JASC 2011a GulyasSmegiVEG

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/210370143
Early Neolithic man and riparian environment

in the Carpathian Basin
Article · January 2003
CITATIONS READS
15 69
1 author:
Pál Sümegi
University of Szeged
192 PUBLICATIONS 2,848 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Evolution, Palaeobiogeography, Palaeoecology and Palaeoclimatology of the Quaternary Mollusca

fauna in the Carpathian Basin View project
All content following this page was uploaded by Pál Sümegi on 17 December 2013.
The user has requested enhancement of the downloaded file.

Journal of Archaeological Science xxx (2011) 1e17
Contents lists available at ScienceDirect
Journal of Archaeological Science

journal homepage: http://www.elsevier.com/locate/jas
Farming and/or foraging? New environmental data to the life and economic
transformation of Late Neolithic tell communities (Tisza Culture) in SE Hungary
Sándor Gulyás*, Pál Sümegi
University of Szeged, Department of Geology and Paleontology, H-6722 Szeged, Egyetem u. 2-6, Hungary
a r t i c l e i n f o a b s t r a c t
Article history: The turn of the 6th and 5th millennia BC witnessed probably the largest economic and cultural
Received 7 December 2010 transformation of SE Europe giving rise to a new techno-complex occupying the alluvial plains of the Tisza
Received in revised form River and its tributaries in the southern parts of the Carpathian Basin. Representatives of the Tisza Culture
18 July 2011
were engaged in intensive farming complemented with foraging creating a complex system of hierarchical
Accepted 20 July 2011
multi-layered settlements (tells). The favorable endowments of the sites with a large variety of multiple
ecotones ideal for multifocal subsistence, as well as the introduction of new farming techniques ensured
Keywords:
the establishment of long-term sedentary lifeways. However, according to the archeology, a major shift in
Mollusks
Foraging
subsistence happened toward the end of the Late Neolithic marking the terminal part of the evolution of
Multiproxy paleoecological analysis the culture. Traditional crop cultivation was increasingly complemented with hunting, animal husbandry
Social & economic crisis gaining importance. Other second-line resources like fish and shellfish followed the same pattern. Finally,
Riparian environment tells were disintegrated and a new cultural group of the Copper Age emerged. The exact background of
Late Neolithic these transformations is still unknown. In order to see whether or not potential transformations in the
Carpathian Basin local riparian environment had some role in shaping human behavior, a multiproxy paleoecological
analysis was implemented on mollusk material of one of the largest tell sites of SE Hungary. Freshwater
mollusks collected by humans in themselves characterize the quality of the water body from which they
derive. They are also an excellent marker of socioeconomic response to environmental stress. According to
our findings the emergence of new settlement phases and the intensified foraging could have been
correlated with alteration of stream properties yielding successively higher floods. This was initially
beneficial creating lush pasturelands for large bodied prey infiltrating the area during the referred period
like aurochs, red-deer. But ultimately it might have reduced areas suitable for agriculture and living most
likely leading to social disruption besides other cultural, social processes.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction Gödrösök and Öcsöd-Kováshalom along the Tisza’s major southern

tributary, the Körös River (Fig. 1). The favorable endowments of the
The turn of the 6th and 5th millennia BC witnessed one of the sites with a large variety of multiple ecotones of backswamps,
most significant economic and cultural transformations of South- gallery forests, wet meadows on the floodplain and dry meadows on
eastern Europe giving rise to a new techno-complex occupying the the natural highs ideal for multifocal subsistence, as well as the
alluvial plains of the Tisza River and its tributaries in the southern introduction of new farming techniques ensured the establishment
parts of the Carpathian Basin (Korek, 1989). Representatives of the of long-term sedentary lifeways dated between 5000 and 4500-
Tisza Culture with strong southern cultural ties were engaged in 4400 cal BC years in the area of the Great Hungarian Plains. During
intensive farming complemented with foraging creating a complex the later mature phase of the evolution of the Tisza Culture new
system of hierarchical multi-layered settlements (tells) in the cultural groups emerged coevally in the Upper Tisza region just
southeastern part of the Great Hungarian Plains (Bognár-Kutzián, northeast of the site Polgár known as the Cso }szhalom Culture
1966; Kalicz and Makkay, 1977; Raczky, 1986, 1988; Horváth, 1987, (Raczky, 1988; Raczky et al., 2004; Horváth, 2003, 2005). Another
1988, 1989, 1994, 2003, 2005). The most illustrative examples are contemporary occupied the north-central part of the Great
found at Szegvár-Tu } zköves, Hódmezo } vásárhely-Gorzsa, Tápé-Leb- Hungarian Plains near Berettyóújfalu known as the Herpály Culture
}halom near the city of Szeged or at Vészto
o }-Mágor, Battonya- (Raczky, 1986, 1988; Kalicz and Makkay, 1977; Kalicz and Raczky,
1987) (Fig. 1, Table 1.) In contrast to their southern contemporaries
* Corresponding author. these northern sites were mainly characterized by single-layer
E-mail address: [email protected] (S. Gulyás). occupation and a dominance of animal husbandry and hunting of
0305-4403/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jas.2011.07.019
Please cite this article in press as: Gulyás, S., Sümegi, P., Farming and/or foraging? New environmental data to the life and economic trans-
formation of Late Neolithic tell communities (Tisza Culture) in SE Hungary, Journal of Archaeological Science (2011), doi:10.1016/
j.jas.2011.07.019
2 S. Gulyás, P. Sümegi / Journal of Archaeological Science xxx (2011) 1e17
Fig. 1. The distribution of horizontal and multi-layered (tell) settlements in the Great Hungarian Plain during the evolution of the Late Neolithic Tisza culture with contemporary
cultural groups marked (modified after Gulyás and Sümegi, 2011).
large bodied domesticates (auroch and cattle) (Fig. 1). The final climatic perturbation were postulated among potential trigger
phase of the Tisza Culture in the southern Great Hungarian Plains is factors but not yet justified. In order to see whether or not potential
represented by the Gorzsa group with a network of interacting transformations in the local riparian environment had some role in
settlements in the vicinity of Hódmezo } vásárhely and Szeged shaping human behavior, a multiproxy paleoecological analysis was
(Figs. 1e3). According to the archeology, a major shift in subsistence implemented on mollusk material of one of the largest tell sites of SE
happened toward the end of the Late Neolithic marking the terminal Hungary: Hódmezo } vásárhely-Gorzsa. Since freshwater mollusks
part of the evolution of the culture. Traditional crop cultivation was collected by humans characterize the quality of the water body from
increasingly complemented with hunting, animal husbandry of which they derive on the one hand. Furthermore, they are also an
large bodied animals (cattle, pig) (Horváth, 2005; Szabó, 2007, excellent marker of socioeconomic response to environmental
Bartosiewicz, 2005) (Fig. 2). Probably other second-line sources stress. Terrestrial mollusks on the other hand may shed light onto
like fish and shellfish showed the same pattern. Finally tells were compositional changes in the vegetation near the sites themselves.
disintegrated and a new cultural group e the so-called Early Copper
Age Tiszapolgár Culture- emerged (Gyucha et al., 2009; Parkinson
2. General characteristics of the studied Late Neolithic tell
et al., 2004, 2010). The exact background of these transformations
site of Hódmezo} vásárhely-Gorzsa
is still to be revealed. Social and economic changes as well as a major
Table 1 One of the most remarkable multi-layered (tell) settlements of

Chronology of the Tisza Culture and the study site. the Late Neolithic in the area of the SE Great Hungarian Plains is
known as Hódmezo } vásárhely-Gorzsa Czukor Homestead. The site
Settlement Culture Age (cal BC) Remarks
stages after phase itself is located in the environs of the city of Hódmezo } vásárhely
Horváth near the satellite farm center of Gorzsa at the confluence of the
(1988, 2005) Maros and Tisza Rivers about 25 km NE of the city of Szeged. The
e Tisza I. 5030e4970 Formative phase, age from nearby settlement lies on a loess-covered ancient levee of Pleistocene age
site Szegvár about 3e4 m above the floodplain along the channel of the former
D1 Tisza II. 4970e4850, Early Tisza, beginning of D1,
Kéró Brook (Fig. 3). This Pleistocene watercourse, which was acting
second half 4800 culture’s largest distribution period
D3 Tisza III. 4850e4700 Classical Tisza, development of as a floodwater drainage channel during the Holocene passed the
Herpály-Cso }szahalom groups in N referred levee on the west providing a constant hydrological link
C1 Tisza IV. 4700e4550 Late Tisza, withdrawal of between the Tisza River and the isolated backswamps, oxbow-
C2 groups to the SE Great Hungarian lakes, marshlands and other minor watercourses found on the
C3 Plain from the north down to the
Körös-Maros Interfluve and Aranka
floodplain of the three major rivers of the area: the Tisza, the Körös
valley and the Maros. The natural endowments of the vicinity of the site
B Tisza V. 4550,4500e Uncertain position and age made up of flood-free natural highs and ridges and intervening
4450 marshlands and minor watercourses was equally suited to
A Tisza VI. 4450- Final phase, beginning of
a sedentary way of life based on crop cultivation and animal
Prototiszapolgár Culture
husbandry, as well as traditional foraging activities.
j.jas.2011.07.019
S. Gulyás, P. Sümegi / Journal of Archaeological Science xxx (2011) 1e17 3
Fig. 2. The ratio of domestic and wild animals during the Neolithic of Hungary (after Bartosiewicz, 2005).
The site covers an area of 5 ha of which 3.5 represents roughly (Fig. 3e). Correlation between the individual profiles was imple-
the tell settlement itself. Out of the 2.6e3 m thick sequence about mented on the basis of relative archeostratigraphy (Horváth, 1988,
1.8e2 m corresponds to the period of the Late Neolithic Tisza 2005) and absolute chronology (Hertelendi and Horváth, 1992;
Culture overlain by layers of Early and Late Copper Age, Early and Hertelendi et al., 1998; Gulyás et al., 2010). Faunal changes
Late Bronze Age, Iron Age, Sarmatian Age as well as Medieval observable in the archeological material must be attributed to
deposits. The inner chronology of the site established by Horváth various factors, if we appreciate the human or cultural context of
(1982) corresponds to four successive settlement phases marked the material. Abundance of certain faunal elements in our fresh-
as D to A. Phase C yielded the thickest 1 m deposit of the sequence water shellfish material is the outcome of the complex interplay of
(Horváth, 1988). According to the available test-probe data, the various biotic (predation including humans) and abiotic ecological
settlement reached its greatest extent of 35,000 m2 during this components (substrate conditions, hydraulic parameters, water
phase. Houses were arranged in a circle around a large open space temperature, pH, bioproduction, sediment accumulation etc)
and burnt down several times. Based on the evaluation of succes- (Byers and Broughton, 2004; Wolverton, 2005). Thus changes
sive finds the longest periods of D and C were further subdivided observable in faunal abundances can indicate subsistence change
(D1, D3 and C1, C2, C3) (Horváth, 1994, 2005). Phase B is repre- and resource selection resulting from increased foraging pressure
sented by a couple of graves dug into the occupation layers of phase on the one hand. On the other hand, environmental change can also
C. Finally the last A phase is given by the artefacts of some scattered reduce the abundance of certain faunal elements independent of
pits and occupation levels. human predation. Thus to resolve issues of temporal change
According to the available radiocarbon dates (Hertelendi and observable in the studied archeofauna a complex analysis of several
Horváth, 1992; Hertelendi et al., 1998; Gulyás et al., 2010) and the lines of information is needed (Peacock and Seltzer, 2008; Rustioni
archeology of the site (Horváth, 1982, 2005), out of the 6 major et al., 2007) (Fig. 4). Faunal composition based on abudance was
phases of the Tisza Culture (Table 1.) 4 typical (Tisza IIeV) and 1 determined using the measure minimum number of individuals
transitional phase to the Early Copper Age (Tisza VI) has been (MNI) via taking into account the valves present in larger numbers
identified (Horváth, 2005). per taxon (Peterson, 1989; Lyman, 1994; Claasen, 1998). The taxa
present were identified to the lowest possible level using the works
3. Material and methods of Richnovszky and Pintér (1979) and Soós (1943). The ecological
characteristics of the individual species were gleaned from studies
Fine stratigraphic sampling yielded mollusk samples in adjacent of the modern fauna using the works of Sümegi (1989), Krolopp and
profiles along micro-horizons corresponding to occupation levels Sümegi (1992, 1995), Sümegi and Krolopp (1995), Evans (1972) and
j.jas.2011.07.019
} vásárhely-Gorzsa. a) wider surroundings of the site with tell sites of the same culture marked (dark dots). b) conditions preceding the
Fig. 3. Location of the study site of Hódmezo
19th century river regulations and, c, d) geomorphology of the site. e) squares of excavated profiles by Horváth (1988, 2005).
ek (1964). Several widely adopted measures are known to track

Loz Butler and Campbell, 2004; Peacock, 2000, 2002; Stiner and Munro,
faunal changes in terms of taxonomic composition: the use of 2002; Kelly, 1995; Jones, 2004). High evenness values indicate that
diversity indices (DIs) and abudance indices (AIs) is the most the pattern of exploitation is generalized (Broughton, 1994;
common (Magurran, 1988; Peat, 1974; Hammer and Harper, 2005). Mellars, 1996; Grayson and Delpech, 1998, 2002; Waguespack
As we are dealing with “artefacts” these indices tend to express the and Surovell, 2003; Cannon and Meltzer, 2004), while low even-
cultural aspect of the fauna termed as foraging efficiency (Nagaoka, ness values indicate a specialized pattern of exploitation. High
2001, 2002; Kintigh, 1984, 1989; Leonard and Jones, 1989; evenness values were generally interpreted to reflect the increasing
Magurran, 1988; Grayson and Delpech, 1998; Neeley and Clark, importance of lower rank taxa, attributed to reduced foraging
1993). Evenness measures can characterize whether exploitation efficiency reflecting the emergence of some sort of an economic
patterns are more generalized or specialized (Nagaoka, 2001, 2002; crisis (Kelly, 1995; Stiner and Munro, 2002; Jones, 2004). Shannon’s
j.jas.2011.07.019
Fig. 4. The steps of multiproxy paleoecological analysis of freshwater shells.
diversity index (H) was calculated for the sample assemblages to Prey demographic structure was determined via univariate
measure resource specialization as it exists along a continuum (Zar, statistical analysis of mussel shell parameters (shell height) (Gulyás
1996; Hammer and Harper, 2005). It is also important to recognize and Sümegi, 2004). The received descriptive statistics was used to
that similar evenness values do not mean that resource utilization identify foraging strategies (Charnov and Orians, 1973; McArthur
or targeted habitat conditions are unchanging. Sample proportions and Pianka, 1966; McArthur and Levins, 1967; Broughton, 1999;
can remain the same but the taxa corresponding to those propor- Lupo, 2007; Bird and Bliege-Bird, 2000) and its outcome on the
tions may change; i.e. a shift in habitat conditions or targeted patch harvested fauna.
may appear. In order to identify units of similar paleobiological In order to assess the fidelity of information gained, taphonomy
content reflecting similar environmental or habitat conditions, the yielding potential bias was considered (Behrensmeyer and Hill,
collective handling of ecological information of the studied taxa 1980; Behrensmeyer et al., 2000; Peacock, 2000; Claasen, 1998;
and faunal abudance values is needed. For this several widely Pike-Tay et al., 2004; Bartosiewicz, 2006). Here the indices of the
adopted explorative multivariate statistical methods are known rate of fragmentation depicting the ratio of quantifiable elements
(Digby and Kempton, 1987; Chatfield and Collins, 1980; Davies, (Kent, 1988; Nichol and Wild, 1984; Claasen, 1998; Lyman, 1984)
1998). In our work DCA (detrended correspondence analysis) was was considered along with valve representation (Koike, 1979;
implemented on a taxon/abudance data matrix per sample Peacock and Seltzer, 2008; Gulyás and Sümegi, 2004; Gulyás
following the methods of Hill and Gauch (1980); Behrensmeyer et al., 2007). To asses potential bias from sample size, Spearman
(1991) and Rustioni et al. (2007). Eigenvalues received were used rho was calculated between the gained parameters and the number
to describe information expressed by the faunal composition of the of identified elements (Simpson et al., 1960; Cochrane, 2003).
sample as whole. Visual evaluation of the scatterplot enabled the In order to retain further ecological parameters of the studied
interpretation of the ecological meaning of the received eigen- aquatic habitats, shells were subjected to geochemical analysis.
values. Thus the received axes were identified to reflect major Some of the routinely recorded trace elements in shells are excel-
ecological gradients; i.e. changes in water velocity, substrate type in lent markers or proxies of certain parameters of the studied
case of the freshwater mollusks or habitat and humidity changes in waterways. Mn2þ has been proposed as indicator of algal produc-
case of the terrestrial mollusks. However, faunal frequency or tivity (Vander Putten et al., 2000; Langlet et al., 2006, 2007; Caroll
abundance on its own may not be sufficient to capture all aspects of and Romanek, 2008; Siegele et al., 2001; Lazareth et al., 2003) and
harvest strategies deployed. It is necessary to use additional lines of as such an ideal proxy of the rate of dissolved oxygen in the water
information such as changes in prey demographic structure (body (Mutvei and Westermark, 2001; Ravera et al., 2003, 2007; Caroll
size, age) as well reflecting foraging pressure on the fauna and its and Romanek, 2008). Fe2þ, Zn2þ are soluble in relation to the
background causes-harvest strategy employed-(McAulife, 1984; acidity of the medium and as such may record fluctuations in the
Broughton, 1994), fluctuations in habitat conditions due to over- pH (Stumm and Morgan, 1996; Mutvei and Westermark, 2001).
harvest or other natural causes (Erman and Erman, 1984; Murphy According to Lee and Wilson (1969) and Hill (1975) rivers with
and Davy Bowker, 2005; Weber, 2005; Bauer and Wachtler, 2001; a larger discharge tend to be more dilute and as such relatively
Peacock, 2000; Reitz and Wing, 1999), and/or climatic parameters deprived of dissolved elements compared to rivers with a lower
(Klein and Cruz-Uribe, 1984; Nagaoka, 2002, 2005; Cannon, 2000; discharge. Consequently, a negative shift in general concentration
Jerardino and Navarro, 2008; Yamazaki and Oda, 2009). values of trace elements may also reflect a transformation in
j.jas.2011.07.019
Table 2
Composition of the identified archeomalacofauna.
Paleoecological group Taxa NISP MNI Proportional Abudance by

abudance (%) ecological group (%)
Freshwater fauna
Moving-water (rheophyllic) taxa Unio crassus Philipsson, 1788 597 353 6.55 14.27
Viviparus acerosus Bourgniat 1862 378 378 7.01
Viviparus contectus Millet 1813 38 38 0.71
Stagnant water or slowly Unio pictorum Linnaeus 1758 7151 3692 68.51 85.73
moving-water taxa Unio tumidus Philipsson, 1788 1590 878 16.29
Anodonta cygnea Linnaeus 1758 346 49 0.91
Lymnaea stagnalis Linnaeus 1758 1 1 0.02
Terrestrial fauna
hygrophylous, shade-loving taxa Helix pomatia Linnaeus 1758 56 56 23.93 23.93
xerophylous forest-steppe taxa Cepea vindobonensis Férussac 1821 178 178 76.07 76.07
Freshwater fauna total 10101 5389 95.84
Terrestrial fauna total 234 234 4.16
TOTAL bivalves 9685 4973 88.44
TOTAL gastropods 650 650 11.56
TOTAL FAUNA 10335 5623 100.00
discharge values of a fluvial system. Shells of the same mussel taxon Statistical analysis including KruskalleWallis test and correla-
(Unio pictorum (Linnaeus)) were chosen for analysis in order to tion analysis were performed following the methods of Sokal and
avoid discrepancies deriving from taxon-specific incorporation of Rohlf (1995) to explore and clarify the relationship among the
the studied elements. Samples were acid digested on hot plate gained variables using programs SPSS 11 and PAST 1.8 (Hammer
following a modified procedure of Szöo } r et al. (1992) and Peacock and Harper, 2005). Statistical significance was assumed when
and Seltzer (2008). Mn, Fe and Zn concentrations were recorded p < 0.05 and 0.01, respectively. The gained proxies were also
using the flame AAS technique in a PerkineElmer AAS type 100 at depicted graphically using the program Golden Software Grapher
the Department of Geology and Paleontology, University of Szeged, 7.0 to study shifts in the values in relation to archeostratigraphy.
Hungary. Calibrations and measurements were performed in
accordance with international standards following Pradyot (2004). 4. Results
Samples were measured 3 times to ensure reproducibility of the
results is better than 0.1%. Shell concentrations of elements are 4.1. Taxonomic composition of the studied fauna
reported to dry weight to enable comparison with other published
data on modern and subfossil unionid taxa. The studied archeofauna yielded 10,335 specimens (NISP) of 7
Finally additional social and cultural aspects of the fauna freshwater (4 bivalve, 3 gastropod) and 2 terrestrial mollusk taxa
(shellfish use in sacrificial and everyday activities) was also incor- (Table 2). Freshwater elements are clearly dominant (95.84%), while
porated into our analysis. the proportion of terrestrial gastropods is relatively negligible
Fig. 5. Results of multivariate ordination (DCA) for freshwater taxa.
j.jas.2011.07.019
(4.16%) due to the nature of the deposit. In the freshwater fauna the paleotemperatures based on malacothermometry (Sümegi, 1989)
dominant ecotypes (85.73%) are those of so-called ditch elements were somewhat above the modern value (22e23 C ). This is
preferring standing or temporarily moving-water conditions and congruent with the idea of Holocene climatic optimum during the
a muddy substrate (painter’s mussel ¼ Unio pictorum (Linnaeus Atlantic (Denton and Karlen, 1973; Davis et al., 2003; Kalis et al.,
1755), swollen river mussel ¼ Unio tumidus (Philipsson 1788)). Taxa 2003; Voigt et al., 2008; Wolverton, 2005). The fauna was domi-
showing preference for moving-water conditions (rheophylic) (river nated by the xerophylous, forest-steppe- dweller Cepea vindobo-
mussel ¼ Unio crassus (Philipsson 1788), river snails ¼ Viviparus nensis (Férussac 1821) (76%). Conversely, the proportion of the
acerosus (Bourguinat 1862) and Viviparus contectus (Millet 1813)) are hygrophilous, shade-loving Burgundy snail (Helix pomatia (Lin-
relatively subordinate (14.27%). Freshwater snails populating minor naeus)) is subordinate (w24%). This data is in line with that shown
ponds of the floodplain (Lymnaea stagnalis (Linnaeus 1758), are also by the freshwater fauna; i.e. the presence of wide-open grasslands
subordinate in the fauna similarly to pond mussel (Anodonta cygnea on the flood-free natural high hosting the settlement (Fig. 3).
(Linnaeus 1755) (w1%). This fact plus the univocal dominance of If we examine the composition of the fauna by sample we can
painter mussel in our material is a clear proof of the artificial nature clearly see that there is only a slight scatter between samples
of the deposit. Nevertheless, the composition of the mollusk fauna along Axis 1 of the DCA scatterplot (Fig. 5). Bubble size corre-
ideally describes a habitat in accordance with the location of the site. sponds to the magnitude of the eigenvalue along Axis 2. Circles
Namely, a protected floodplain away from the main channel (Fig. 3) with dotted line depict scatter in sample values along major
with relatively calm conditions, and a muddy substrate as well as ordination axes. Dotted lines mark inferred ecotype boundaries
dense aquatic vegetation and higher bioproduction rates. The along major gradient axes. Rheophylic elements like river snails
presence of moving-water elements points to recurrent minor are not sharply separated from typical ditch elements of painter’s
floods carrying coarser material onto the floodplain as well. Two mussel and swollen river mussel along Axis 1 apart from thick
thermophylous terrestrial snail taxa were identified in the material shelled river mussel, which is characterized by a highly positive
with about 234 specimens (Table 2). The inferred mean July eigenvalue (w2). This somewhat complicates the interpretation
Fig. 6. Fluctuations of proxy values describing major environmental parameters throughout the life of the study site (broken lines depict major trend in data).
j.jas.2011.07.019
Table 3
Correlations between received variables for habitat quality, foraging strategy, taphonomy, archeostratigraphy.
Correlation
Decoration (PCA) Temper (PCA) Imported ceramics (PCA) Harvested mussels Harvested riverine snails Water velocity
Spearman’s Decoration Correlation 1 0.5091 0.743662595748901 0.412253201007843 0.372900277376175 0.346557408571243
rho coefficient
(PCA) Sig. (2-tailed) 0.0110 0.0000311744406644721 0.0453013330698013 0.072704866528511 0.0971125289797783
N 24 24 24 24 24 24
Temper Correlation 0.5091 1 0.672626554965972 0.25179386138916 0.106637716293335 0.281739115715027
coefficient
(PCA) Sig. (2-tailed) 0.0110 0.000317014986649156 0.235257282853127 0.619932770729064 0.182282775640488
N 24 24 24 24 24 24
Imported Correlation 0.743662595748901 0.672626554965972 1 0.559844672679901 0.469998359680176 0.576287925243377
coefficient
Ceramics Sig. (2-tailed) 0.0000311744406644721 0.000317014986649156 0.00444378657266498 0.0204772986471653 0.00320466374978423
(PCA) N 24 24 24 24 24 24
Mussel Correlation 0.412253201007843 0.25179386138916 0.559844672679901 1 0.70069682598114 0.206566646695137
harvest coefficient
Intensity Sig. (2-tailed) 0.0453013330698013 0.235257282853127 0.00444378657266498 0.000137150083901361 0.332823514938354
N 24 24 24 24 24 24
Harvested Correlation 0.372900277376175 0.106637716293335 0.469998359680176 0.70069682598114 1 0.060500580817461
coefficient
Riverine Sig. (2-tailed) 0.072704866528511 0.619932770729064 0.0204772986471653 0.000137150083901361 0.778844058513641
gastropods
N 24 24 24 24 24 24
Water Correlation 0.346557408571243 0.281739115715027 0.576287925243377 0.206566646695137 0.060500580817461 1
velocity coefficient
Sig. (2-tailed) 0.0971125289797783 0.182282775640488 0.00320466374978423 0.332823514938354 0.778844058513641
N 24 24 24 24 24 24
Shannon H Correlation 0.0623367428779602 0.433043479919434 0.0802095159888268 0.0474016107618809 0.144504979252815 0.0243478268384933
coefficient
Sig. (2-tailed) 0.772301971912384 0.0345363691449165 0.709469139575958 0.825913190841674 0.500511765480041 0.910087406635284
N 24 24 24 24 24 24
Average shell Correlation 0.0418484434485435 0.141739130020142 0.281605154275894 0.488801926374435 0.249401673674583 0.291304349899292
coefficient
Height Sig. (2-tailed) 0.846054673194885 0.508832454681396 0.182499662041664 0.0153595302253962 0.239884734153748 0.167254343628883
N 24 24 24 24 24 24
Fragmentation Correlation 0.0575541220605373 0.221352472901344 0.010682393796742 0.247716397047043 0.00718328543007374 0.124374866485596
coefficient
Sig. (2-tailed) 0.789372324943542 0.298582792282104 0.960488677024841 0.243180274963379 0.973425447940826 0.562563121318817
N 24 24 24 24 24 24
Right/left Correlation 0.0257810801267624 0.245805889368057 0.102468870580196 0.235833168029785 0.0796248465776443 0.10764903575182
valve coefficient
Ratio Sig. (2-tailed) 0.904817402362823 0.24695186316967 0.633752286434173 0.267255455255508 0.711498260498046 0.616598844528198
N 24 24 24 24 24 24
Size range Correlation 0.182036712765694 0.15090237557888 0.212121829390526 0.635928690433502 0.366565227508545 0.0621874332427979
of dominant coefficient
mussel Sig. (2-tailed) 0.39458429813385 0.481526523828506 0.31969279050827 0.000837902654893696 0.0780950710177422 0.772833406925201
N 24 24 24 24 24 24
Mn2þ Correlation 0.140533775091171 0.0181818176060915 0.0325203686952591 0.252029895782471 0.217744663357735 0.224675327539444
coefficient
Sig. (2-tailed) 0.54345291852951 0.937650084495544 0.88870894908905 0.270402312278748 0.343040108680725 0.327510237693787
N 21 21 21 21 21 21
Fe2þ Correlation 0.14248563349247 0.0103896101936698 0.243252351880074 0.256576836109161 0.285342991352081 0.068831168115139
coefficient
Sig. (2-tailed) 0.537811577320098 0.964349031448364 0.2879958152771 0.261560827493668 0.209913671016693 0.766877889633178
N 21 21 21 21 21 21
*Statistically significant correlation p ¼ 0.05 (2-tailed).

**Statistically significant correlation p ¼ 0.01 (2-tailed).
of the ecological meaning of Axis 1. However, thick shelled river a more important role than considerable shifts in substrate condi-
mussel is the only taxon with a preference for a sandy substrate. tions during the history of the tell complex.
The remaining rheophylic river snail and the ditch species like For the entire profile the inferred two components (DCA1 &
painter’s mussel, swollen river mussel and pond mussel with DCA2) tend to display a gradually increasing trend regarding their
a muddy substrate preference are characterized by eigenvalues intensities not absolute values (Fig. 6). For substrate conditions
below 0.5 and zero. The typical pond dweller pond mussel has an the first settlement phase (D) is in sharp contrast with the
extremely negative eigenvalue (2.7) in similar magnitude to the successive phases of C and B with a prevalence of muddy
thick shelled river mussel (w2). So this gradient (Axis 1) giving substrate and only minor fluctuations between the two subphases
78.25% of the total variance must represent habitat/substrate type D1 and D3. Conversely, in phase C, apart from some minor fluc-
along which elements with a preference for muddy substrate are tuations, generally coarse-grained substrate conditions develop in
clearly separated from those of sandy substrate. comparison with the preceding phase probably as a result of
Conversely, separation of taxa according to their ecological heightened floods carrying more suspended load onto the distal
characteristics is more straightforward along Axis 2 giving 22.75% parts of the floodplain located at a larger distance from the active
of total variance. Here ditch elements are characterized by eigen- channel hosting the settlement (Fig. 3). A similar pattern was
values close to zero (w 0.5 and 0.25). Rheophylic elements on the observable in the parameter expressing fluctuations in the
other hand have high positive eigenvalues (w1 and >1). So this hydraulic conditions of the floodplain in the vicinity of the site.
gradient must correspond to variation in preference shown for The beginning of phase D1 is characterized by relatively higher
hydraulic conditions of the habitat. Samples tend to display a larger water velocities, which is gradually reduced by the middle part of
scatter along Axis 2 implying that fluctuations in hydraulic phase D3. From the second half of phase D3 there is a marked
parameters attributable to recurrent floods must have played increase in water velocity remaining relatively high during almost
j.jas.2011.07.019
Correlation Fe2þ
Shannon H Average shell height Fragmentation Right/Left valve ratio Size range of dominant mussel Mn2þ
0.0623367428779602 0.0418484434485435 0.0575541220605373 0.0257810801267624 0.182036712765694 0.140533775091171 0.1424
0.772301971912384 0.846054673194885 0.789372324943542 0.904817402362823 0.39458429813385 0.54345291852951 0.537
24 24 24 24 24 21 21
0.433043479919434 0.141739130020142 0.221352472901344 0.245805889368057 0.15090237557888 0.0181818176060915 0.0103
0.0345363691449165 0.508832454681396 0.298582792282104 0.24695186316967 0.481526523828506 0.937650084495544 0.964
24 24 24 24 24 21 21
0.0802095159888268 0.281605154275894 0.010682393796742 0.102468870580196 0.212121829390526 0.0325203686952591 0.2432
0.709469139575958 0.182499662041664 0.960488677024841 0.633752286434173 0.31969279050827 0.88870894908905 0.2879
24 24 24 24 24 21 21
0.0474016107618809 0.488801926374435 0.247716397047043 0.235833168029785 0.635928690433502 0.252029895782471 0.256576836109161
0.825913190841674 0.0153595302253962 0.243180274963379 0.267255455255508 0.000837902654893696 0.270402312278748 0.261560827493668
24 24 24 24 24 21 21
0.144504979252815 0.249401673674583 0.00718328543007374 0.0796248465776443 0.366565227508545 0.217744663357735 0.285342991352081
0.500511765480041 0.239884734153748 0.973425447940826 0.711498260498046 0.0780950710177422 0.343040108680725 0.209913671016693
24 24 24 24 24 21 21
0.0243478268384933 0.291304349899292 0.124374866485596 0.10764903575182 0.0621874332427979 0.224675327539444 0.068831168115139
0.910087406635284 0.167254343628883 0.562563121318817 0.616598844528198 0.772833406925201 0.327510237693787 0.766877889633178
24 24 24 24 24 21 21
1 0.0391304343938828 0.0948032215237617 0.459796488285065 0.523157238960266 0.112987011671066 0.315584421157837
0.855947911739349 0.659473419189453 0.0237850397825241 0.00870951171964407 0.625809609889984 0.16345351934433
24 24 24 24 24 21 21
0.0391304343938828 1 0.423135459423065 0.164306417107582 0.295716464519501 0.0571428574621677 0.28311687707901
0.855947911739349 0.0393800139427185 0.442962676286697 0.160622596740723 0.805660963058471 0.213647812604904
24 24 24 24 24 21 21
0.0948032215237617 0.423135459423065 1 0.02048827894032 0.245759025216103 0.185774609446526 0.0461188703775406
0.659473419189453 0.0393800139427185 0.924296498298645 0.247044876217842 0.420097708702087 0.842648386955261
24 24 24 24 24 21 21
0.459796488285065 0.164306417107582 0.02048827894032 1 0.13251993060112 0.0501631312072277 0.119870342314243
0.0237850397825241 0.442962676286697 0.924296498298645 0.537045240402221 0.829037904739379 0.604771375656127
24 24 24 24 24 21 21
0.523157238960266 0.295716464519501 0.245759025216103 0.13251993060112 1 0.420915901660919 0.119519330561161
0.00870951171964407 0.160622596740723 0.247044876217842 0.537045240402221 0.057415671646595 0.605837166309356
24 24 24 24 24 21 21
0.112987011671066 0.0571428574621677 0.185774609446526 0.0501631312072277 0.420915901660919 1 0.0525127053260803
0.625809609889984 0.805660963058471 0.420097708702087 0.829037904739379 0.057415671646595 0.816469132900238
21 21 21 21 21 21 21
0.315584421157837 0.28311687707901 0.0461188703775406 0.119870342314243 0.119519330561161 0.0525127053260803 1
0.16345351934433 0.213647812604904 0.842648386955261 0.604771375656127 0.605837166309356 0.816469132900238
21 21 21 21 21 21 21
the entire period of phase C again apart from some relatively freshwater fish was studied unfortunately only in a single sub-
negligible fluctuations. profile (profile no.18) of the excavated area. Nevertheless, in this
This again may point to the development of higher floods from profile a relatively strong correlation was observed between
the second part of D3, which gradually die off as we reach the foraging intensities of fish and shellfish (Spearman rho ¼ 0.415)
terminal part of phase C (Fig. 6). Fluctuations in relative humidity (Appendix:Table 3). So there is every reason to believe that this
on the natural high inferred from the ratio of xerophylous terres- pattern holds true for the rest of the studied subprofiles from the
trial gastropods in our material seem to be congruent with the excavation site as well. Not to mention that the correlation of both
trend seen in the parameters discussing the most important parameters is likewise strong and significant with those expressing
features of the studied fluvial system, especially that of water hydraulic conditions on the floodplain (Spearman rho ¼ 0.967
velocity. The first phase (D1) was characterized by generally lower p ¼ 0.01) (Appendix:Table 3). It is also important to note that the
humidity compared to the successive phases (Fig. 6), followed by majority of fish were given by such taxa as carp (70%), catfish (15%)
a sharp increase by phases D3 and C and a generally decreasing and pike (7%), all of which like to dwell in ponds and lakes with lush
trend toward the terminal part of phase C (C3) and B. So fluctua- vegetation and a muddy substrate. The majority of the specimens
tions in humidity values on the natural high are congruent with were extraordinary in size, especially catfish based on biometry of
changes in flood intensity. Thus heightened floods onto the flood- headbones, yielding substantial amount of meat to the community
plain must have resulted in higher humidity values even in the (Horváth, 2005). The number of harvested river snails and that of
flood-free drier areas, which is an excellent expression of the hygrophilous shade-loving terrestrial snails displays a strong
greater intensity of fluvial activities during the referred settlement statistically significant correlation as well (Spearman rho ¼ 0.579
phases after subphase D1. p ¼ 0.01) (Appendix: Table 3). So the increased harvest of moving-
Harvest intensity seems to have followed the trend seen in water river snails is linked to periods of higher humidity marking
water velocity, substrate type and humidity values as well (Fig. 6). the contemporarity of heightened floods and intensified foraging
Altogether 9 harvest peaks was observed (Neolithic micro-levels on the floodplain. Heightened floods thus must have carried more
24, 22, 19, 17, 16, 13, 10, 8, 5) with a strong correlation between nutrition onto the wet meadows, with sufficient water for primary
the harvest of dominantly ditch mussel elements and rheophylic production of plants as well, creating lush grasslands and a some-
river snails (Spearman rho ¼ 0.701 p ¼ 0.01) (Appendix:Table 3). what more closed forest-steppe vegetation in the vicinity of the
During phase D foraging seems to have been subordinate both site. This must have been ideal for large bodied ruminants like
regarding freshwater mussels and river snails (Figs. 6 and 7). cattle, auroch, deer as well as wild boar. According to the pattern
Conversely, phase C was characterized by an intensive exploitation observable, second-line aquatic resources seemed to have played
of aquatic habitats displaying 4 peak harvest periods, each of which a gradually increasing importance in the economy of the local
tends to fall in the periods of the individual subphases (C1, C2. C3). community toward the second half of its life and as such might
The highest peak was recorded during subphase C3 after which have been able to provide enough meat together with that gained
a sharp decrease is observable. Foraging intensity regarding from hunting to be a staple in subsistence. The heightened reliance
j.jas.2011.07.019
Table 4
Results of KruskaleWallis test for significant diffence between mean size of the dominantly harvested freshwater mussel.
Kruskal-Wallis test Unio pictorum size
H¼71.39
Hc¼71.39
p(same)¼4.157E-06
Sarmatian Bronze N1 N2 N3 N5 N6 N7 N8 N9 N10

Sarmatian 0 0.03104a 0.03576 0.06437 0.2424 0.1836 0.1121 0.3616 0.02902 0.1963 0.01671
Bronze 1 0 0.8363 0.5578 0.1678 0.3177 0.6745 0.1011 0.5962 0.4045 0.645
N1 1 1 0 0.518 0.1261 0.2799 0.7252 0.05 0.2117 0.3909 0.3309
N2 1 1 1 0 0.3264 0.678 0.9817 0.2468 0.199 0.8349 0.1934
N3 1 1 1 1 0 0.6713 0.5684 0.6822 0.0755 0.7963 0.05329
N4 1 1 1 1 1 0.4625 0.7351 0.2301 0.8762 0.5465 0.8724
N5 1 1 1 1 1 0 0.6642 0.5003 0.09803 0.9232 0.09328
N6 1 1 1 1 1 1 0 0.282 0.4594 0.6935 0.3827
N7 1 1 1 1 1 1 1 0 0.01709a 0.432 0.01632
N8 1 1 1 1 1 1 1 1 0 0.1345 0.9961
N9 1 1 1 1 1 1 1 1 1 0 0.146
N10 1 1 1 1 1 1 1 1 1 1 0
N11 1 1 1 1 1 1 1 1 1 1 1
N12 0.7086 1 1 1 0.8993 0.7868 1 0.1218 1 1 1
N13 0.6636 1 0.03079 0.4871 0.2342 0.1874 1 0.01209 1 0.3738 1
N14 1 1 1 1 1 1 1 1 1 1 1
N15 0.1518 1 0.04197 0.194 0.1123 0.08772 1 0.006156 1 0.2863 1
N16 1 1 1 1 1 1 1 1 1 1 1
N17 1 1 1 1 1 1 1 1 1 1 1
N18 1 1 1 1 1 1 1 1 1 1 1
N19 1 1 1 1 1 1 1 1 1 1 1
N20 0.7801 1 1 1 1 1 1 1 1 1 1
N21 1 1 1 1 1 1 1 1 1 1 1
N22 1 1 1 1 1 1 1 1 1 1 1
N23 1 1 1 1 1 1 1 1 1 1 1
N24 1 1 1 1 1 1 1 1 1 1 1
N25 1 1 1 1 1 1 1 1 1 1 1
a
Denote statistically significant difference at the level of 0.05.
on second-line aquatic and terrestrial resources developed paral- 4.2. The size composition of the studied fauna
lely with the observed transformation of hydraulic parameters of
the floodplain. It is also important to examine whether or not the previously
In the next part we shall examine proxies describing some mentioned shift in foraging strategy regarding harvested taxa is
aspects of foraging strategy employed (harvest evenness) (Figs. 6 equally observable in the size composition of the fauna; i.e. foraging
and 7). During subphases D1 and the first half of D3, character- was size-specific or not and strategy employed had influence on the
ized by low foraging intensities, a relatively specialized foraging was foraged natural populations (Fig. 7). Peak harvest periods are
employed targeting only a low number of available taxa (Shannon H characterized by high average size values (Spearman rho ¼ 0.489
low) (Fig. 7). However, from the second half of subphase D3, when p ¼ 0.05) (Appendix:Table 3), while there is a general decrease in
foraging gradually gains more importance toward the terminal part modal size after each harvest peak, which is statistically significant
of phase C, we can see a dominantly generalist type of foraging only after the largest peaks in subphases C3 and C2 (Appendix:
compared to the previos phase (Shannon H values close to 1) tar- Table 4). This implies an increased pressure on the natural pop-
geting all available shellfish taxa on the floodplain, including those ulation from foraging; a clear sign of overharvest after which the
as well which originally had a lower currency value in line with OFT shellfish population had time to recover. Other statistical parame-
prey-choice models. This is also a clear expression of the heightened ters describing size distribution of the fauna (skewness and
role of aquatic resources in subsistence during the referred period. kurtosis) also point to a size-specific foraging strategy in times of
The referred foraging proxies tend to have a moderately strong intensified harvest. Positive skewness values generally refer to the
positive correlation with abundance values of shade-loving hygro- dominance of larger size classes in the material; i.e. foraging for
phylous terrestrial snails (Spearman rho ¼ 0.308), and a likewise dominantly larger specimens due to higher meat yields. Positive
moderate but negative correlation with abundances of xerophylous kurtosis values equally refer to a peaked distribution and as such
steppe elements (Appendix:Table 3). So periods with increased a size-specific harvest implying the presence of a small number of
humidity were characterized by the deployment of generalist, while size classes concentrated close to the mean. Conversely, negative
those of lower humidity, a specialist type of foraging activity. As it skewness indicates the dominance of smaller size classes. Negative
has been mentioned before higher humidity periods could have kurtosis on the other hand refers to a flattened distribution due to
developed on the floodplain as a result of increased evaporation the presence of several size classes in “relatively equal” proportions
attributable to higher temperatures and larger constant water (Bailey and Milner, 2000).
coverage on the floodplain, which might have developed only Peaked distributions are linked to the major harvest peaks,
during larger floods (Fig. 3) as mentioned earlier in line with while intervening periods are characterized by relatively flatter
patterns seen from other proxies. distributions (Fig. 7). The larger trend seen in this parameter is also
j.jas.2011.07.019
Kruskal-Wallis test Unio pictorum size
H¼71.39
Mann-Whitney pairwise comparisons
Bonferroni corrected/uncorrected
N11 N12 N13 N14 N15 N16 N17 N18 N19 N21 N22 N23 N24 N25
0.04761 0.002019 0.00189 0.302 0.0004326 0.04088 0.01719 0.3502 0.03374 0.6504 0.01086 0.2002 0.07515 0.08529
0.3975 0.0323 0.02156 0.2498 0.009351 0.6236 0.3753 0.4088 0.6851 0.394 0.4479 0.6801 0.4174 0.9003
0.2076 0.002867 0.00008773 0.1552 0.0001196 0.2981 0.1449 0.4695 0.4817 0.3915 0.2965 0.7732 0.3714 0.9931
0.174 0.005473 0.001388 0.4759 0.0005526 0.2283 0.1292 0.6608 0.3402 0.5743 0.23 0.8729 0.909 0.7981
0.08687 0.002562 0.0006672 0.9928 0.0003201 0.1057 0.03655 0.8597 0.1088 0.8764 0.04651 0.4731 0.3768 0.3798
0.6319 0.2332 0.2481 0.413 0.1193 0.8988 0.6862 0.5334 0.8309 0.4278 0.7159 0.637 0.5411 0.9278
0.1076 0.002242 0.0005338 0.7768 0.0002499 0.1375 0.05251 0.8025 0.2061 0.6931 0.07004 0.9068 0.7901 0.6266
0.3138 0.03791 0.0313 0.4398 0.01248 0.4826 0.2621 0.5519 0.528 0.6543 0.33 0.8899 0.8861 0.7417
0.02603 0.000347 0.00003443 0.7771 0.00001754 0.02635 0.007525 0.9837 0.06678 1 0.009001 0.4159 0.2371 0.2681
0.5045 0.03317 0.01127 0.04853 0.007395 0.9907 0.5694 0.4546 0.8821 0.3696 0.7388 0.3869 0.1131 0.8014
0.1177 0.004617 0.001065 0.5682 0.0008157 0.1817 0.06357 0.8166 0.264 0.7065 0.1347 0.9657 0.8977 0.5986
0.6754 0.05364 0.02866 0.0464 0.007967 0.9046 0.6971 0.3697 0.9404 0.3467 0.9279 0.449 0.1389 0.822
0 0.2606 0.3295 0.04662 0.1973 0.5404 0.8728 0.4615 0.676 0.3579 0.7646 0.299 0.1372 0.8021
1 0 0.6687 0.0007931 0.8054 0.03812 0.09981 0.2228 0.1424 0.1577 0.1243 0.04595 0.002251 0.3221
1 1 0 0.0001995a 0.4364 0.01676 0.09591 0.1904 0.1548 0.1267 0.08326 0.03622 0.0002342 0.3157
1 0.2784 0.07004 0 0.0001239a 0.05192 0.02069 1 0.1517 0.8571 0.03259 0.6675 0.4843 0.3942
1 1 1 0.04348 0 0.01341a 0.04129 0.1048 0.0698 0.08465 0.03996 0.02101 0.0001501 0.1507
1 1 1 1 1 0 0.5639 0.4701 0.9639 0.3985 0.7484 0.4056 0.1659 0.9285
1 1 1 1 1 1 0 0.3864 0.7866 0.2752 0.9263 0.2789 0.06224 0.7709
1 1 1 1 1 1 1 0 0.51 1 0.3789 0.3929 0.543 0.4941
1 1 1 1 1 1 1 1 0 0.339 0.7856 0.4464 0.2369 0.7052
1 1 1 1 1 1 1 1 1 0.1437 0.3307 0.109 0.04719 0.05132
1 1 1 1 1 1 1 1 1 0 0.3372 0.5023 0.5488 0.6261
1 1 1 1 1 1 1 1 1 1 0 0.4092 0.106 0.887
1 1 1 1 1 1 1 1 1 1 1 0 0.8283 0.9759
1 0.7902 0.08221 1 0.05268 1 1 1 1 1 1 1 0 0.6725
1 1 1 1 1 1 1 1 1 1 1 1 1 0
congruent with observed fluctuations in harvest intensity. Apart biased, taphonomical characteristics of the samples were also
from peak harvest periods, subphases D1 and D3 are generally assessed (Fig. 7). The proportions of fragments not entering
characterized by a flattening trend toward D3 in kurtosis. As we statistical analysis is generally below 50% and even lower (40% or
move to phase C and foraging turns more and more intensive, 20%) during major harvest peaks. So our data does not seem to be
kurtosis gets positive with peak values observed during the largest significantly biased as a result of taphonomic processes as shown
harvest peak of the entire profile. The same trend develops for by a strong significant negative correlation between our fragmen-
skewness, although fluctuations are not as great here as in case of tation proxy and the modal size of the dominantly harvested
kurtosis. But here too peak harvests are right skewed, while shellfish (Spearman rho ¼ 0.423 p ¼ 0.05) as well (Appendix:-
intervening periods are somewhat left-skewed compared to the Table 3). The relatively equal proportion of valves in the samples
preceding phase. also seems to corroborate this statement.
Nevertheless the major trend is just the same as for kurtosis. The number of modified shells was very low both for mussels
These interpretations are also seen in the values of general size and snails, implying that the majority of shellfish must have been
range of the studied material and the strong correlation between used for dietary purposes dominantly, either by humans or as
this parameter and the former two describing size distributions of fodder. This must be attributed to the relatively high protein
the fauna (Spearman rho ¼ 0.482 p ¼ 0.05 and Spearman content of the meat and the presence of essential trace elements
rho ¼ 0.750) (Fig. 7) (Appendix:Table 3). Peak harvests are char- and vitamins not available from other food sources.
acterized by higher range values again pointing to a relatively
generalized foraging strategy congruently with the pattern seen in 4.3. Results of geochemical analysis of freshwater shells
evenness of taxa (Spearman rho ¼ 0.636 p ¼ 0.05), yet the selection
of dominantly larger size classes yielding more meat (Appen- Shells subjected to geochemical analysis derive from a single
dix:Table 3). This strategy again has significantly brought down the subprofile of the site profile No.18 (Fig. 3e). The reason behind this
mean size of the population as mentioned, as a result of over- choice was the availability of information on other aquatic second-
harvesting, which is seen in a negative shift of skewness and line resources (fish) from this subprofile alone (Fig. 7). Concentra-
kurtosis values. Finally it’s also important to mention that fluctu- tions of Mn2þ ranged between 125 and 471 ppm in the samples.
ations in modal size are moderately correlated with that of habitat Difference between the individual micro-horizons was statistically
type and water velocity proxies (Spearman rho ¼ 0.322) (Appen- significant (KruskaleWallis p ¼ 0.0055321). Within the entire
dix:Table 3). This parameter is also moderately correlated with our profile there are major fluctuations in the concentration of this
humidity proxy values (Spearman rho ¼ 0.253) (Appendix:T- element between values around 150 ppm and 400 ppm implying
able 3). So in general larger specimens were usually foraged for pronounced shifts in the hydraulic parameters of the studied water
in periods of higher floods resulting in higher stream velocities and body with recurrent phases of increased and decreased dissolved
humidity on the floodplain. oxygen content. The lowest value is reached in the topmost part of
In order to test whether or not our size compositional param- the profile assigned to phase B (125.202 ppm). In general we can say
eters and the interpretations deduced from them are potentially that apart from these major shifts there is a gradual upward decrease
j.jas.2011.07.019
Fig. 7. Fluctuations of proxy values describing foraging strategy throughout the life of study site.
in the concentration of this element. The most pronounced peaks of statistically significant (KruskaleWallis p ¼ 0.00165). The major
low manganese concentrations (w130e150 ppm) are connected to trend observable here is the same as for manganese. There is
the peak harvest periods and subphases D3, C1, C2, C3 and B. Fluc- a gradual upward decrease in iron concentration of the studied
tuations in manganese concentrations tend to display a moderately shells, where minima corresponds roughly to the individual
strong negative correlation with the number of harvested mussels subphases D3, C1, C2, and C3. Fluctuations of iron concentrations had
(Spearman rho ¼ 0.252), that of harvested river snails (Spearman a moderately strong negative correlation with the number of har-
rho ¼ 0.218), that of hygrophilous snails (Spearman rho ¼ 0.173), vested mussels and river snails, similarly to manganese (Spearman
stream velocity (DCA2) and substrate type (DCA1) (Spearman rho ¼ 0.257 & 0.285) (Appendix:Table 3). All in all concentration
rho ¼ 0.225 illetve 0.315), and the size range of harvested changes in manganese and iron point to a frequent shift in water
mussels (Spearman rho ¼ 0.421). Conversely, this parameter has quality throughout the entire history of the tell. The general decrease
a positive correlation with the number of xerophylous snails in the concentration of both studied elements in accordance with
(Spearman rho ¼ 0.104) (Appendix:Table 3). This means that parallel Lee and Wilson (1969) and Hill (1975) refer to increased stream
with a decrease in manganese concentration of the studied shells discharge toward the terminal part of the evolution of the culture.
indicating elevated levels of dissolved oxygen in the water, the
number of harvested mussels and river snails, as well as the targeted 4.4. Comparison of findings with archeology proxies of the site
size range of the foraged mussels increased along with humidity and
stream velocity. So again periods of intensive foraging were char- Finally a comparison of our proxies was made with quantitative
acterized by an increase in stream velocity, a slight shift in substrate information on the archeology of the site on the basis of raw data by
type and stream velocity, which is a clear mark of more pronounced published by Horváth (2005) including such parameters as vessel
floods onto the floodplain. Concentrations of iron display a similarly decoration, type of temper used for pottery making and type of
large fluctuation in the profile as that of manganese with values pottery imported to the site. In order to elucidate the most
ranging between 36 and 175 ppm. It also points to a frequent fluc- important components, raw data of Horváth (2005) was subjected
tuation of the pH conditions near the site during the history of the to multivariate ordination (PCA) and the received eigenvalues were
tell complex. Difference between the individual micro-horizons was taken to accurately describe the interpreted archeologies of vessel
j.jas.2011.07.019
Fig. 8. Fluctuations of proxy values describing archeostratigraphy of the study site (raw data from Horváth (2005)).
decoration, temper, imported ceramics (Fig. 8). There is a strong period spanning about 100e150 years marked the greatest
and statistically significant correlation between all three referred expansion of the representatives of the Tisza culture in the Great
archeological parameters (Spearman rho ¼ 0.744 p ¼ 0.01 Hungarian Plains (Fig. 1, Table 1), most likely as a result of favorable
& 0.509 p ¼ 0.05) (Appendix:Table 3). So the observed pattern warm climatic conditions and relative hydrological stability. This is
unambiguously reflect a similar social and economic evolution of seen in our local data at this site and another nearby contemporary
the tell complex. In addition all three parameters show a strong or tell site (Szegvár-Tu} zköves (Figs. 1e3)) as well (Gulyás and Sümegi,
moderate correlation with foraging intensities (Spearman 2011). As a result of the warmer climate, probably with somewhat
rho ¼ 0.412 p ¼ 0.05 & Spearman rho ¼ 0.373). Furthermore, the reduced humidity compared to previous phases, and increasing
parameter used for establishing the archeostratigraphy; i.e vessel human activities, vegetation on the loess-covered lag surface must
decoration is moderately correlated with fluctuations in stream have become more open as seen on the high number of xerophy-
velocity (Spearman rho ¼ 0.347) and substrate type (Spearman lous mollusk elements. The expansion of open-vegetation areas
rho ¼ 0.354) (Appendix:Table 3). This means that transformations and grasslands in the vicinity of the site, as well as some higher-
in the economy are hallmarked by shifts in the quality of the lying floodplain areas must have favored the creation of extensive
surrounding riparian environment. arable lands and pasturelands. Furthermore, in a period of relative
water shortage (lowstand in rivers) settling beside watercourses
5. Discussion must have been quite advantageous in the heart of the Carpathian
Basin, the driest region in Central Europe. The number of harvested
The studied Late Neolithic tell site Hódmezo }vásárhely-Gorzsa shellfish was low in this period. This points to a minimal and
from the Lower-Tisza region spans almost the entire period of the alternative use of shellfish at the site other than dietary purposes.
evolution of the Late Neolithic Tisza culture from its formative According to the available archeology, pottery with straw, shell and
phase (Tisza I. Horváth, 2005) up to its transitional phase to the crushed ceramic temper is confined to this period alone (Horváth,
Early Copper Age Tiszapolgár culture (Tisza VI. Horváth, 2005). The 2005). So shellfish might have been used primarily as a fertilizer
life of the Tisza culture starts somewhat later in Gorzsa (during in a period of low floods to compensate for the loss of organic
Tisza II. or locally D1 phase) and entails all the remaining phases of matter generally transported by floodwaters onto the floodplain.
cultural evolution. This enabled us to compare transformations Then the empty shells were crushed and utilized as temper in
observable in the riparian environment and the role of second-line pottery.
freshwater resources with the socioeconomic evolution of the Conversely, a marked transformation in the hydraulics of the
referred tell complex (Figs. 6 and 7.). floodplain is observable from the second half of the next settlement
Phase D1 starting around 4900 cal BC in Gorzsa (later phase of stage D3 (Tisza III.). This brought about an increase in water velocity
the Tisza II. period) is characterized by calm conditions on the on the one hand, and a shift in the geochemistry of the water seen
floodplain, with a prevalence of muddy substrate around the site, in the reduced ratio of dissolved manganese and iron in the shells
a high rate of primary production, low oxygen levels, and relatively as well. All these factors indicate an elevated freshwater supply to
stable pH conditions in the pools of the backswamp (see Fig. 9.a). As the backswamp areas near the site. Transformations inferred for the
shown by the archeology (Horváth, 1982, 1987, 1989, 2005), this substrate also point to the development of intensified flood activity
j.jas.2011.07.019
Fig. 9. Inferred environment of the study site during major settlement phases.
during the referred period. Higher floods must have carried more humidity values as a result of the relatively higher temperatures.
organic matter onto the floodplain boosting the fertility of gallery This and higher fertility of soils enhanced bioproduction in the
woodlands and wet meadows, which was beneficial for grazing gallery forest and adjacent wet meadows of the floodplain creating
animals wild or domestic (Fig. 9.b). Floods must have reached ideal conditions for large bodied ruminants like auroch or red-deer
higher levels in the regions of the Upper and Middle Tisza, so their (Fig. 9.c and d.). Their hunting seems to have played a continuously
fertilizing work must have been more beneficial there compared to increasing role in the lives of Late Neolithic communities
the southern areas. It was this period when new contemporaries of (Bartosiewicz, 2005; Szabó, 2007) along with fishing and shellf-
the Tisza culture, the Herpály and Cso } szhalom groups, character- ishing. The most prominent harvest peaks for shellfish seem to
ized by a subsistence mainly based on auroch hunting and animal correspond to major shifts observable in our environmental proxies
husbandry, first appear (Horváth, 2005; Raczky et al., 2004). The in relation to the individual subphases. Similary to the proxies for
referred period also hallmarks the so-called second migration wave pH and dissolved oxygen levels a dual peak characterizes the
of wild fauna (Vörös, 2003, 2005), when numerous large bodied intensity of shellfishing in subphase C2. Similar opportunistic
gallery woodland elements like auroch and red-deer surged into economic activities with an increased reliance on fish and shellfish
the area of the Carpathian Basin from the neighboring areas, most were recorded in numerous Late Neolithic sites in Romania con-
likely as a result of the appearance of lush pasturelands on the nected to the so-called Gulmenita culture in Bordusani and Harsova
floodplain attributed to these recurrent floods. The blessings of in a riparian setting as our site (Radu, 1997; Sárkány-Kiss and Bolos,
these iterative floods on agricultural activities are clearly seen in 1996a,b; Venczel, 1997).
the low amount of harvested shellfish during this period. The inferred iterative floods in the referred periods must have
The next settlement stage (phase C or (Tisza IV.)), giving the been initially beneficial for the Late Neolithic communities creating
thickest occupational deposit at the site of Gorzsa (Horváth, 1982), extensive, fertile pasturelands on the floodplain, and attracting
is characterized by rapidly changing floodplain environment, large bodied wild fauna into the region. The mass appearance of
where the emergence of almost every subphase (C1, C2, C3) was easily available protein sources tempted communities to increas-
accompanied by the development of stream conditions and ingly turn to such originally second-line activities like hunting,
a transformation of the substrate as well marking the appearance of fishing, shellfishing and domestication of wild ruminants (Horváth,
prominent iterative floods. All this is seen not only in our stream 2005; Raczky, 1988; Bökönyi, 1974, 1988; Sherratt, 1982; Szabó,
velocity proxies, but the cyclical increases of the dissolved oxygen 2007, Bartosiewicz, 2005). The dominant animals in level 3 corre-
content of the water body, and the fluctuations in the pH values sponding to settlement phases CeB in Gorzsa were cattle and
(Fig. 6). These latter proxies have a dual peak in the level corre- auroch. The number of aurochs was greatly reduced by level 1
sponding to subphase C2 and single peaks at subphases C1 and C3 (phase A) most likely as a result of the depletion of the herd as an
each. outcome of massive hunting of large bodied bulls. Nevertheless, the
The inferred iterative floods seem to get more pronounced ratio of other wild components like red-deer remained relatively
upwards in the sequence accompanied by a sharp increase in constant marking the elevated importance of hunting in the
j.jas.2011.07.019
subsistence of these communities here (Szabó, 2007). Similar the groundlevel (Kalicz and Raczky, 1987). This must have been an
economic transformations have been documented in connection important device in fight against elevated groundwater tables and
with the northern Vra culture, where people have given up farming inland waters. The mentioned prominent floods and accompanying
turning to traditional hunting, fishing, gathering activites in their inland waters must have trigerred similar structural changes in
subsistence as a result of the development of favorable conditions Gorzsa seen in a multiple displacement of the settlement center
for prey animals. The referred transformations in the lives of the and most likely led to the final abandonment of the settlement after
people of the Tisza culture appeared in their cult lives and material phase B around 4500 cal BC besides other social, cultural processes
culture as well (Horváth, 1982, 1987, 1988, 2005). There are an (Horváth, 2005; Parkinson et al., 2004, 2010).
increasing number of animal idols appearing from levels of the
referred settlement phases pointing to the increasing importance Acknowledgments
of hunting and animal husbandry in subsistence. Several times
foundation deposits turn up in connection with the erection of new The authors would like to express their gratitude to all
houses. Numerous pits with thick layers of freshwater shells and colleagues working at the Department of Geology and Paleon-
auroch horns and skulls under the threshold of houses are recorded tology, University of Szeged for assisting in the technical part of the
from phase C in Gorzsa (Horváth, 1982). Plus several clay bull paper preparation and a part of the labwork. Special thanks to
figurines appear on the walls of houses expressing the importance Ferenc Horváth at the Móra Museum of Szeged for granting access
of these creatures in the lives of the community. Sacrificial pits with to the material subjected to analysis. Work was supported from the
a similar role and content were recorded in Tisza I. phase at the site research fund of the Department of Geology and Paleontology,
of Öcsöd-Kováshalom (Kalicz and Raczky, 1987; Raczky, 1987, 1988). University of Szeged, Hungary. Finally, thanks are due to the
Initially beneficial iterative floods turn more pronounced upwards editorial board of The Journal of Archaeological Science and the five
in the profile toward the terminal part of the evolution of the tell. reviewers whose comments made the final version of this paper
This is clearly visible in all our proxies recording water velocity, stronger and clearer than early drafts. Work was supported from
oxygen level, pH, humidity, or substrate conditions. The most the research fund of Departments of Geology and Geography,
prominent floods appear from the middle part of the phase University of Szeged, Hungary no. TÁMOP-4.2.1/B-09/1/KONV-
(subphase C2) up to subphase C3 and the succeeding phase B (Tisza 2010-0005.
V. after Horváth, 2005) (Fig. 9d). Similarly to the modern conditions,
floodwaters must have been significantly higher in the Middle and
References
Upper Tisza Region than the southern parts of the Great Hungarian
Plains. This may explain why settlements from the northern areas Bökönyi, S., 1988. Környezeti és kulturális hatások késo}-neolitikus kárpát-medencei
of the Great Hungarian Plains retreat south to the Maros-Körös és balkáni lelo} helyek csontanyagán. Akadémiai Kiadó, Budapest.
Intefluve, the Lower-Tisza region and the valley of the Aranka river Bökönyi, S., 1974. The history of domestic mammals in central and eastern Europe.
Akadémiai Kiadó, Budapest.
by the end of the Tisza IV phase (phase C) (Horváth, 2005) (Fig. 1, Bailey, G., Milner, N., 2000. The marine molluscs from the Norsminde shell midden.
Table 1). Heightened floodwaters must have finally significantly In: Andersen, S. (Ed.), Stone Age Settlement in the Coastal Fjord of Norsminde
reduced the original area of pasturelands, arable lands and living Jutland, Denmark.
Bartosiewicz, L., 2005. Plain talk: animals, environment and culture int he Neolithic
quarters with the recurrent appearance of high subsurface of the Carpathian Basin and adjacent areas. In: Bailey, D.W., Whittle, A.,
watertables and persistant inundations. This may have finally Cummings, V. (Eds.), ) (Un)settling the Neolithic: Breaking down concenpts,
posed a crisis inducing frequent alteration of the settlement boundaries and origins. Oxbow, Oxford, pp. 51e63.
Bartosiewicz, L., 2006. Régenvolt háziállatok. L’Harmattan Kiadó, Budapest.
structure and displacement of the houses. The inferred largest Bauer, G., Wachtler, K., 2001. Ecology and evolution of the freshwater mussels
spatial extent of the tell during this phase (Horváth, 2005), most Unionidae, Ecological studies. Springer, Berlin, p. 545.
likely was the result of these environmental transformations In the Behrensmeyer, A.K., 1991. Terrestrial vertebrate accumulation. In: Allison, P.A.,
Briggs, D.E.G. (Eds.), Taphonomy, releasing the data locked in the fossil record.
north where floodwaters were higher this process might have been
Plenum, New York, pp. 291e335.
faster and must have appeared earlier as we have seen in connec- Behrensmeyer, A.K., Hill, A.P. (Eds.), 1980. Fossils in the Making: Vertebrate
tion with the site of Polgár-Cso }szhalom (Raczky et al., 2004; Taphonomy and Paleoecology. University of Chicago Press, Chicago.
Behrensmeyer, A.K., Kidwell, S.M., Gastaldo, M.A., 2000. Taphonomy and paleobi-
Sümegi, 2009). Such negative influences are visible in trans-
ology. Paleobiology 26, 103e144.
formations of the settlement structure of another northerly site Bird, D.W., Bliege-Bird, R.L., 2000. The Ethnoarchaeology of Juvenile Foragers:
Berettyóújfalu-Herpály (Kalicz and Raczky, 1987). In its initial phase shellfishing strategies among Meriam Children. Journal of Anthropological
the settlement was confined to a minor core area on the banks of Archeology 19, 461e476.
Bognár-Kutzián, I., 1966. Das Neolithikum in Ungarn. Archeologia Austriaca 40,
the river surrounded by a double moat on its western waterfront 249e280.
side, most likely functioning as a drainage channel for extensive Broughton, J.M., 1994. Declines in mammalian foraging efficiency during the late
groundwater and floodwaters. The core was made up of closely Holocene. San Francisco Bay, California. Journal of Anthropological Archeology
13, 371e401.
packed houses. This compact settlement structure was preserved in Broughton, J.M., 1999. Resource depression and intensification during the late
the later phases as well (early, classical and transitional Herpály Holocene, San Francisco Bay. University of California Press, Berkeley.
stages) with a slight displacement of the moat (Kalicz and Raczky, Butler, V.L., Campbell, S.K., 2004. Resource intensification and resource depression
in the Pacific Northwest of North America: a zooarchaeological review. Journal
1987). A major transformation visible in the creation of a looser of World Prehistory 18 (4), 327e405.
arrangement of the houses and the emergence of a wider central Byers, D.A., Broughton, J.M., 2004. Holocene environmental change, Artiodactyl
space was documented only in the late phase of the evolution of the abundances, and human hunting strategies in the great Basin. American
Antiquity 69, 251e271.
site. The initial 3 periods roughly correspond to the CeB settlement Cannon, M.D., 2000. Large mammal relative abundance in Pithouse and Pueblo
stages at our site of Gorzsa. While the latter late Herpály period is period archeofaunas from southwestern New Mexico: resource depression
roughly coeval with our B-A phases. Numerous foundation deposits among the Mibres-Mogollon? Journal of Anthropological Archaeology 19,
17e347.
with auroch skulls have been documented in the early classical
Cannon, M.D., Meltzer, D.J., 2004. Early Paleoindian foraging: examining the faunal
phases, highlighting the importance of these animals in subsistence evidence for large mammal specialization and regional differences. Quaternary
during this time of the evolution of the settlement, which was Science Reviews 23, 1955e1987.
characterized by frequent floods as we have seen. Furthermore, Caroll, M., Romanek, C.S., 2008. Shell layer variation in trace element concentration
for the freshwater bivalve Elliptio complanata. Geo-Marine Letters 28, 369e381.
numerous subfloor wooden frame structures are documented from Charnov, E.L., Orians, G.H., 1973. Optimal Foraging: Some theoretical explorations.
this period, which enabled the elevation of the housefloor above MS Department of Biology, University of Utah, Salt Lake City, Utah.
j.jas.2011.07.019
Chatfield, C., Collins, A.J., 1980. Introduction to Multivariate Analysis. Chapman and Jones, E.L., 2004. Dietary evenness, prey choice and human environment interac-
Hall, New York. tions. Journal of Archeological Science 31, 307e317.
Claasen, C., 1998. Shells. Cambridge University Press, Cambridge. Kalicz, N., Makkay, J., 1977. Die Linienbandkeramik in der Grossen Ungarischen
Cochrane, G.W.G., 2003. Artefact atribute richness and sample size adequacy. Tiefebene, vol. VII. Studia Archaeologica, Budapest.
Journal of Archeological Science 30, 837e848. Kalicz, N., Raczky, P. 1987. The Late Neolithic in the Tisza region: a survey of recent
Davies, P., 1998. Numerical analysis of sub-fossil wet-ground molluscan taxocenes archeological research. In: Tálas, L., Raczky, P. (Eds.), The Late Neolithic in the
from overbank alluvium at Kingsmead Bridge, Wiltshire. Journal of Archeo- Tisza region. Budapest-Szolnok, pp. 11e31.
logical Science 25, 39e52. Kalis, A.J., Merkt, J., Wunderlich, J., 2003. Environmental changes during the
Davis, B.A.S., Brewer, S., Stevenson, A.C., Guiot, J., Contributors, Data, 2003. The Holocene climatic optimum in central Europe - human impact and natural
temperature of Europe during the Holocene reconstructed from pollen data. causes. Quaternary Science Reviews 22, 33e79.
Quaternary Science Reviews 22, 1701e1716. Kelly, L.R., 1995. The Foraging Spectrum: Diversity in Hunter-Gatherer Lifeways.
Denton, G.H., Karlen, W., 1973. Holocene climatic variations: their pattern and Smithsonian Institution Press, Washington, D.C.
possible cause. Quaternary Research 3, 155e205. Kent, B., 1988. Making dead oysters talk. Maryland Historical Trust, St. Mary City.
Digby, P.G.N., Kempton, R.A., 1987. Population and Community Biology Series: Kintigh, K.W., 1984. Measuring archeological diversity by comparison with simu-
Multivariate Analysis of Ecological Communities. Chapman and Hall, London. lated assemblages. American Antiquity 49, 44e54.
Erman, D.C., Erman, N.A., 1984. The response of stream macroinvertebrates to Kintigh, K.W., 1989. Sample size, significance and measures of diversity. In:
substate size and heterogeneity. Hydrobiologia 108, 75e82. Leonard, R.D., Jones, G.T. (Eds.), Quantifying diversity in archeology. Cambridge
Evans, J.G., 1972. Landsnails in archeology. Seminar Press, London. University Press, Cambridge, pp. 25e36.
Grayson, D.K., Delpech, F., 1998. Changing diet breadth in the early Upper Paleolithic Klein, R.G., Cruz-Uribe, K., 1984. The analysis of animal bones from archeological
of southwestern France. Journal of Archeological Science 25, 1119e1129. sites. University of Chicago Press, Chicago.
Grayson, D.K., Delpech, F., 2002. Specialized early Upper Paleolithic Hunters in Koike, H., 1979. Seasonal dating and the valve-pairing technique in shell-midden
southwestern France. Journal of Archeological Science 29, 1439e1449. analysis. Journal of Archeological Science 6, 63e74.
Gulyás, S., Sümegi, P., 2004. Kagylógyu } jtés a korai neolitikumban Magyarországon Korek, J., 1989. Die Theiss-kultur in der mittleren und nördlichen Theissgegend
egy Körös lelo }hely Tiszapüspöki példáján. SOOSIANA Hungarian Malacological Budapest.
Newsletter Supplement 32, 1e60. Krolopp, E., Sümegi, P., 1992. A magyarországi löszök képzo } désének paleoökológiai
Gulyás, S., Sümegi, P., 2011. Riparian environment in shaping social and economic rekonstrukciója mollusca fauna alapján. In: Szöo }r, Gy. (Ed.), Fáciesanalitikai,
behaviour during the first phase of the evolution of Late Neolithic tell paleobiogeokémiaI és paleoökoógiai kutatások. MTA Debreceni Akadémiaia
complexes in SE Hungary (6th/5th millenia BC). Journal of Archaeological Bizottság, Debrecen, pp. 263e278.
Science. doi:10.1016/j.jas.2011.06.005. Krolopp, E., Sümegi, P., 1995. Paleoecological reconstruction of the late Pleistocene
Gulyás, S., Sümegi, P., Molnár, M., 2010. New radiocarbon dates from the Late based on loess Malacofauna in Hungary. GeoJournal 36, 213e222.
Neolithic tell settlement of Hódmezo }vásárhely-Gorzsa, SE Hungary. Radio- Langlet, D., Alleman, L.Y., Plisnier, P.D., Hughes, H., André, L., 2006. Mn seasonal
carbon 52, 1458e1464. upwellings recorded in Lake Tanganyika mussels. Biogeosciences Discussions 3,
Gulyás, S., Tóth, A., Sümegi, P., 2007. The zooarcheological analysis of freshwater 1453e1471.
bivalve shells and their relevance regarding the life of a Neolithic community. Langlet, D., Alleman, L.Y., Plisnier, P.D., Hughes, H., André, L., 2007. Manganese
In: Whittle, A. (Ed.), The Early Neolithic on the Great Hungarian Plain- Inves- content records seasonal upwelling in Lake Tanganyika mussels. Biogeosciences
tigations of the Körös culture site of Ecsegfalva 23, Békés County, Varia 4, 195e203.
Archeologica 21. MTA, Budapest, pp. 395e413. Lazareth, C.E., Vander Putten, E., André, L., Dehairs, F., 2003. High resolution trace
Gyucha, A., Parkinson, W.A., Yerkes, R.W., 2009. A multi-scalar apprach to settle- element profiles in shells of the mangrove bivalve Isognomon ephippium:
ment pattern analysis: the transition from the late neolithic to the early Copper a record of environmental spatio-temporal variations. Estuarine. Coastal and
age on the great Hungarian Plain. In: Thurston, T., Salisbury, R. (Eds.), Reima- Shelf Science 57, 1103e1114.
gining regional analysis: the archaeology of spatial and social dynamics. Cam- Lee, G.F., Wilson, W., 1969. Use of chemical composition of freshwater clamshells as
bridge Scholars Publishing, New Castle Upon Tyne, pp. 100e129. indicators of paleohydrologic conditions. Ecology 50, 990e997.
Hammer, O., Harper, D.A.T., 2005. Paleontological Data Analysis. Wiley-Blackwell, Leonard, R.D., Jones, G.T. (Eds.), 1989. Quantifying diversity in archeology. Cam-
New York. bridge University Press, Cambridge.
Hertelendi, E., Horváth, F., 1992. Radiocarbon chronology of the late neolithic Lo
zek, V., 1964. Quartermollusken der Tschechoslowakei. Rospravy Ustredniho
settlements in the Tisza-Maros region. Radiocarbon 34 (3), 859e866. Ustadu Geologickeho 31 Praha.
Szöo} r, Gy., Balázs, É, Cserháti, Cs., Dinya, Z., Hertelendi, E., Sümegi, P., Szanyi, J., 1992. Lupo, K.D., 2007. Evolutionary foraging Models in zooarcheological analysis: recent
Quarter és Neogén mollusca-héjak kemotaxonómiai és paleoökológiai ele- Applications and Future Challenges. Journal of Archaeological Research 15,
mzése. In: Szöo } r, Gy (Ed.), Fáciesanalitikai, paleobiogeokémiai és paleoökológiai 143e189.
kutatások. MTA Debreceni Akadémiai Bizottság, Debrecen, pp. 111e183. Lyman, R.L., 1984. Bone density and differential survivorship of fossil classes. Journal
Hertelendi, E., Svingor, É., Raczky, P., Horváth, F., Futó, I., Bartosiewicz, L., Molnár, of Anthropological Archaeology 3, 259e299.
M., 1998. Radiocarbon chronology of the Neolithic and timespan of tell Lyman, R.L., 1994. Quantitative units and terminology in Zooarcheology. American
settlements in Eastern Hungary based on calibrated radiocarbon dates. In: Antiquity 59, 36e71.
Költo } , L., Bartosiewicz, L. (Eds.), Archeometrical Research in Hungary II, Buda- Magurran, A.E., 1988. Ecological diversity and its measurement. Princeton Univer-
pest, pp. 61e69. sity Press.
Hill, F., 1975. Effects of the environment on animal exploitation by Archaic Inhab- McArthur, R.H., Levins, R., 1967. The limiting similarity, convergence and divergence
itants of the Koster site, Illinois. Ph.D. dissertation, Biology, University of of coexisting species. American Naturalist 101, 377e385.
Louisville. McArthur, R.H., Pianka, E.R., 1966. On optimal use of patchy environment. American
Hill, M.O., Gauch Jr., H.G., 1980. Detrended Correspondence Analysis: an improved Naturalist 100, 603e609.
ordination technique. Vegetatio 42, 47e58. McAulife, J.R., 1984. Competition for space, disturbance and the structure of the
Horváth, F., 1982. A Gorzsai halom késo } neolit rétege. Archeológiai Értesíto} I (109), benthic stream community. Ecology 65, 894e908.
201e220. Mellars, P., 1996. The Neanderthal Legacy. Princeton University Press, Princeton.
Horváth, F. 1987. Hódmezo } vásárhely-Gorzsa. A settlement of the Tisza culture. In: Murphy, J.F., Davy-Bowker, J., 2005. Spatial structure in lotic macroinvertebrate
Tálas, L., Raczky, P. (Eds.), The Late Neolithic in the Tisza Region. Budapest- communities in England and Wales: relationship with physical, chemical and
Szolnok, pp. 31e46. anthropogenic stress variables. Hydrobiologia 534, 151e164.
Horváth, F., 1988. Late Neolithic ditches, fortifications, and tells in the Hungarian Mutvei, H., Westermark, T., 2001. How environmental information can Be Obtained from
Tisza region. In: Tasic, N., Petrovic, J. (Eds.), Chronologie und stratigraphie der Naiad shells. In: Bauer, G., Wachtler, K. (Eds.), Ecology and evolution of
vorgeschichtlichen und Antiken kulturen der Donauniederung und Südos- the freshwater mussels Unionidae, Ecological Studies. Springer, Berlin, pp. 367e379.
teuropas. Band I, Symposium Ruma, Novi Sad, pp. 145e149. Nagaoka, L., 2001. Using diversity indices to measure changes in prex choice at the
Horváth, F., 1989. A survey on the development of neolithic settlement pattern and Shag River Mouth site, southern New Zeeland. International Journal of Osteo-
house types in the Tisza region. In: Bökönyi, S. (Ed.), Neolithic of Southestern archeology 11, 101e111.
Europe and its Near Eastern Connections, vol. 2. Varia Archaeologica Hungarica, Nagaoka, L., 2002. The effects of resource depression on foraging efficiency, diet
Budapest, pp. 85e101. breadth, and patch use in southern New Zealand. Journal of Anthropological
Horváth, F., 1994. A dél-alföldi újko }kor kutatás új szempontjai, módszerei és Archaeology 21, 419e442.
eredményei. unpublished PhD thesis, MTA Budapest-Szeged. Nagaoka, L., 2005. Declining foraging efficiency and moa carcass exploitation in
Horváth, F., 2003. Az újko }kori fejlo } dés vázlata és ido } keretei a Dél-Alföldön. In: southern New Zealand. Journal of Archeological Science 32, 1328e1338.
Visy, Zs (Ed.), Hungarian Archeology at the turn of the Millenium. Ministry Neeley, M.P., Clark, G.A., 1993. The human food niche in the Levant over the past
of National Cultural Heritage. Teleki László Foundation, Budapest, 150,000 years. In: Peterkin, G.L., Bricker, H., Mellars, P. (Eds.), Hunting and
pp. 100e101. animal exploitation in the Late Paleolithic and Mesolithic of Eurasia, vol. 4.
Horváth, F., 2005. Gorzsa. Elo } zetes eredmények az újko }kori tell 1978 és 1996 közötti Archeological Papers of the American Anthropological Association, pp. 221e240.
feltárásából. In: Bende, L., Lo } rinczy, G. (Eds.), Hétköznapok vénuszai. Móra Nichol, R.K., Wild, C.J., 1984. Numbers of individuals" in faunal analysis: the decay of
Ferenc Múzeum, Szeged, pp. 51e73. fish bone in archaeological sites. Journal of Archeological Science 11, 35e51.
Jerardino, A., Navarro, R., 2008. Shell morphometry of seven limpet species from Parkinson, W.A., Yerkes, R.V., Gyucha, A., 2004. The transition from the neolithic to
coastal shell middens in southern Africa. Journal of Archaeological Science, the Copper age: excavations at Vészto }-Bikeri, Hungary 2000-2002. Journal of
1023e1029. Field Archeology 29, 101e121.
j.jas.2011.07.019
Parkinson, W.A., Yerkes, R.V., Gyucha, A., Sarris, A., Morris, M., Salisbury, R., 2010. Sárkány-Kiss, A., Bolos, F., 1996b. Malacological aspects of the archeological exca-
Early Copper age settlements in the Körös region of the great Hungarian plains. vations in the Neolithic-Eneolithic settlement of BorduŞani. Communicare
Journal of Field Archeology 35 (2), 164e183. sustinuta la Colocviul International “ Viata pe malui Dunárii acum 6500 ani".
Peacock, E., 2000. Assessing bias in archaeological shell assemblages. Journal of National de Istorie a Romaniei, Bucuresti.
Field Archaeology 27 (2), 183e196. Sümegi, P., 1989. A Hajdúság felso }-pleisztocén fejlo }déstörténete finomrétegtani
Peacock, E., 2002. Shellfish use during the woodland period in the middle south. In: (o}slénytani, szedimentológiai) vizsgálatok alapján. unpublished PhD thesis,
Anderson, D.G., Mainfort, R. (Eds.), The Woodland Southeast, Tuscaloosa. The KLTE, Debrecen.
University of Alabama Press, pp. 444e460. Sümegi, P., 2009. Ember és környezet kapcsolata a középso }-bronzkorban: az o
}skori
Peacock, E., Seltzer, J.E., 2008. A comparison of multiple proxy data sets for pale- gazdasági tér fejlo }dése egy bronzkori tell geoarcheológiai és környezettörténeti
oenvironmental conditions as derived from freshwater bivalve (Unionid) shell. feldolgozása során. Tisicum XIX, 457e480.
Journal of Archaeological Science 35, 2557e2565. Sümegi, P., Krolopp, E., 1995. A magyarországi würm korú löszök képzo }désének
Peat, R.K., 1974. The measurement of species diversity. Annual Review of Ecology paleoökológiai rekonstrukciója. Földtani Közlöny 124, 125e148.
and Systematics 5, 285e307. Sherratt, A., 1982. The development of Neolithic and Copper Age settlement in the
Peterson, M.L., 1989. Marine invertebrate analysis of CA-SLO-165: Quantification, Great Hungarian Plain. Part I and II. The regional setting; site survey and
measurement theory, formation and ecological history. Report for Singer and settlement dinamics. Oxford Journal of Archeology 1, 287e316.
Associates, California, p. 110. Siegele, R., Orlic, I., Cohen, D.D., Markich, S.J., Jeffree, R.A., 2001. Manganese profiles
Pike-Tay, A., Bartosiewicz, L., Gál, E., Whittle, A., 2004. Body part representation and in freshwater mussel shells. Bulletin of Nuclear Instruments and Methods in
seasonality: sheep/goat, bird and fish remains from Early Neolithic Ecsegfalva Physics Research 181, 593e597.
23, SE Hungary. Journal of Taphonomy 2, 221e222. Simpson, G.G., Roe, A., Lewontin, R.C., 1960. Quantitative zoology. Harcourt, New York.
Pradyot, P., 2004. Dean’s analytical chemistry handbook, 2nd edition. McGraw-Hill, Sokal, R.R., Rohlf, F.J., 1995. Biometry. Freeman Publishers, New York.
New York. Soós, L., 1943. A Kárpát-medence mollusca faunája e I. Állattani rész, Magyarország
Raczky, P., 1986. Megjegyzések az "alföldi vonaldíszes kerámia" kialakulásának természetföldrajz. MTA, Budapest.
kérdéséhez. Régészeti tanulmányok Kelet-Magyarországról. Folklór és Etnog- Stiner, M.C., Munro, N.D., 2002. Approaches to prehistoric diet breadth, demog-
ráfia 24, 45e59. raphy and prey ranking systems in time and space. Journal of Archeological
Raczky, P., 1987. Öcsöd-Kováshalom. In: Tálas, L., Raczky, P. (Eds.), The Late Neolithic Method and Theory 9/2, 181e214.
in the Tisza Region. Budapest-Szolnok, pp. 61e85. Stumm, W.M., Morgan, J.J., 1996. Aquatic chemistry. John-Wiley, New York.
Raczky, P., 1988. A Tiszavidék kulturális és kronológiai kapcsolatai a Balkánnal és az Szabó, D., 2007. A Tiszai-kultúra állattartásának jellegzetességei Hód-
Égeikummal a neolitikum, rézkor ido }szakában. Újabb kutatási eredmények és mezo } vásárhely-Gorzsa késo } neolit tell csontanyagának vizsgálata alapján.
problémák. The cultural and chronological connections of the Tisza region with unpublished MSc thesis, University of Szeged, Szeged.
the Balkan and Aegean at the time of the Neolithic and Chalcolithic. Recent Venczel, M., 1997. Amphibians and reptiles. Archeological researches at Borduşani-
research results and problems, Szolnok. Popina (Ialomita County). Preliminary report 1993e94. Cerceta ri Arheologice
Raczky, P., Meier-Arendt, W., Anders, A., Hajdú, Zs., Nagy, E., Kurucz, K., 10, 126e127.
Domboróczki, L., Sebo }k, K., Sümegi, P., Magyari, E., Szántó, Zs., Gulyás, S., Vörös, I., 2003. The mammals of Hungary. In: Visy, Zs (Ed.), Hungarian Archeology
Dobó, K., Bácskay, E., T.Bíró, K., Schwartz, Ch, 2004. Polgár-Cso }szhalom at the turn of the Millenium. Ministry of National Cultural Heritage. Teleki
(1989e2000): summary of the Hungarian-German excavations on a neolithic László Foundation, Budapest, pp. 73e74.
settlement in Eastern Hungary. In: Aslan, R. (Ed.), Mauerschau, Festschrift für Vörös, I., 2005. Neolitikus állattartás és vadászat a Dél-Alföldön. In: Bende, L.,
Manfred Korfmann, Remshalden-Grumbach. Verlag, Stuttgart, pp. 833e860. Lo} rinczy, G. (Eds.), Hétköznapok vénuszai. Móra Ferenc Múzeum, Szeged,
Radu, V., 1997. Archeological researches at Bordus ani-Popina (Ialomita County). pp. 203e228.
Preliminary report 1993e94. Pisces. Cerceta ri Arheologice 10, 96e125. Vander Putten, E., Dehairs, F., Keppens, E., Baeyens, W., 2000. High resolution
Ravera, O., Cenci, R., Beone, G.M., Dantas, M., Lodigiani, P., 2003. Trace element distribution of trace elements in the calcite shell of modern Mytilus edulis:
concentrations in freshwater mussels and macrophytes as related to those in environmental and biological controls. Acta Geochimica et Cosmochemica 64,
their environment. Journal of Limnology 62 (1), 61e70. 997e1011.
Ravera, O., Beone, G.M., Trincherini, P.R., Riccardi, N., 2007. Seasonal variations in Voigt, R., Grüger, E., Baier, J., Meischner, D., 2008. Seasonal variability of Holocene
metal content of two Unio pictorum mancus populations from two lakes of climate: a paleolimnological study of varved sediments in Lake Jues (Harz Mts.
different trophic state. Journal of Limnology 66, 28e39. Germany). Journal of Paleolimnology 40, 1021e1052.
Reitz, E.J., Wing, E.S., 1999. Zooarcheology. Cambridge University Press, Cambridge. Waguespack, N.M., Surovell, T.A., 2003. Clovis hunting strategies, or How to Make
Richnovszky, A., Pintér, L., 1979. A vízi csigák és kagylók (mollusca) kishatározója. out on Plentiful resources. American Antiquity 68 (2), 333e352.
Vízügyi hidrológia, 6. Vízügyi Dokumentáció és Továbbképzo } Intézet, Weber, E., 2005. Population size and structure of three mussel species (Bival-
Budapest. via:Unionidae) in a northeastern German river with special regard on influ-
Rustioni, M., Mazza, P., Magnatti, M., 2007. Multivariable analysis of an Italian late ences of environmental factors. Hydrobiologia 537, 169e183.
neolithic archaeofauna. Journal of Archaeological Science 34, 723e738. Wolverton, S., 2005. The effects of the hypsithermal on prehistoric forgaing effi-
Sárkány-Kiss, A., Bolos, F., 1996a. Malacological aspects of the archeological exca- ciency in Missouri. American Antiquity 70, 91e106.
vations in the Neolithic-Eneolithic settlement of H arŞova. Communicare sus- Yamazaki, T., Oda, S., 2009. Changes in shell gathering in an early agricultural society
tinuta la Colocviul International “Viata pe malui Dunárii acum 6500 ani”. at the head of Ise Bay, Japan. Journal of Archaeological Science 36, 1578e1589.
National de Istorie a Romaniei, Bucuresti. Zar, J.H., 1996. Biostatistical Analysis. Prentice Hall, New Jersey.
j.jas.2011.07.019
View publication stats

JASC 2011a GulyasSmegiVEG

Uploaded by

Copyright:

Available Formats

JASC 2011a GulyasSmegiVEG

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

JASC 2011a GulyasSmegiVEG

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Early Neolithic man and riparian environment

Article · January 2003

Evolution, Palaeobiogeography, Palaeoecology and Palaeoclimatology of the Quaternary Mollusca

The user has requested enhancement of the downloaded file.

Contents lists available at ScienceDirect

Journal of Archaeological Science

1. Introduction Gödrösök and Öcsöd-Kováshalom along the Tisza’s major southern

Table 1 One of the most remarkable multi-layered (tell) settlements of

ek (1964). Several widely adopted measures are known to track

Fig. 4. The steps of multiproxy paleoecological analysis of freshwater shells.

Paleoecological group Taxa NISP MNI Proportional Abudance by

Fig. 5. Results of multivariate ordination (DCA) for freshwater taxa.

*Statistically signiﬁcant correlation p ¼ 0.05 (2-tailed).

Kruskal-Wallis test Unio pictorum size

Sarmatian Bronze N1 N2 N3 N5 N6 N7 N8 N9 N10

Kruskal-Wallis test Unio pictorum size

Mann-Whitney pairwise comparisons

You might also like