposing a reconstruction of three activity areas
in the interior of the 12-m2
hut: processing of food centered on a grinding stone; a flint knapping area;
and an access area in between. Finally, it is suggested that these activity areas might represent malefemale
division of labor.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
Detailed analyses of intrasite spatial patterns are the key to
reconstructing activity areas, the finest scale of past spatial behavior
revealed by archaeological assemblages. The reconstruction
of activity areas and their locations, in turn, can help us understand
the internal social and economic organization of ancient households
(Balme and Beck, 2002; Hayden, 1997). Intrasite studies are
predicated upon the assumption that an association of archaeological
remains in a limited space is indicative of a particular human
activity. Most such clusters identified at prehistoric sites are related
either to hearths or to stone-tool manufacturing (see references in
Balme and Beck, 2002; Gowlett, 1997). Due to the relative scarcity
of plant macrofossils in archaeological deposits, to this date only
animal-food preparation activities associated with hearths have
been reported in the literature on Near East prehistory. The present
paper interprets patterned distributions of plant remains as
evidence of a specialized plant processing area on a brush hut floor,
at the Upper Paleolithic site of Ohalo II in Israel.
In recent years, archaeologists have studied patterns of intrasite
spatial organization at Middle Paleolithic sites as proxies of evolving
human cognitive capabilities (see references in Vaquero and
Pasto, 2001). However, ideas and models of ‘‘high resolution’’ or
‘‘high definition’’ archaeology go back much longer (see Gowlett,
1997 for review). Eventually, the rise of the New Archaeology
during the 1960s, with its quest for objectivity in analyzing
archaeological data, led to the development of quantitative, formal
approaches to spatial archaeology (Clark and Stafford, 1982). This
approach is exemplified by the works of Clarke (1977) and Binford
(1981, 1983). The meticulous recording and analysis of archaeological
finds proposed by these scholars was expected to reveal
patterned spatial relationships between the finds that would
provide insights into their sociological, economic, or ecological
meaning.
Understanding the formation processes that affected the
archaeological deposits both during and after a site’s occupation is
crucial for high-resolution intrasite spatial analysis. Meaningful
interpretations of the spatial relationships between objects depend
on there having been no more than minimal disturbance after
initial deposition. However, many studies have documented
sources of site disturbance that may alter the original spatial distribution
of archaeological materials. They include both factors that
act at the time of initial deposition or shortly thereafter (cf. Binford,
* Corresponding author. Kimmel Center for Archaeological Sciences, Weizmann
Institute of Science, Rehovot 76100, Israel.
E-mail address: eweiss@mail.biu.ac.il (E. Weiss).
Contents lists available at ScienceDirect
Journal of Archaeological Science
journal homepage: http://www.elsevier.com/locate/jas
0305-4403/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jas.2008.03.012
Journal of Archaeological Science 35 (2008) 2400–2414
1978) and post-depositional processes, such as human trampling,
displacements caused by wind and water, and animal activity
(Balme and Beck, 20021). Many of the sites where these sources of
disturbance have been examined in detail are cave sites. Goldberg’s
studies (Goldberg, 2000; Goldberg and Bar-Yosef, 1998) of geomorphological
and geochemical profiles of cave sites demonstrate
the significance of geogenic, biogenic, and anthropogenic factors in
causing horizontal and vertical movements of small archaeological
remains.
In contrast to cave sites, Ohalo II is uniquely suited for finegrained,
intrasite spatial analysis. While there are studies that
analyze the distribution of charred plant macrofossils to interpret
human use of space (good examples are Hayden, 1997, 2000), we
are not aware of any other Levantine or, for that matter, Upper
Paleolithic site that has produced a similarly sealed, ‘‘Pompeii-like’’
floor, rich enough in plant remains to allow detailed spatial
analysis. Its short-term, open-air occupation further distinguishes
Ohalo II from contemporaneous cave sites. This paper examines
floor II in hut 1 at Ohalo II. The plant assemblage from all excavation
units of this 12-m2
floor includes close to 60,000 analyzed
specimens. This assemblage puts us in a position to reconstruct the
use of space inside an Upper Paleolithic hut.
Due to the uniquely favorable conditions of preservation at
Ohalo II, as well as the complete excavation and fine-grained
recording of the hut 1 floor, the plant assemblage analyzed here is
so large and its distribution so clear that statistical manipulation of
the data can be kept to a minimum. Most concentrations stand out
immediately on the distribution maps, without any massaging of
the data. The only fundamental quantitative assumption made here
is one that underlies any archaeobotanical study (without always
being made explicit)dnamely that different taxa have roughly
equal preservation rates.
2. Ohalo II
2.1. The site and its setting
Ohalo II is a submerged late Upper Paleolithic (locally termed
Early Epipalaeolithic) site radiocarbon dated to 22,500–23,500 cal
B.P. (Nadel et al., 1995, 2002). The site is located on the southwestern
shore of the Sea of Galilee (Lake Kinneret), Rift Valley,
Israel (Fig.1). This hunter-gatherer-fisher camp, covering more than
2000 m2
(0.2 ha), includes the remains of six brush huts, open-air
hearths, and a human grave (Fig. 1) (Nadel, 2002a; Nadel et al.,
2002). The site was occupied during the Last Glacial Maximum
(LGM), a period of cold and dry climate when ice sheets covered
parts of North America and Europe.
Three successive seasons (1989–1991) of excavations at Ohalo
II were conducted immediately following its discovery, when the
lake level had dropped and much of the site was exposed.
Subsequently, the site was submerged again for several years,
rendering fieldwork difficult. However, additional three seasons of
fieldwork took place in 1999–2001, after droughts and heavy
water pumping from the lake had caused the water level to drop
to ca. 213.80 m below msl.
The archeological remains are located on the clay/silt Lisan
Formation bedrock. The Lisan Formation into which the Ohalo II hut
floors were dug is composed of lacustrine sediments, dated
between 70,000 and 15,000 years B.P. (Bartov et al., 2002). The huts
were located on the shores of the newly formed Sea of Galilee
(Belitzky and Nadel, 2002a).
Ohalo II is an exceptional site on a number of accounts. No other
Levantine Upper Paleolithic site has produced such well preserved
huts, hearths, a grave, and large quantities of archaeological materials.
These include flint and ground stone tools, a broad spectrum of
animal remains (mammals, birds, rodents, fish, mollusks), and a rich
plant assemblage (Bar-Yosef Mayer, 2002; Belmaker et al., 2001;
Kislev et al., 1992; Nadel and Werker, 1999; Nadel et al., 1994, 2002;
Rabinovich, 2002; Rabinovich and Nadel, 1994, 2002; Simmons,
2002; Simmons and Nadel, 1998; Weiss, 2002).
The probable reason for the outstanding preservation is the rise
of the lake level, which sealed the site immediately after its
abandonment (Nadel, 2002a; Nadel et al., 1995, 2002, 2004). The
plant material, in particular, was protected by two successive
events. The first was charring by fire, which preserved most of the
archaeobotanical remains. Then, very shortly after the site was
abandoned, the water level of the newly formed lake rose. We do
not know whether the rise of the lake level was the reason for the
abandonment. It is clear, however, that the deposition of silt and
clay by the lake sealed the submerged site. This combination of
charring, evidently in low-oxygen conditions (perhaps under the
cover of the hut’s walls), and sealing under lake sediments and
water apparently created ideal conditions for preservation of
organic materials (Weiss, 2002). The vast majority of the site’s plant
remainsdall but 152 seedsdwere charred (Kislev et al., 1992,
2002; Simchoni, 1998; Weiss, 2002). The botanical remains are
distinguished by exquisite preservation. As a result, they can be
identified to the family, genus, and in many cases even to the
species level.
2.2. Site formation processes
Inferences from spatial relationships between archaeological
finds are legitimate only if these finds are contemporaneous and
were laid down by the same deposition-accumulation process.
Tsatskin and Nadel (2003) reconstructed Ohalo II’s site formation
history and distinguished anthropogenic from natural deposits.
They used soil-geomorphic analysis coupled with micromorphology,
SEM/EDS, magnetic susceptibility, and conventional
sedimentological techniques, such as measurements of organic
matter and soluble salts. Their research confirms the field observation
(Nadel and Werker, 1999) of three successive floors in hut 1
and shows that the dark organic layers of brush hut 1 are indeed
human-made floors, rather than effects of natural forces, such as
water and wind (Figs. 2,3) (Tsatskin and Nadel, 2003). Our
reconstruction of human activities that were responsible for the
deposition of copious plant remains on one of these floors points in
the same direction.Fig. 1. Location map of Ohalo II and central area of excavation at the site.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–2414 2401
2.3. The Grinding stone
A grinding stone found in situ in the northern part of floor II in
hut 1 (Figs. 3, 4) apparently was a focal point of activities inside
the hut. This stone is a roughly trapezoidal, flat, 40-cm long slab
of basalt, deliberately embedded in horizontal position in the
floor. It was tentatively named ‘‘the anvil’’ by the excavator
(Nadel, 1997), who remarks on the special attention it received
from the inhabitants of the hut. To keep the stone horizontal and
firmly set within the floor, non-local sand was brought in to form
a bed underneath the grinding stone, further stabilized with small
cobbles (Nadel, 1997). These efforts suggest that its intended use
required the stone to be in a stable, horizontal position inside the
hut.
In order to ascertain whether the stone was used to grind plant
material and specifically establish which taxa were processed,
Piperno et al. (2004) conducted a starch grain study. A total of 150
starch grains were recovered from the grinding stone; 23 of these
were identified as Hordeum sp. grains (Piperno et al., 2004). This
study produced direct evidence demonstrating that the stone did
indeed function as a grinding stone. However, comparative
ethnographic evidence shows that grinding grass grains is not
usually the sole function of grinding stones in hunter-gatherer
societies. Several African and Australian groups employ grinding
stones as multi-purpose tools that serve such diverse functions as
flint tool manufacturing, pounding of ochre and bones, cracking of
nuts and eggs, and processing of bark (Kraybill, 1999; McCarthy,
1946). In light of this comparative evidence and the distributions of
plant remains to be discussed here, it is likely that the Ohalo II
grinding stone was used both as a grinding stone and as a generalpurpose
working surface.
2.4. Sub-sampling floor II
The second of three superimposed floors of hut 1 received
special care in the analysis of the Ohalo II plant remains. This floor is
a kidney-shaped, ca. 3-by-4 m feature, measuring some 12 m2
in
area. Like the rest of the site, it was excavated in 0.5-by-0.5 m units.
Botanical remains from this floor were sampled in a two-step
process. First, we examined all samples from the floor and rated
them according to relative richness in plant remains. Subsequently,
we analyzed the 80% richest samples. To ensure full coverage of the
entire floor area, we also analyzed samples from those units that
initially scored as poor. The last 20%, which were either poor or
originated from squares that had already been sampled, were
skipped. This procedure resulted in a sample of almost 60,000 plant
remains from this floor (Weiss, 2002).
3. Results: plant distribution
Figs. 5–19 and Figs. 22, 23 illustrate the spatial distributions of
various taxa of plant remains in form of continuous kernel density
plots interpolated between the centroids of the excavation grid
squares. As used here, the term taxon (plural taxa) designates any
level of plant identification from species to family. The excavation
squares devoid of botanical samples are marked on the figures by
a diagonal hatch pattern.
Almost 100 taxa are represented among the identified plant
remains from floor II (Simchoni, 1998; Weiss, 2002). The spatial
distributions of those 13 taxa that display distinct concentrations in
restricted floor areas will be discussed in the following sections. As
expected of archaeological assemblages, other taxa are represented
by small quantities with no discernible patterns of distribution. The
13 taxa that will be examined here represent over 80% of the total
plant assemblage from the floor (Table 1).
3.1. Distribution of the total plant assemblage
Before we search for concentrations of individual taxa, we first
need to examine the distribution of the plant assemblage as
a whole. An even distribution of the entire plant assemblage,
without any notable clusters or gaps, would imply that the distribution
of plant remains is not skewed by restricted or unused floor
areas or by recovery bias. In this case, any concentrations or clusters
of remains of one or more individual taxa would be distinctive of
those taxa rather than of the assemblage as a whole. On the other
hand, such concentrations would be much less meaningful if the
bulk of the plant assemblage originated from a limited number of
Fig. 2. Geo-archaeological cross-section through the floors of brush hut 1. square E80b.
1, dark cultural layer and soil; 2, cultural layer mixed with ash; 3, homogeneous lake
marl; 4, bedded lacustrine sand; 5, surface covered with pebbles and sand from Lake
Kinneret, prior to excavation. The numbers along the vertical axis represent meters
below msl (after Tsatskin and Nadel, 2003).
Fig. 3. An east–west cross-section through the floors of brush hut 1. Note the
exaggerated vertical scale. I, II, III, the three floors, stone on rightdthe grinding stone
discussed in text; GB, grass bedding.
Fig. 4. Hut 1 during excavation of floor II, facing east. Note the diamond-shaped
grinding stone in the northern (left) part of the floor.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–24142402
grid squares or from a particular sector of the hut. Under these
circumstances, the distribution might have been created by any one
or a combination of factors during and after the site’s occupation, as
well as during field recovery and laboratory processing, such as
accumulation, preservation, excavation, extraction, sorting, etc.
Patterned distributions might be attributable to differential
preservation rates, too, but given the intractable nature of this
problem it is common practice to disregard it, treating it as yet
another uncontrollable vagary of preservation whose effects
hopefully will cancel each other out.
Fig. 5 shows the distribution of the complete plant assemblage
across the floor. It is clearly a regular, anti-clustered distribution:
most squares contain no more than 3% and even the
highest number of plant remains in any one square (F78c)
represents less than 10% of the total. Since a single taxon,
Suaeda palaestina/fruticosa (sea-blite) (nomenclature in this
paper after Danin, 2004), constitutes around 35% of the total
assemblage, we checked whether and to what extent its distribution
determines that of the entire assemblage. Fig. 6 shows
the distribution of the total plant assemblage from floor II excluding
S. palaestina/fruticosa seeds. The removal of the single
most common taxon does not significantly alter the general
distribution. This suggests that the regular distribution
described is representative of the whole assemblage and no
single contributor is responsible for it. Therefore, we may
conclude that any clusters of a single taxon that may be found
will be indicative of the distribution of that particular taxon and
the activities in which it was involved.
3.2. The distributions of individual taxa
Figs. 7–19 and Figs. 22, 23 display the distributions of major taxa
found on floor II. It is readily apparent that these distributions are
not random. Different taxa are concentrated in different, clearly
delimited zones. The following discussion will demonstrate that
there are correlations both among the patterned distributions of
several plant taxa and between these taxa and the structure and
internal features of the hut. These correlations will then allow us to
offer an explanation of the observed distributions of plant remains.
Since different plants produce different amounts of seeds and
fruits, it is important to distinguish between concentrations of
plant remains resulting from a single deposition of one plant or
fruiting branch and cumulative concentrations that represent
multiple plants or deposition events. Our discussion will take this
distinction into consideration. Please note that each excavation grid
square is approximately 2.5% of the floor area.
The most salient characteristic that leaps to the eye when
examining the different plant distributions is the correlation
between several taxa that cluster around the grinding stone toward
Fig. 5. Distribution of the total plant assemblage, some 60,000 seeds and fruits, on
floor II. Continuous kernel density plot interpolated between the centroids of the
excavation grid squares. Squares devoid of botanical samples are marked by a diagonal
hatch pattern. Scale in units of percent standard deviation from mean density (10%
steps). The following distribution maps, Figs. 5–18 and Figs. 21, 22, follow the same
convention.
Fig. 6. Distribution of the total plant assemblage, excluding S. palaestina/fruticosa
seeds, on floor II (N ¼ 29,451).
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–2414 2403
the north wall of the hut. The densest concentrations are in square
F78c, the northeastern one of four squares that touch the grinding
stone. This square alone produced 15% of all 9904 Bromus pseudobrachystachys/tigridis
(brome) grains (Fig. 7); 45% of the 505
Hordeum marinum/hystrix grains (Fig. 8); 16% of the 606 Hordeum
spontaneum (wild barley) grains (Fig. 9); and 36% of the 819
Piptatherum holciforme (syn. Oryzopsis holciformis) (millet grass)
grains (Fig. 10). In addition, it contained 13% of the 115 Silybum
marianum achenes (Fig. 11); 24% of the 594 Malva parviflora mericarps
(Fig. 12); and 58% of the 134 Melilotus indicus seeds (Fig. 13).
Visual inspection of distribution maps thus suggests that these
seven taxa form a tightly defined cluster around the grinding stone,
with secondary concentrations in the center-south of the floor.
K-means clustering of the excavation grid squares on the densities
(count/m2
) of these seven taxa, assuming two clusters (high vs. low
concentrations), confirms this interpretation of the distribution
maps. Squares with extremely high probabilities of membership in
the ‘‘high concentration’’ cluster form a U-shaped group around the
grinding stone and another block in the center south of the floor.
The result is spatially identical when only the densities of the four
grass (Poaceae ¼ Gramineae) taxa (H. spontaneum, B. pseudobrachystachys/tigridis,
P. holciforme, and H. marinum/hystrix,) are
included in the cluster analysis. Clustering on the remaining three
taxa produces a ‘‘high concentration’’ cluster that includes only one
or perhaps two squares immediately to the north of the grinding
stone. Similarly, local spatial autocorrelation analysis (Anselin’s
Local Moran’s I) (Anselin, 1995) reveals a tight and statistically
highly significant cluster of squares with exceptionally high
densities of the four grass taxa next to the grinding stone (Fig. 14).
However, the secondary concentration of grass grains in the center
south of the floor is not statistically distinguishable from grain
counts in the adjacent squares. The following paragraphs will
discuss the meaning of these associations in light of possible uses of
the plants involved.
The grains of the four grass taxa were part of the diet of the site’s
inhabitants (Weiss et al., 2005), and their grains most probably
were ground on the grinding stone or processed in other ways near
it (Piperno et al., 2004).
Holy thistle (Silybum marianum) has many potential uses. Its
leaves and stalks are eaten both fresh and boiled and its achenes
(fruitlets) have medicinal properties. Pharmaceutical analyses have
discovered that glycoside compounds in its achenes help to cure
liver inflammation and poisoning. Extracts of S. marianum achenes
are a common alternative medicine today (Dafni, 1984; Hedrick,
1972; Palevitch and Yaniv, 2000).
Mallow (Malva parviflora) is an important food plant. Its leaves
and young fruits are eaten fresh, and all its parts are edible when
boiled. The immature mericarps, which are sometimes gathered
separately, are edible and also used in traditional medicine. In India,
the roots are used for washing hair and wool (Beily and Danin,
1981; Dafni, 1984; Zohary and Feinbrun, 1930).
As a note of caution, we must point out that medicinal uses by
present or historic societies, as documented in the literature, do
not prove that prehistoric people used these plants in the same
Fig. 7. Distribution of Bromus pseudobrachystachys/tigridis (9904 grains) on floor II. Fig. 8. Distribution of Hordeum marinum/hystrix (505 grains) on floor II.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–24142404
way. Only archaeological contexts such as gut contents, coprolites
(palaeofaeces), or stew remains within a cooking pot, provide
direct evidence for human use. Given that there is no such
evidence from Ohalo II, the references to medicinal uses are
meant merely to call attention to the wide range of potential uses
of the plants found on floor II. However, we do want to suggest
that concentrations of these plants as dense as those we found on
floor II in direct association with a grinding stone constitute the
strongest of indirect evidences of their use. Additionally, it is
important to notice that the relative quantities needed in
medicinal plants are very small.
Melilotus indicus displays the most densely packed distribution
of all taxa in this group (Fig. 13). Eighty-four percent of the 134
Melilotus seeds come from two squares, F78a and F78c. M. indicus
pods grow on multi-pod racemes; each pod usually contains
a single seed. Thus, the concentration on floor II may represent
a single deposition event. M. indicus is used in India to stop bleeding
and as narcotic (Boulos, 1983; Townsend and Guest, 1974). In Egypt,
seeds of this species are used to treat diseases of the genital organs
of both sexes (Boulos, 1983).
Two additional taxa were heavily concentrated in squares F78a
and F78c. 85% of the 31 fumitory (Fumaria macrocarpa) nutlets and
60% of the 55 Aegilops spp. (including Ae. peregrina and Ae.
geniculata/peregrina) grains were found in square F78c.
The dense concentrations of food plants in square F78c point to
food preparation or storage activities in this area. Medicinal plants
may also have been processed on or near the grinding stone. We
previously mentioned ethnographic evidence of grinding stones
serving as multipurpose working surfaces in hunter-gatherer
societies (Kraybill, 1999; McCarthy, 1946). The fact that the
concentrations of plant remains around the floor II grinding stone
include both plants that do and others that do not require pulverizing
as preparation for human use makes sense in light of these
ethnographic observations.
Similarly high densities of plant remains as in F78a and F78c
were observed in two adjacent grid squares, including H. spontaneum,
P. holciforme, and S. marianum in E80b and B. pseudobrachystachys/tigridis
and H. marinum/hystrix in F79c. These two
squares are adjacent to each other and to the grinding stone. A
fourth square with much lower densities of botanical remains is
between the high-density squares and also adjacent to the
grinding stone. A plausible explanation of this U-shaped
constellation of densely distributed plant remains around
a grinding stone and a single square notably poorer in plant
remains is a person squatting in the low-density square F78a,
operating the grinding stone.
Also clustered around the grinding stone were the dispersal
units of Atriplex rosea and A. leucoclada (orache) (Fig. 15). Due to
their morphological proximity, only half of these could be identified
to the species level, while the rest was identified as A. rosea/
leucoclada. Fig. 15 therefore illustrates the pooled distribution of
Atriplex on the floor. It is uniformly sparse across the floor, with one
major concentration in the squares surrounding the grinding stone,
Fig. 9. Distribution of Hordeum spontaneum (606 grains) on floor II. Fig. 10. Distribution of Piptatherum holciforme (819 grains) on floor II.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–2414 2405
which contained about 50% of the total of 1300 Atriplex fruits. This
large number of fruits clearly represents several fruiting branches.
Young A. rosea plants are eaten pickled (Zohary and Feinbrun,
1930). All local species of this genus are eaten either fresh or
pickled (Dafni, 1984; Palevitch and Yaniv, 2000).
It is likely that Atriplex branches were used in the construction
of the Ohalo II hut. Charcoal analysis by Nadel and Werker (1999)
of remains of the walls of hut 1 identified Atriplex/Seidlitzia
charcoal. According to these authors, Atriplex is the more plausible
of the two identifications since the only Seidlitzia species in Israel,
S. rosmarinus, thrives in hot deserts and is not native to the Ohalo
II region.
Thus, there are two plausible explanations for the presence of
Atriplex fruits at Ohalo II. Either the fruits were used for food, which
was prepared on or near the grinding stone, or the branches might
have been used for roofing, more so on the northern side of the hut,
and their fruits accidentally fell to the floor. These two explanations
are not mutually exclusive.
The distribution of Rubus sanguineus/canescens (bramble)
nutlets is also densely clustered (Fig. 16). Seventy-two percent of
the 175 nutlets on floor II are concentrated in four adjacent
squares, i.e. a single square meter. Rubus fruits are apocarpic;
their many carpels develop into many nutlets covered with juicy
pulp. Since Rubus fruits grow in groups, the 175 nutlets on floor II
might represent several fruits from one branch or several
branches. About half of the nutlets are not charred. Due to the
very high lignin content in their seed coating, some
archaeological Rubus seeds preserve in an uncharred state (G.
Hillman, personal communications). The fleshy Rubus fruits
prevent fire from touching most of the nutlets, but this protection
was not sufficient for all them. The concentration of the
Rubus nutlets within a one-square-meter area near the southwest
end of the grinding stone points to a single deposition
event. Rubus fruits are edible and nutlets of several Rubus species
have been found at many European sites, from the Neolithic
period onward (Renfew, 1973; Zohary and Hopf, 1994). It is most
likely that these fruits were eaten by the Ohalo II dwellers, as
they are eaten today.
Rubus fruits are juicy, fragile, and difficult to transport. Usually,
they are consumed right in the field on an eat-as-you-go-along
basis. Why, then, were these berries carried back to the Ohalo II
hut? They may have been dried by fire or in the sun for storage,
a practice reported for several native North American groups (e.g.
Moerman, 1999). In any event, this is likely an early case of
advanced planning of plant food consumption.
Uncharred plant remains reveal an interesting pattern of
distribution on floor II. They form two concentrated patches, one
mainly composed of the Rubus nutlets, the other in the northeast
corner of the floor. Besides these 82 Rubus nutlets, only 56
uncharred seeds have come to light at Ohalo II, and all of them are
from this floor. Apparently, the conditions of preservation on floor II
were exceptional even by Ohalo II standards.
Adonis dentata/microcarpa (peasant’s eye) is another taxon with
a clustered distribution (Fig. 17). Some 30% out of the 71 mericarps
Fig. 11. Distribution of Silybum marianum (115 achenes) on floor II. Fig. 12. Distribution of Malva parviflora (594 mericarps) on floor II.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–24142406
were found in two adjacent squares, or one-half square meter, and
the bulk of these mericarps tend to come from the center-north of
the floor, rather than from its perimeter. Adonis spp. fruits are
spike-like, dense, and apocarpic, bearing many mericarps. Thus, it is
possible that all these remains stem from a single plant. A. vernalis
contains the glycoside Adonidin, which in Europe has been used as
an effective medicine for the heart. It increases the activity of all
types of muscle tissue, regulates the pulse, and elevates blood
pressure (Grieve, 1984). Local traditional healers use a decoction of
the whole dry plant for the same purposes (Heyn and Pazy, 1989).
Umbels (flowers) and umbellules (partial umbel) of the
Apiaceae (¼Umbelliferae) family were concentrated in and south of
the entrance to the hut (Fig. 18). All these umbels and umbellules
are similar and apparently belong to the same species. 36% of them
were found in a single square (F79d) and some 60% came from
three adjacent grid squares (0.75 m2
). We were unable to identify
this Apiaceae beyond the family level, as these are flowers and not
fruits, and some of them had their petals rolled inward. The
inflorescence of the Apiaceae family is an umbel that is often made
up of several umbellules. The Ohalo II umbels were composed of
eight umbellules. Therefore, they may be the remains of one or two
plants, if they were laid down in a single deposition event.
Apiaceae were not the only family of flowers represented in
the Ohalo II botanical assemblage. Square F80d, located in the
entrance area of the hut, south of square F79d where the
Apiaceae flowers concentrated, produced two Senecio
glaucus (groundsel) capitula (heads) (Compositae (¼Asteraceae))
(Figs. 19, 20). In addition, 20 Compositae heads were found just
outside the entrance (square G80c). S. glaucus flowers from
February to April, and during seed dispersal its involucre bracts
bend backward (Mabberley, 2002). The involucres of the Ohalo II
specimens are erect (Fig. 19), and their achenes are mature
(Fig. 20). Both observations suggest that at the time of charring
these heads were near the end of their flowering period. Because
we could not identify the 20 Compositae heads to their species,
we could not reconstruct their exact flowering months. However,
since all of these heads came from a single excavation grid
square and spring is the flowering season of most local plants, it
is safe to infer that these flowers were picked in the spring.
Thus, we have both indirect spatial and direct evidence for the
seasonality of the hut’s occupation. S. glaucus provides direct
flowering dates, and the provenience of the two heads points to
a single deposition episode. Therefore, the botanical macro
remains in the entrance area support an occupation of the hut
during the spring or early summer.
The appearance of S. glaucus in Ohalo II is not self-evident. First,
its identification relies on minute differences in bracts whorls on
the flower head. In S. glaucus, thin and pointed bracts are present in
Fig. 13. Distribution of Melilotus indicus (134 seeds) on floor II.
Fig. 14. Local spatial autocorrelation (Anselin’s Local Moran’s I) of counts of four grass
taxa (H. spontaneum, B. pseudobrachystachys/tigridis, P. holciforme, and H. marinum/
hystrix,) on floor II: cluster membership probabilities. The grid squares next to the
grinding stone display exceptionally high densities of the four grass taxa that are
statistically distinguishable from those of surrounding grid squares. In contrast, the
secondary concentration in the center south of the hut (see Fig. 14) is not statistically
significant.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–2414 2407
a lower whorl in the base of the head, attaches, and go beyond the
base of the upper whorl, while in other species in the same genus
they are shorter (Fig. 19). Second, S. glaucus chorotype is IranoTuranian
and Saharo-Arabian, growing today as north as Samarian
Desert and Lower Jordan Valley, some 40–60 km south of the site
(Danin, 2004). Therefore, the appearance of southern plant in Ohalo
represent the drier conditions prevailed in that time (Weiss et al.,
2005).
While the distribution of Puccinellia cf. convoluta grains is
non-uniform like those of the other taxa discussed so far (Fig. 21),
the pattern is a markedly different one, with a mean center at the
southern extreme of the hut, close to the perimeter wall. 34% of the
673 Puccinellia grains were found in squares E82b and E82d, an
additional 14% in square F82b just east of these two, and no more
than four 0.25-m2
squares at the south end of the hut account for
54% of all Puccinellia grains. The possible use of Puccinellia grains for
food is discussed elsewhere (Weiss et al., 2005). If these grains were
actually eaten, their preparation and/or consumption must have
differed in some way from those of other food plants, which were
deposited in the northern part, near the grinding stone. Weiss et al.
(2005) and Piperno et al. (2004) point out that the minute Puccinellia
grains are similar to the smallest known grains used for food
today, those of Eragrostis tef (tef). Moreover, the Puccinellia
inflorescence is a panicle bearing hundreds of grains, so the grains
found at Ohalo II may conceivably originate from as few as one or
two plants. We suggest, therefore, that the Puccinellia distribution
attests to a pattern of use different from that of other grasses,
possibly due to the minute size of Puccinellia grains.
The concentration of Puccinellia grains near the south wall is
reminiscent of the Puccinellia grass bedding found in the same
area directly underneath on floor III (Nadel et al., 2004). On floor
III, the residents of the hut had created a comfortable cover by
spreading a layer of Puccinellia shoots oriented with their base
toward the center of the hut and the heads toward the walls.
However, the southern part of floor II was also used for flint
knapping (see below). Considering the typically opportunistic
nature of hunter-gatherer economic behavior, the use of the
shoots as bedding material does not preclude the use of the
grains as a source of food.
A different pattern of distribution was observed for Suaeda
palaestina/fruticosa seeds (Fig. 22). Although the relative distribution
of the 28,373 seeds is rather uniform across most of the
squares, those squares that display modestly higher densities are
located along the hut’s perimeter walls. Moreover, while most
taxa are absent from at least a few grid squares, S. palaestina/
fruticosa seeds are ubiquitous. Villagers in the Euphrates valley
in northern Syria used to collect large quantities of S. fruticosa
branches for soap making, in much the same way as Salsola kali
(Hillman, 1995; Mabberley, 2002). Since it is hard to imagine
that the kind of fire required for soap making was lit inside the
hut, it is likely that the tens of thousands of Suaeda palaestina/
fruticosa seeds came from the branches that formed the walls of
the hut, as suggested by Nadel and Werker (1999). The huge
number of seeds found on floor II represents at least several
branches.
Fig. 15. Distribution of Atriplex rosea/leucoclada (1228 fruits and seeds) on floor II.
Fig. 16. Distribution of Rubus sanguineus/canescens (175 nutlets) on floor II.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–24142408
3.3. Reconstruction of activity areas
The above review of the distributions of various plant taxa on
floor II makes one point abundantly clear. All taxa whose
distributions are not uniform are concentrated in the northern
and southern parts of the floor, separated by a west-east triangle
of relatively low density of plant remains. The base of this
triangle is the middle portion of the hut’s west wall, its apex is
near the middle of the east wall. Given that the entire floor was
excavated in the same way and the same retrieval and analysis
techniques were used, recovery or analytical biases clearly do not
account for the patterned distribution of botanical remains. Since
there are good reasons to believe that this floor was rapidly
abandoned and sealed (see above), and the plant assemblage
unequivocally represents a single occupation, the patterned
distribution of botanical remains allows us to infer two plantrelated
activity areas inside the hut, one in the north and one in
the south, separated by an entrance area (Fig. 23). The northern
activity area is centered around the grinding stone, which was
used for the preparation of food (Piperno et al., 2004) and
possibly medicinal plants. There is limited evidence to suggest
that the southern activity area was used for sleeping. Recently,
Nadel et al. (2004) described the use of stems and leaves of
Puccinellia cf. convoluta as bedding material on the third floor of
this hut, just below this floor. Most of floor III was covered with
grasses, with the exception of the central part, where a hearth
was found. We suggest, therefore, that the concentration of
Puccinellia cf. convoluta grains near the southern wall of the hut
on floor II, represents a comparable arrangement of bedding
material. Apparently, the bedding material covering the floor did
not prevent other activities; as flint knapping was practiced there
as well, see below.
4. Discussion: implications for the economy and social
organization of Upper Paleolithic Ohalo II
Our exhaustive analysis of most of the Upper Paleolithic plant
assemblage from floor II of hut 1 paints a consistent, representative
picture of relevant activities on this floor. These findings,
along with the excellent preservation of plant remains retrospectively
validate our decision to embark on such an extensive
analysis of the plant assemblage. Any sub-sampling technique
would likely have failed to reveal the spatial patterns described
in this paper. The Ohalo II assemblage proved deserving of
exhaustive analysis, as most taxa represented by an adequate
number of specimens, say a few dozen or more, produced
meaningful patterns of distribution.
These results allowed us to understand some aspects of the way
of life of the hut’s late Upper Paleolithic residents and their use of
its interior space, which was consciously planned rather than
haphazard. In our opinion, the ability to reconstruct past human
activities with such a high degree of spatial resolution from the
Fig. 17. Distribution of Adonis dentata/microcarpa (71 mericarps) on floor II. Fig. 18. Distribution of Apiaceae (Umbelliferae) family (253 umbels and umbellules) on
floor II.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–2414 2409
distribution of plant remains makes an exhaustive analysis of
a large plant assemblage well worth the effort.
We have demonstrated elsewhere (Piperno et al., 2004) that
the stone placed on the floor functioned as a grinding stone.
The distribution of the plant assemblage relative to the grinding
stone, in particular the concentrations of grass grains surrounding
it, reinforces this functional interpretation of the
stone. The clusters of two groups of plant remains around the
grinding stonedfood and medicinal plantsdsuggest that this
stone was used for grinding and as a general-purpose working
surface. Australian aborigines, for example, reportedly use
grinding stones for an array of tasks (Kraybill, 1999; McCarthy,
1946). Inferences of medicinal use of plants in a prehistoric
context are inherently uncertain. However, the massive presence
of plants with medicinal qualities clearly shows that such
plants were available to and collected by the Upper Paleolithic
residents of the site. The presence of hundreds of seeds of this
group in the archaeological assemblage makes it exceedingly
unlikely that these plants were accidentally deposited or drifted
into the hut. Some of these plants are still in use in the area, but
we do not know whether this is an unbroken tradition transmitted
from generation to generation since the area’s early
prehistory.
Our spatial analysis revealed several patterns in the distribution
of plant remains on floor II. Different groups of plants display
distinct distributions. Food and possibly medicinal plants tend to
cluster around the grinding stone in the north, while plants
probably used as roofing material, such as Atriplex and Suaeda, tend
to be distributed uniformly across the floor. Some taxa show
concentrations in restricted areas, sometimes as small as one
square meter or less; these include A. dentata/microcarpa, B. pseudobrachystachys/tigridis,
F. microcarpa, M. indicus, P. holciforme, P.
convoluta, R. sanguineus/canescens, and Apiaceae. This pattern of
distribution is in agreement with the findings of Tsatskin’s
(Tsatskin, 2002; Tsatskin and Nadel, 2003) soil micromorphology
and magnetic susceptibility studies. Both approaches independently
show that the accumulation of materials on this floor
is the result of human activities rather than natural forces. Wind
and water, for example, would hardly have selectively concentrated
some taxa but not others around the grinding stone. Differential,
non-uniform, clustered distributions of several taxa cannot be
attributed to such forces. The apparent stockpiling of juicy Rubus
berries, which are fragile and usually eaten on the spot, suggests
long-term planning and storage.
There is, in fact, evidence that such an individual did live at
Ohalo II. Locus 5 (Fig. 2) is a grave of an adult male who, in addition
to suffering from a disabled hand, also sustained a penetrating
Fig. 19. SEM photograph of archaeological Senecio glaucus (groundsel) head. The head
is seen from the outside. It has five upper whorl bracts; the one on the left is broken.
Attached to these bracts is another, thinner and pointed bract from a lower whorl
(Weiss, 2002).
Table 1
Representative plant assemblage from floor II
Taxon (organ) Quantity Figure no.
Adonis dentata/microcarpa (mericarp) 71 Fig. 17
Apiaceae (¼Umbelliferae) (umbel and umbellule) 253 Fig. 18
Atriplex rosea/leucoclada (fruit and seed) 1228 Fig. 15
Bromus pseudobrachystachys/tigridis (grain) 9904 Fig. 7
Hordeum marinum/hystrix (grain) 505 Fig. 8
Hordeum spontaneum (grain) 606 Fig. 9
Malva parviflora (mericarp) 594 Fig. 12
Melilotus indicus (seed) 134 Fig. 13
Piptatherum holciforme (grain) 819 Fig. 10
Puccinellia cf. convoluta (grain) 673 Fig. 21
Rubus sanguineus/canescens (nutlet) 175 Fig. 16
Silybum marianum (achene) 115 Fig. 11
Suaeda palaestina/fruticosa (seed) 28373 Fig. 22
Fig. 20. SEM photograph of another archaeological S. glaucus head. This photo shows
the outside of the head. In the foreground, three of the originally five achenes, with
bracts behind them. The achenes are covered with short hairs; the pappus’ attachment
scar can be seen on their top. The depressions in front of the achenes indicate where
the missing achenes were attached to the head (Weiss, 2002).
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–24142410
wound to the rib cage (Hershkovitz et al.,1993). The survival of such
a disabled and heavily wounded individual is strong evidence of
medicinal care giving by members of the Ohalo II late Upper
Paleolithic group.
The accumulation of flowers in the entrance area demonstrates
both the residents’ appreciation of different types of
flowers and the use of the hut during the spring. It is hard to say
whether these flowers served practical or aesthetic purposes.
The S. glaucus capitula are in a stage after the beginning of
flowering, maybe not strictly a flower at the time they were
preserved by fire. However, since these stages succeed one
another quite rapidly and there is no known economic use for S.
glaucus seeds, we have to conclude that the Upper Paleolithic
residents of Ohalo II brought them home as flowers. This is, to
the best of our knowledge, the earliest evidence of human use of
flowers. Three decades ago, the discovery of the Shanidar IV
‘‘flower burial’’ (Leroi-Gourhan, 1975a,b; Solecki and Solecki,
1974) made enormous waves, opening a cultural and even
spiritual dimension of Neanderthal life. It has since been argued
convincingly, however, that a local rodent, Meriones persicus
(Persian Jird), was most likely responsible for the introduction of
pollen into the cave (Sommer, 1999), making the Ohalo II floral
assemblage the earliest known human use of flowers. Given that
the residents of Ohalo II were Anatomically Modern Humans
(Hershkovitz et al., 1995), their appreciation of flowers is much
less surprising and sensational than was that by the Shanidar
Neanderthals.
5. Distribution of flint tools and debitage
Like the plant remains, the flint assemblage from floor II also
displays a patterned distribution (Nadel,1997, 2002a,b). The analyzed
assemblage includes all retrieved pieces (N ¼ 8127). Among these,
the debitage category (N ¼ 1840) is dominated by bladeletsdthe
target products, accompanied by blades, flakes, primary elements,
core trimming elements, and cores. There are 132 retouched tools,
and the assemblage as a whole reflects all stages of core reduction.
The class of minute flints (N ¼ 6155) includes several categories,
among which tiny bladelets and flakes, as well as fragments of
regular bladelets are conspicuous. There are also many angular and
fire-cracked fragments. All of these categories display high find
densities within the 2-m2
area near the entrance (Fig. 24; Nadel
et al., 2006). Furthermore, square E81b has one core, the highest
per-square numbers of bladelets (62) and core trimming elements
(7), the second highest of retouched tools (11), and the highest of
minute pieces (340).
The analysis of the flint assemblage reveals a concentration of
flint products spatially dissociate from those of plant remains. Flints
are concentrated in the southern part of the hut, in front of the
entrance. This concentration includes all stages of flint production,
from heavy cores and primary elements to a wealth of bladelets
(the principal knapping product) and thousands of tiny, millimetersized
specimens. No natural processes could have created the
observed pattern. Rather, it points to two or three persons knapping
flint near the entrance, facing the light from the door (61).
Fig. 21. Distribution of Puccinellia cf. convoluta (673 grains) on floor II. Fig. 22. Distribution of Suaeda palaestina/fruticosa (28,373 seeds) on floor II.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–2414 2411
6. Conclusions: spatial distribution to suggest
gender-related use of space
The well-defined distribution patterns of plant remains and flint
products on the same floor reveal a structured division of space
inside this Upper Paleolithic brush hut. One part was dedicated to
plant processing, with a grinding stone firmly set in a unique way.
The other part was devoted to flint knapping. Naturally, seeds and
flints are not restricted to the two concentrations, and it is possible
that in several occasions certain relevant activities were not
exclusive to the two areas.
While this spatial separation of tasks is abundantly evident, it is
not clear whether it maps onto a gender-specific division of labor
(Nadel et al., 2006). The division of labor between male and female
is deep-rooted in many human societies, though culture-specific
variability is high. The anthropological literature indicates that the
two groups of tasks differ in their degree of male-female affinity.
Plant-food preparation is reported to be an almost exclusively
female task. Flint knapping, on the other hand, is commonly a male
task, but exceptional cases of female knappers have been
documented as well (Hays-Gilpin and Whitley, 1998; Kelly, 1995;
Kent, 1998; Kuhn and Stiner, 2006).
Consequently, the spatial separation of such tasks is to be interpreted
with caution. We can only suggest that distinct locations
dedicated to seed processing and flint knapping within the Ohalo II
hut reflect a gendered division of floor space. This would be the oldest
documented case of a gendered division of domestic space, with men
preparing flint tools in the well-lit area near the entrance and women
processing food in the darker northern part of the hut (Fig. 25).
The combined picture of two distinct indoor activities provides
a rare glimpse into the way in which Upper Paleolithic huntergatherers
in the Jordan Valley organized and conceived of their
Fig. 23. Suggested use of space on floor II, as inferred from distributions of plant
remains.
Fig. 24. Distribution plan of flint objects on floor II, brush hut 1: cores and bladesbladelets.
Presented as densities calculated for 0.5 Â 0.5 excavation units (Nadel et al.,
2006).
Fig. 25. A reconstruction of brush hut 1 showing the location of two activity areas on
the floor: flint knapping on the left and seed processing on the right. Drawing by
Rachel Brown-Goodman (Nadel et al., 2006).
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–24142412
dwelling space. It might also supply us with an insight into their
male–female interrelationship.
Acknowledgments
We would like to thank Ofer Bar-Yosef and Wilma Wetterstrom
(Harvard University), for their comments and improvements of
earlier versions of this paper. This Center of Excellence (No. 300/06)
was supported by the Israel Science Foundation. The Ohalo II
project was generously supported by the Irene-Levi Sala CARE
Archaeological Foundation, the Israel Science Foundation (NO. 831/
00), the Jerusalem Center for Anthropological Studies, the L.S.B.
Leakey Foundation, the Stekelis Museum of Prehistory in Haifa, the
MAFCAF Foundation, the National Geographic Society and the Israel
Antiquities Authority.
References
Anselin, L., 1995. Local indicators of spatial association. Geographical Analysis 27,
93–115.
Balme, J., Beck, W.E., 2002. Starch and charcoal: useful measures of activity
areas in archaeological rock shelters. Journal of Archaeological Science 29,
157–166.
Bartov, Y., Stein, M., Enzel, Y., Agnon, A., Reches, Z., 2002. Lake levels and sequence
stratigraphy of Lake Lisan, the Late Pleistocene precursor of the Dead Sea.
Quaternary Research 57, 9–21.
Bar-Yosef Mayer, D., 2002. The use of Mollusk shells by fisher-hunter-gatherers at
Ohalo II. In: Nadel, D. (Ed.), Ohalo IIdA 23,000-Year-Old Fisher-HunterGatherers’
Camp on the Shore of the Sea of Galilee, Hecht Museum. University
of Haifa, pp. *39–*41. Catalogue No. 20pp.
Beily, C., Danin, A., 1981. Bedouin plant utilization in Sinai and the Negev. Economic
Botany 35, 145–162.
Belitzky, S., Nadel, D., 2002. The Ohalo II prehistoric camp (19.5 Ky): new evidence
for environmental and tectonic changes at the Sea of Galilee. Geoarchaeology:
An International Journal 17, 453–464.
Belmaker, M., Nadel, D., Tchernov, E., 2001. Micromammal taphonomy in the site of
Ohalo II (19 ky., Jordan Valley). Archaeofauna 10, 125–135.
Binford, L.R., 1978. Dimensional analysis of behavior and site structure: learning
from an Eskimo hunting stand. American Antiquity 52, 30–61.
Binford, L.R., 1981. Bones: Ancient Man and Modern Myths. Academic Press, New
York.
Binford, L.R., 1983. In Pursuit of the Past. Academic Press, London.
Boulos, L., 1983. Medicinal Plants of North Africa. Reference Publications, Algonac,
Michigan.
Clark, G.A., Stafford, C.R., 1982. Quantification in American archaeology: a historical
perspective. World Archaeology 14, 98–119.
Clarke, D.L., 1977. Spatial information in archaeology. In: Clarke, D.L. (Ed.), Spatial
Archaeology. Academic Press, London, pp. 1–32.
Dafni, A., 1984. Edible wild plants. Society for the Protection of Nature in Israel,
Jerusalem. in Hebrew.
Danin, A., 2004. Distribution Atlas of Plants in the Flora Palaestina Area. The Israel
Academy for Sciences and Humanities, Jerusalem.
Goldberg, P., 2000. Micromorphology and site formation at Die Kelders Cave I, south
Africa. Journal of Human Evolution 38, 43–90.
Goldberg, P., Bar-Yosef, O., 1998. Site formation processes in Kebara and Hayonim
caves and their significance in Levantine prehistoric caves. In: Azakawa, T.,
Aoki, K., Bar-Yosef, O. (Eds.), Neanderthals and Modern Humans in Western
Asia. Plenum Press, New York, pp. 7–125.
Gowlett, J.A.J., 1997. High definition archaeology: ideas and evolution. World
Archaeology 29, 152–171.
Grieve, M., 1984. A Modern Herbal. The Medicinal, Culinary, Cosmetic, and Economic
Properties, Cultivation and Folklore of Herbs, Grasses, Fungi, Shrubs and
Trees with all their Modern Scientific Uses. Tiger Books International, London.
Hayden, B., 1997. Observation on the prehistoric social and economic structure of
the North American Plateau. World Archaeology 29, 242–261.
Hayden, B. (Ed.), 2000. Socioeconomy. Archaeology Press, Burnaby, B.C.
Hays-Gilpin, K., Whitley, D.S. (Eds.), 1998. Reader in Gender Archaeology. Routledge,
London.
Hedrick, U.P., 1972. Sturtevant’s Edible Plants of the World. Dover, New York.
Hershkovitz, I., Edelson, G., Spiers, M., Arensburg, B., Nadel, D., Levi, B., 1993.
Ohalo II mandunusual finding in the anterior rib cage and shoulder girdle of
a 19,000 year-old specimen. International Journal of Osteoarchaeology 3,
177–188.
Hershkovitz, I., Spiers, M.S., Frayer, D., Nadel, D., Wish-Baratz, S., Arensburg, B., 1995.
Ohalo II H2: A 19,000-year-old skeleton from a water-logged site at the Sea of
Galilee, Israel. American Journal of Physical Anthropology 96, 215–234.
Heyn, C.C., Pazy, B., 1989. The annual species of Adonis (Ranunculaceae): a polyploid
complex. Plant Systematics and Evolution 168, 181–193.
Hillman, G.C., 1995. Lecture Notes. Institute of Archaeology, UCL.
Kelly, R.L., 1995. The Foraging Spectrum: Diversity in Hunter-Gatherer Lifeways.
Smithsonian Institution Press, Washington and London.
Kent, S. (Ed.), 1998. Gender in African Prehistory. Sage Publications, London.
Kislev, M.E., Nadel, D., Carmi, I., 1992. Epipalaeolithic (19,000 BP) cereal and fruit
diet at Ohalo II. Sea of Galilee. Review of Palaeobotany and Palynology 73,
161–166.
Kislev, M.E., Simchoni, O., Weiss, E., 2002. Reconstruction of the landscape, human
economy, and hut use according to seeds and fruit remains from Ohalo II. In:
Nadel, D. (Ed.), Ohalo IIdA 23,000-Year-Old Fisher-Hunter-Gatherers’ Camp on
the Shore of the Sea of Galilee, Hecht Museum. University of Haifa, Haifa, pp.
*21–*23. Catalogue No. 20pp.
Kraybill, N., 1999. Pre-agricultural tools for the preparation of foods in the Old
World. In: Anderson, P. (Ed.), Prehistory of Agriculture, New Experimental and
Ethnographic Approaches, Institute of Archaeology. University of California, Los
Angeles, pp. 489–519.
Kuhn, S.L., Stiner, M.C., 2006. What’s a mother to do? The division of labor
among Neanderthals and modern humans in Eurasia. Current Anthropology
47, 953–980.
Leroi-Gourhan, A., Shanidar Iv, A., 1975a. Neanderthal flower burial in Northern
Iraq. Science 190, 880–881.
Leroi-Gourhan, A., 1975b. The flowers found with Shanidar IV, A Neanderthal burial
in Iraq. Science 190, 562–564.
Mabberley, D.J., 2002. The Plant-Book, A Portable Dictionary of the Vascular Plants.
Cambridge University Press, Cambridge.
McCarthy, F.A., 1946. The Stone Implements of Australia. Australian Museum
Memoir 9.
Moerman, D.E., 1999. Native American Ethnobotany. Timber Press, Portland,
Oregon.
Nade, D., 1997. The spatial organization of prehistoric sites in the Jordan Valley:
Kebaran, Natufian and Neolithic case studies, Ph.D. Thesis, Institute of
Archaeology, The Hebrew University, Jerusalem.
Nadel, D., 2002a. Ohalo IIdA 23,000-Year-Old Fisher-Hunter-Gatherers’ Camp on
the Shore of the Sea of Galilee, Hecht Museum. University of Haifa, Haifa.
Catalogue No. 20.
Nadel, D., 2002b. Indoor/outdoor flint knapping and minute debitage remains: the
evidence from the Ohalo II submerged camp (19.5 ky, Jordan Valley). Lithic
Technology 26, 118–137.
Nadel, D., Werker, E., 1999. The oldest ever brush hut plant remains from Ohalo II,
Jordan Valley, Israel (19 ka BP). Antiquity 73, 755–764.
Nadel, D., Danin, A., Werker, E., Schick, T., Kislev, M.E., Stewart, K., 1994. 19,000-yearold
twisted fibers from Ohalo II. Current Anthropology 35, 451–458.
Nadel, D., Carmi, I., Segal, D., 1995. Radiocarbon dating of Ohalo II: archaeological
and methodological implications. Journal of Archaeological Science
22, 811–822.
Nadel, D., Tsatskin, A., Bar-Yosef Mayer, D.E., Belmaker, M., Boaretto, E.,
Kislev, M.E., Hershkovitz, I., Rabinovich, R., Simmons, T., Weiss, E., Zohar, I.,
Asfur, O., Emmer, G., Ghraieb, T., Grinberg, U., Halabi, H., Weissbrod, L.,
Zaidner, Y., 2002. The Ohalo II 1999–2000 seasons of excavation: a preliminary
report, Mitekufat Haeven. Journal of the Israel Prehistoric Society
32, 17–48.
Nadel, D., Weiss, E., Simchoni, O., Tsatskin, A., Danin, A., Kislev, M., 2004. Stone
Age hut in Israel yields world’s oldest evidence of bedding. Proceedings of
the National Academy of Sciences of the United States of America 101,
6821–6826.
Nadel, D., Weiss, E., Tschauner, H., 2006. Gender specific space division of domestic
activities during the Upper Palaeolithic? A brush hut floor as a case study. UISPP
Conference, Lisbon.
Palevitch, D., Yaniv, Z., 2000. Medicinal Plants of the Holy Land. Modan Publishing
House, Tel-Aviv.
Piperno, D.R., Weiss, E., Holst, I., Nadel, D., 2004. Starch grains on a ground stone
implement document Upper Paleolithic wild cereal processing at Ohalo II,
Israel. Nature.
Rabinovich, R., 2002. The mammal bones: environment, food and tools. In: Nadel, D.
(Ed.), Ohalo IIdA 23,000-Year-Old Fisher-Hunter-Gatherers’ Camp on the Shore
of the Sea of Galilee, Hecht Museum. University of Haifa, Haifa, pp. *24–*27.
Catalogue No. 20pp.
Rabinovich, R., Nadel, D., 1994–5. Bone tools from Ohalo IIda morphological
and functional study, Mitekufat Haeven. Journal of Israel Prehistoric Society
26, 32–63.
Rabinovich, R., Nadel, D., 2002. Broken mammal bones: taphonomy and food
sharing at the Ohalo II submerged prehistoric camp, Archaeozoology of the
Near East. ARC-Publicaties 62, Groningen Six symposium, The Netherlands/
Irbid London, pp. 32–63.
Renfew, J.M., 1973. Palaeoethnobotany, the Prehistoric Food Plants of the Near East
and Europe. Columbia University Press, New York.
Simchoni, O., 1998. Reconstruction of the Landscape and Human Economy 19,000
BP in the Upper Jordan Valley by the Botanical Remains Found at Ohalo II, Ph.D.
Thesis, Faculty of Life Sciences, Bar-Ilan University, Ramat Gan.
Simmons, T., 2002. The birds from Ohalo II. In: Nadel, D. (Ed.), Ohalo IId A
23,000-Year-Old Fisher-Hunter-Gatherers’ Camp on the Shore of the Sea of
Galilee, Hecht Museum. University of Haifa, Haifa, pp. *32–*36. Catalogue No.
20pp.
Simmons, T., Nadel, D., 1998. The avifauna of the early Epipalaeolithic site of Ohalo II
(19,400 years B.P.), Israel: Species diversity, habitat, and seasonality.
International Journal of Osteoarchaeology 8, 79–96.
Solecki, R.L., Solecki, R.S., 1974. Letters: Shanidar Cave. Science 184, 937.
Sommer, J.D., 1999. The Shanidar IV ‘Flower Burial’: a re-evaluation of Neanderthal
burial ritual. Cambridge Archaeological Journal 9, 127–129.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–2414 2413
Townsend, C.C., Guest, E., 1974. Melilotus. In: Townsend, C.C., Guest, E. (Eds.), Flora
of Iraq, pp. 142–149.
Tsatskin, A., 2002. Geoarchaeology of prehistoric campsite on the shore of Lake
Kinneret. In: Nadel, D. (Ed.), Ohalo IId A 23,000-Year-Old Fisher-HunterGatherers’
Camp on the Shore of the Sea of Galilee, Hecht Museum. University
of Haifa, Haifa, pp. *56–*59. Catalogue No. 20pp.
Tsatskin, A., Nadel, D., 2003. Formation processes at the Ohalo II submerged
prehistoric campsite, Israel, inferred from soil micromorphology and magnetic
susceptibility studies. Geoarchaeology: An International Journal 18, 409–432.
Vaquero, M., Pasto, I., 2001. The definition of spatial units in Middle Palaeolithic
sites: the hearth-related assemblages. Journal of Archaeological Science 28,
1209–1220.
Weiss, E., 2002. Issues in reconstruction the human economy and society of the
Epipalaeolithic site Ohalo II inhabitants by the macrofossil botanical remains,
Ph.D. Thesis, Department of Land of Israel Studies and Archaeology, Bar-Ilan
University, Ramat-Gan, 2002.
Weiss, E., Kislev, M.E., Simchoni, O., Nadel, D., 2005. Small-grained wild grasses
as staple food at the 23,000 year old site of Ohalo II, Israel. Economic Botany,
125–134.
Zohary, M., Feinbrun, N., 1930. How to recognize beneficial plants growing wild
in Eretz-Israel. Hassadeh 10, 298–301. 363-368, 433-437, 608-612. (in
Hebrew).
Zohary, D., Hopf, M., 1994. Domestication of Plants in the Old World, second ed.
Oxford Science Publications, Oxford.
E. Weiss et al. / Journal of Archaeological Science 35 (2008) 2400–24142414