What did grinding stones grind? New light on Early Neolithic subsistence economy in the Middle Yellow River Valley, China.
Liu, Li ; Field, Judith ; Fullagar, Richard 等
[ILLUSTRATION OMITTED]
Introduction
The Peiligang culture (c. 7000-5000 BC) represents the earliest
Neolithic settlement in the Middle Yellow River Valley and signals the
emergence of food production and ritual complexity in the region. These
developments indicate a transformation from mobile hunter-gatherer
society, characterised by microlithic technology, to an
agricultural-based Neolithic economy (Lu 1999), which eventually led to
the formation of Chinese dynastic states several thousand years later
(Chang 1986; Liu, L. 2004). Archaeological research in China has for
decades focused on the Peiligang culture for insight into the origins of
plant and animal domestication (Shi 1992; Henan Institute 1999; Lee et
al. 2007; Luo 2007), sedentary settlement and the full development of
cereal (mainly millet) farming (e.g. Bellwood 2005: 120-22; Smith, B.
1995: 134-9; Underhill 1997). However, the argument for the development
of a cereal-based agricultural society needs to be further evaluated by
scientific methods.
One approach to investigating ancient subsistence is to study the
function of grinding stones, which are frequently found at Early
Neolithic sites in north China, and constitute a high proportion of
Peiligang stone tool assemblages (Liu, L. 2008). The interpretations of
their function are based mainly on ethnographic analogy and range from
hide-working (Zhao, S. 2005) and wild plant processing (Wu 1986; Liu, L.
2008), to dehusking domesticated cereals (Song 1997; Chen 2002). The
last view has been widely accepted and thus grinding stones are claimed
to provide key evidence for early agriculture in China (Chang 1986: 91;
An 1989; Yan 1992: 114; Smith, B. 1995: 134; Bellwood 2005: 121; Higham
2005: 240). In this paper we present the results of functional analyses
on six grinding stones from two Peiligang culture sites, Shigu and Egou,
which challenge this view.
The sites and their assemblages
The Shigu site in Changge county is situated at the confluence of
the Shiliang and Xiaohong rivers. To the west of the site are the Funiu
Mountains, and to the east is the Central Henan floodplain (Figure 1). A
total of 214[m.sup.2] was excavated, revealing three subterranean house
foundations, 189 ash pits, 69 burials and 440 artefacts. Carbonised
plant remains were identified as hazelnuts, walnuts, elm fruit and
jujube. Eleven grinding stones, mostly unearthed from burials, account
for 14 per cent of the lithic assemblage (Henan Institute 1987). The
Egou site in Mixian is situated on a tableland at the confluence of the
Wei and Sui rivers and surrounded by low hills. The site is
8000[m.sup.2] in extent and an area of 2781[m.sup.2] was excavated in
the 1970s. The cultural deposits are thin at 0.3-0.5m in depth. Six
subterranean house foundations, 44 ash pits and around 370 artefacts
were uncovered. Among 133 stone tools, 20 are grinding stones, mostly
from burials (18), accounting for 15 per cent of the total lithic
assemblage (Henan Provincial Museum 1981).
Six stones, three from each site, were examined in this study
(Figure 2; Table 1). Grinding stones of the Peiligang culture have
distinctive stylistic characteristics. Mopan slabs are the lower stones
and are normally made of coarse to medium-grained sandstone, elongated
in plan with a flat upper surface and four short legs; they weigh around
10-20kg (Figure 2, nos. 1-3). Many mopan show striations and pitting on
the used surface (Figure 2, no. 7), suggesting an abrading motion and
frequent pecking to maintain an effective working surface. Mobang
(literally grinding roller) are elongated upper stones (Figure 2, no.
4). They are mostly made of medium-grained sandstone, but occasionally
other raw materials such as limestone were used. Mobang show a range of
shapes in cross-section: round, oval, hemispherical and faceted.
Different degrees of wear and various grinding motions, such as
abrading, rocking and rolling have determined the shape at discard. Most
mabang have one or more flat surfaces or facets, suggesting that
abrading was the dominant action. Use-wear on the end of some mobang
indicates they have also been used as pestles (Figure 2, no. 6).
[FIGURE 1 OMITTED]
Methods of analyses
Two methods were used: use-wear analysis and residue analysis based
on the starches surviving on the stone. In each case the identifications
were achieved by comparison with a reference collection.
[FIGURE 2 OMITTED]
Use-wear analysis (Table 1; Figures 3 & 4)
Reference: the reference collection for use-wear analysis was
compiled from two sources: (1) sandstones obtained from China (Henan)
were used in an experimental study for processing cereals and acorns at
La Trobe University, and the use-wear patterns were documented; and (2)
a set of modern millstones (sandstone) at Xiazhuang village in Songxian,
Henan, China, was examined for use-wear. These tools had been used
extensively by the local people to process many different plants,
including cereals and acorns (Figure 3).
The character of wear was determined using Polyvinyl siloxane (PVS)
impressions or peels, of a type commonly used by archaeologists and
other scientists to document surface details of various materials at
high resolution (e.g. Mandikos 1998; Fullagar 2006a; cf. the use of
acetate peels by Young & Syms 1980). The peels provide negative
impressions so that pits and striations appear as raised features.
Grinding, pounding and abrasion experiments undertaken by Fullagar
(unpublished data) have utilised sandstone, granite, dolerite,
quartzite, flint and basalt stones. Materials worked include seeds (i.e.
millet, barley, rice and water lily), tubers, ochre and bone. However,
smoothing, polish and other use-wear can, to a certain extent, be
extrapolated from experiments on flaked stone (Fullagar 1991, 2006b) and
other experimental datasets (e.g. Cunnar 2007; Hamon 2008). The current
database of use-wear on ethnographic and archaeological grinding stones,
in addition to the aforesaid materials from China, also includes
extensive study of granite bedrock grinding patches (for seeds) in
north-western Australia (Fullagar & Wallis 2008); numerous seed
grinding dishes from arid regions (e.g. Fullagar 1985; Fullagar &
Field 1997) including those from museum collections; and archaeological
specimens used for processing tubers in more temperate regions of
Australia (e.g. Hall et al. 1989; Fullagar & Jones 2004).
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Based on our reference samples, stones used to process acorns and
cereals can be readily distinguished. Use-wear on sandstone grinding
stones used exclusively to pound and crush acorns produces far fewer
striations than experimental tools used for grinding seeds such as
millet and oats. Striations are not completely absent but rare, and
compared with grinding seeds, the polish and smoothing from processing
acorns is less flat and more uneven at high magnification (x 200). Small
depressions in the sandstone surface are polished and give the
appearance of a surface that has been pitted and subsequently polished,
which is likely to be what has happened.
Analysis and results: nine PVS peels from five grinding stones,
three from Shigu and two from Egou, were examined microscopically under
a compound (reflected light) microscope at magnifications of x 100 and x
200 (Figure 4). Two PVS samples from Shigu mopan (GS28) have no clear
striations visible at low magnification, although possible pitting is
present (Figure 4A). Striations seem to be present on some macroscopic
images. At high magnification striations are present but rare, and
polish development is generally low, with rare patches of more developed
polish with a reticular pattern. Striations (longitudinal) were rarely
visible at low magnification on two PVS peels taken from slab GS29
(Figure 4B). Striations were similarly rare at higher magnifications (x
100 and x 200). Polish development was low with rare reticular polish
patterns. A PVS sample from mobang R32, which has a convex surface and
an oval cross-section, had no striations visible at low magnification
(Figure 4C). At high magnification (x 100 and x 200), striations are
also rare and polish development is medium with clear reticular patterns
and smoothed surfaces. However, the polished surface is uneven with
apparently polished pits. The scarcity of striations and a polish
pattern with apparent pitting are more typical of pounding and crushing
acorns than cereals, although other plants may have been processed on
the same stone. The two PVS peels from an Egou mopan GS31 show that
longitudinal striations (furrows) are visible at low magnification, on
flat, smoothed surfaces (Figure 4D). Polish development is low-medium
with rare reticular polish visible at magnifications of x 100 and x 200.
Furrows are common also at these higher magnifications and sleeks
(ribbons of polish, see Kamminga 1979: 148) are rare. The polished
surface is uneven and patchy at high magnification (x 200) and the
furrows may be associated with manufacture or surface rejuvenation,
rather than use. The use-wear, particularly polish and striations,
indicates the grinding of plant material, although further grinding
experiments are required with local plants and other materials. Mobang
R25 with a nearly round cross-section showed transverse furrows at low
magnification (Figure 4E). At high magnification (x 100 and x 200), the
surface is smooth with medium polish development with a reticular
pattern, striations, rare sleeks and corrugated alignments. There are
similarities with the stones used for grinding seeds, although at high
magnification (x 200), the polished mobang surface appears slightly more
uneven than seed grinding stones. The corrugated alignments are unusual
and may relate to grinding and pecking during manufacture.
In general, the use-wear, particularly polish and striations,
indicates the processing of plants, although there are differences in
the wear on the Egou and Shigu mobang. The orientation of striations
appear to distinguish mobang (with transverse furrows) and mopan (with
longitudinal furrows and sometimes sleeks). The Egou mobang appears to
have less extensive polish with larger areas of unpolished depressions.
The mobang also appear to have a greater degree of polish development
than mopan from the same site. Compared with those grinding stones used
for processing cereals in our reference collection, Shigu and Egou slabs
have less polish development and fewer striations, but most have some
later stages of polish with a reticular pattern developing, as is the
case on Australian sandstone seed grinding stones (Figure 3). However,
the polished surfaces on mobang and mopan are more uneven at high
magnification (x 200), and are distinct from the smoothed striated
patches typical of experimental stone-on-stone use-wear. It is likely
that all artefacts with this pattern have been used for processing
plants of some kind.
Starch residue analysis (Tables 2 & 3; Figures 5-8)
Reference: a comparative reference collection has been compiled
with which to identify the archaeological residues. It includes acorns,
tubers, grasses and legumes (Table 2; Figure 5).
Analysis and results: dry sediment or water-extractions were
collected from 12 locations on the use-surfaces of mopan and mobang from
Shigou and Egou (Figure 2). The starch residues were extracted by heavy
liquid separation using sodium polytungstate with a specific gravity
(SG) of 2.35. Extractions were mounted on glass slides in 50 per cent
glycerol. Complete scans of the slides were undertaken using a Zeiss
Axioskop II brightfield microscope fitted with polarizing filters and
Nomarski (Differential Interference Contrast) optics. The Nomarski
method is specifically designed to enhance the contrast of unstained
specimens, thus providing good visualisation of surface features. Images
were collected using a Zeiss HRc digital camera and Zeiss Axiovision
software.
[FIGURE 5 OMITTED]
Ten of the twelve samples examined in this study yielded starch
grains. Their measurements (maximum length through the hilum) were
plotted against a range of comparative reference material for comparison
to eliminate those plants that were unlikely to contribute to the
assemblage (Figure 6; Tables 2 & 3). Subsequent identifications were
undertaken on the basis of morphological features of the starch grains.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
Shigu: four of the six Shigu samples yielded starch grains, but of
these only two samples yielded >2 grains (Table 3). GS28A and GS28B
are from the same tool, and yielded only nine grains so are combined and
considered as GS28. On the basis of this preliminary analysis, GS28
starch had dimensions in the Trapa sp. (water caltrop) range, though the
archaeological samples showed greater size variability. It is unlikely
to have been used for cereals as none of the reference material
overlapped to any degree with this sample, nor were the morphological
features observed consistent with cereals. Furthermore as only a small
number of grains were recovered from this artefact, any confident
allocation to species would be speculative and require further
consolidating studies. GS29B yielded 11 starch grains and produced an
assemblage which appears to vary markedly with the starch recovered from
GS28. Size range and morphology of this assemblage correlates to the
acorns (cf. Lithocarpus sp., Quercus sp. and Cyclobalanopsis sp.). Part
of the GS28 assemblage also falls within this size range and some of
these species may have contributed to the final starch assemblage.
Interestingly, of the small amount of starch that was identified from
this sample (GS28) most appeared to be heavily damaged, consistent with
the grinding process, making measurements and possible identification
problematic (Figure 7A & B). For the Shigu material, it is unlikely
that cereals were being processed but attributing any other species or
genera identification to the starch is not possible.
Egou: six samples were examined from the Egou site, of which five
yielded starch grains (Table 3). As with most analyses, the larger the
number of starch grains recovered the greater the possibility of
identification to genera or species for the plants being processed. In
this case, sample GS36C yielded 138 grains which is a good yield for an
archaeological sample. Samples GS31A and B were collected from the
surface of a mopan. The starch grain dimensions from GS31A fall well
within the range of acorns (Figure 6) and the morphology of the
individual starch grains are consistent with this broad grouping. Note
the grains in Figure 8A, B and C show morphologies also observed for a
range of Lithocarpus, Cyclobalanopsis and Quercus species (Figure 5G),
namely the faceting of some grains and the fissures at the hilum. Sample
GS31B had a higher median value than GS31A and may reflect that the
larger sample analysed captured a greater variability in sample sizes.
It is also consistent with a range of acorns with the higher median
value aligning it more closely to Lithocarpus. This mopan (GS31) may
have been used as a multifunctional tool as evidenced by the presence of
kidney shaped starch grains with fissures (Figure 8D), similar to those
produced by beans, e.g. Vigna sp. (Figure 5I), and unidentified faceted
grains (Figure 8F). Two samples, collected from the side (GS36B) and
bottom (GS36C) of a broken slab foot, showed great variability in starch
yields (Table 3). Sample GS36B yielded a tuber grain (Figure 8G)
consistent with the Chinese Yam (Dioscorea opposita Thunb.) (Figure 5H).
Sample GS36C produced the highest number of starch grains (Figure 8I).
The measurements and features of this starch assemblage is similar to
Cyclobalanopsis, though some very distinctive grains observed in the
reference samples were not observed in the archaeological sample. As the
starch was found mostly in clumps on this slide it is likely that the
assemblage represents one plant. Sample R25 was collected from the
elongated surface of a mobang, and yielded 24 starch grains. A range of
morphologies presented in this sample including faceted, round and
irregular shaped grains (see Figure 8L). Five of the grains are strongly
faceted (e.g. Figure 8K) and are consistent with a range of material
from the Poaceae (grass) family, in particular foxtail millet (Setaria
italica). The starch assemblages recovered from the Egou grindstones are
consistent with an acorn origin for most of the starch found there.
However, the presence of possible beans, yam and millet starch suggest
either a multifunctional use for these implements or at least the
associated presence of these plants.
Discussion
The use-wear of these grinding tools show some similarity with
grinding stones used for processing seeds in Australia (Smith, M.A.
1988; Fullagar & Field 1997). Detected differences include surface
unevenness at high magnification and the frequency of striations. For
example, compared with the grinding stones used for processing cereals
(Figure 3A-D), the Peiligang grinding stones have sustained less surface
smoothing, and have polished, pit-like depressions. The role of silica
in development of use polish on these grinding stones is unclear, and
may be less significant than processes of abrasive smoothing (cf.
Fullagar 1991). Unlike many residues from Australian grinding stones
(e.g. Fullagar et al. 2008), phytoliths have not been recovered from the
Peiligang grinding stones. While this may be due to a number of
taphonomic factors such as pH of the enclosing sediments, survivability
of the phytoliths relative to the starch, the lack of phytoliths from
the Peiligang grinding stones may reflect the fact that grass seeds were
not the main plant material being processed.
It is not surprising that millet starch was rarely found on
grinding stones. Millet needs to be dehusked, but does not have to be
ground to flour for human consumption. Experimental studies indicate
that mortar and pestle are the most effective tools for dehusking
cereals (Meurers-Balke & Luning 1999), including millet (Wang 2008).
On the other hand, acorns have to be ground and leached for human
consumption (Mason 1992). Experimental studies indicate that processing
dry and fresh acorns produces polish on grinding tools made of basalt
and sandstone, probably influenced by oil in this plant (Dubreuil 2004;
Wang 2008). The polish appears to be more developed in sandstone than in
basalt, according to Wang's study (Wang 2008). Dubreuil (2004),
like us, noted an irregular surface and that striations were rare on
processing implements used to grind acorns, but suggested that this was
a problem with highly reflected surfaces of polished basalt. We think
that the absence of striations is real in sandstone grinding stones and
that it is probably related to the cushioning effect of large kernels
and minimal contact between upper (i.e. handstone) and lower grinding
stones. There are various ways of processing acorns in different regions
in China. According to our ethnographic observation in Songxian, Henan,
the locals soak the acorns overnight and then grind them with shells
attached on a machined millstone. The millstone was used for grinding
acorns and other plants, including dehusking cereals. Its used surfaces
had patches of wear very similar to our experimental acorn grinding
experiments (Figure 3F), but also had other patches with the relatively
flat heavily striated polish observed on seed grinders.
The starch assemblages in this study appear to be predominantly of
acorns, while starch grains identifiable to yam, bean and millet
occurred much less frequently. This finding is consistent with several
functional studies of grinding stones dating to the early and middle
Holocene periods in other parts of China (Wang 2008; Yang, X. et al.
2009; Liu, L. et al. in press). According to our ethnographic study in
Zhejiang, China, people still make acorn jelly today. Acorn flour is
mixed with starch from other plants, such as sweet potato, in order to
improve the flavour and texture of the product. A similar way of making
acorn foods by combining various plant flours may have already been
practised in prehistory.
The use-wear variability and the recovered residues suggest that
the grinding stones from Peiligang sites are likely to have been used as
multifunctional tools, and they may not have been limited to processing
foodstuffs like seeds and nuts. Other artefact materials that could
cause polish on grinding tools include medicines, wood, bone, ochre,
shell and skin or hides (Cunnar 2007). Further experimental study on
use-wear and residue patterns from more grinding stones is warranted to
understand the range of the functions of these tools.
The results from this study shed light on two important problems in
prehistory: the function of grinding stones in the Peiligang culture of
the Middle Yellow River region and Early Neolithic subsistence
strategies. The current study indicates that a wide range of starchy
plants was exploited by the Peiligang culture people, including acorns
from various species/genera, millet, yam and bean. Chinese yam
(Dioscorea opposita), shanyao, is an indigenous plant, distributed
widely in both north and south China today. Wild yam grows on mountain
slopes, near forests in river valleys, near streams, and in shrubs and
grasses (Flora of China 1985: 103-105). Henan is known to produce high
quality yams, a plant which has been used not only as food but also as
medicine. Starch analysis has the advantage of identifying tuber
remains, which have rarely been found in carbonised forms in
archaeological contexts. It is not currently known when yam was first
domesticated, though the Peiligang people probably exploited wild yam.
In China seeds of beans have been uncovered from many Neolithic sites.
Starch identifiable to species belonging to the genus of Vigna have been
found on grinding stones from the Middle Neolithic site of Shangzhai in
Beijing (Yang, X. et al. 2009). Starch grains from several species of
Vigna show similar morphology; and in this study it has not been
possible to identify the bean starch to species.
The earliest macrobotanical remains of domesticated millet have
been found in the Liao and Yellow rivers in China, including the Henan
region, all dating to the late seventh and early sixth millennia BC
(Zhao, Z. 2004; Liu, C. 2006; Lee et al. 2007). Our starch analysis
confirms that millet was indeed a part of the diet of the Peiligang
people. It is important to point out that starches present on grinding
stones examined here do not represent the whole range of plants used by
the Peiligang people. Soybean, for example, which have been found as
seeds in flotation samples from many Early Neolithic sites, is absent in
our starch assemblages. This is because soybean contains a low
proportion of starch, and its starch grains are very small in size and
lack diagnostic characteristics. Also, the composition of plants shown
in the recovered starch assemblages does not necessarily reflect the
relative proportions of different plants exploited by the Peiligang
people. Except for acorns, almost all plants identified, including
millet, beans and tubers, can be consumed after being boiled in pottery,
without being processed into flours with grinding stones. Nevertheless,
given that grinding stones account for significant proportions in tool
assemblages of the Peiligang culture, it is reasonable to argue that
acorn was an important staple food used by the Peiligang people.
Conclusion
This pilot study of Peiligang culture grinding stone tools greatly
improves our understanding of the Early Neolithic subsistence economy in
the Yellow River region by providing direct evidence of economically
important plant foods. Peiligang grinding stones were used for (but not
limited to) processing a variety of plant foods, predominantly acorns,
followed by millet and beans, among others. Several studies have
previously pointed out that the Early Neolithic Peiligang culture was
characterised by wide-spectrum subsistence economy (Henan Institute
1999: 898; Lee et al. 2007), and our findings certainly support this
proposition.
The presence of grinding stones in the Early Neolithic sites should
no longer be used as an indicator of intensive agriculture based on
cereals; but, instead, it is more likely to suggest a wide-spectrum
subsistence economy, with a particular focus on acorn exploitation.
Acorns are known to have been used as a staple by many
hunting-gathering peoples throughout history in many parts of the world,
such as North America, Japan and Europe (Mason 1992, 1996; Bettinger et
al. 1997; Crawford 1998; Anderson 2005; Matsui & Kanehara 2006).
However, it is a new concept in Chinese archaeology that the Early
Neolithic Peiligang people, who were previously thought to have been
farmers, also relied on acorns as a staple food. This finding challenges
the traditional concept of the Neolithic Revolution in China, which has
often been viewed as a package of inter-related developments, including
cereal farming (millet and rice), animal domestication (pig and dog) and
sedentary settlement. On the contrary, cereal farming, although
practised, did not play a significant role in the subsistence economy
until all the other components of Neolithic culture were already in
place.
Acknowledgements
We thank Zhekun Zhou, Zhijun Zhao, Yaqin Hu, Yonggang Liu, Ming
Wei, Gyoung-Ah Lee, Yanfeng Hou and Duncan Jones for their assistance.
We are grateful to the constructive comments provided by Dolores Piperno
and Geoffrey Cunnar. The authors acknowledge the facilities as well as
scientific and technical assistance from the staffin the Australian Key
Centre for Microscopy and Microanalysis Research Facility (AMMRF) and at
the Australian Key Centre for Microscopy and Microanalysis at the
University of Sydney. The project was funded by an Australian Research
Council Discovery Grant (DP0450025), La Trobe University and the
University of Sydney.
Received: 5 January 2010; Accepted: 1 March 2010; Revised: 29 April
2010
References
AN, Z. 1989. Prehistoric agriculture in China, in D.R. Harris &
G.C. Hillman (ed.) Foraging and farming: the evolution of plant
exploitation: 643-50. London: Unwin Hyman.
ANDERSON, M.K. 2005. Tending the wild: Native American knowledge
and management of California's natural resources. Berkley (CA):
University of California Press.
BELLWOOD, P. 2005. First farmers: the origins of agricultural
societies. Oxford: Blackwell.
BETTINGER, R., R. MALHI & H. MCCARTHY. 1997. Central place
models of acorn and mussell processing. Journal of Archaeological
Science 24: 887-99.
CHANG, K-C. 1986. Archaeology of ancient China. New Haven (CT):
Yale University Press.
CHEN, W. 2002. Nongye Kaogu. Beijing: Wenwu.
CRAWFORD, G.W. 1998. Scheduling and sedentism in the prehistory of
northern Japan, in T.R. Rocek & O. Bar-Yosef (ed.) Seasonality and
sedentism: archaeological perspectives from Old and New Worm sites
(Peabody Museum Bulletin 6): 109-28. Cambridge (MA): Peabody Museum of
Archaeology and Ethnology, Harvard University.
CUNNAR, G. 2007. The production and use of stone tools at the
Longshan period site of Liangchengzhn, China. New Haven (CT): Yale
University Press.
DUBREUIL, L. 2004. Long-term trends in Natufian subsistence: a
usewear analysis of ground stone tools. Journal of Archaeological
Science 31: 1613-29. Flora of China Editorial Committee (ed.) 1985.
Flora of China, Volume 16.1. Beijing: Kexue Press.
FULLAGAR, R. 1985. Use-polish on a muller from Ooraminna Rock Hole,
central Australia. Report prepared for M. Smith, Museums and Art
Galleries of the Northern Territory.
--1991. The role of silica in polish formation. Journal of
Archaeological Science 18: 1-25.
--2006a. Residues and usewear, in J. Balme & A. Paterson (ed.)
Archaeology in practice: a student guide to archaeological analyses:
207-34. Oxford: Blackwell.
--2006b. Starch grains, stone tools and modern hominin behaviour,
in S. Ulm & I. Lilley (ed.) An archaeological life: papers in honour
of Jay Hall (Aboriginal and Tortes Strait Islander Studies Unit Research
Report series 7): 191-202. Brisbane: Aboriginal and Torres Strait
Islander Studies Unit, The University of Queensland.
FULLAGAR, R. & J. FIELD. 1997. Pleistocene seed-grinding
implements from the Australian arid zone. Antiquity 71: 300-307.
FULLAGAR, R. & R. JONES. 2004. Usewear and residue analysis of
stone artefacts from the enclosed chamber, Rocky Cape, Tasmania.
Archaeology in Oceania 39: 79-93.
FULLAGAR, R. & L. WALLIS. 2008. Usewear and residues on
Aboriginal grinding patches from the Fortescue Metals Group Rail
Construction Corridor Project, Pilbara. WAA Confidential Report prepared
for Fortescue Metals Group Pty Ltd.
FULLAGAR, R., J. FIELD & L. KEALHOFER. 2008. Grinding stones
and seeds of change: starch and phytoliths as evidence of plant food
processing, in Y.M. Rowan & J.R. Ebeling (ed.) New approaches to old
stones: recent studies of ground stone artefacts: 159-72. London:
Equinox.
HALL, J., S. HIGGINS & R. FULLAGAR. 1989. Plant residues on
stone tools, in W. Beck, A. Clarke & L. Head (ed.) Plants in
Australian archaeology (Tempus 1): 136-60. St. Lucia: Anthropology
Museum, University of Queensland.
HAMON, C. 2008. Functional analysis of stone grinding and polishing
tools from the earliest Neolithic of north-western Europe. Journal of
Archaeological Science 25: 1502-20.
Henan Institute of Cultural Relics. 1987. Changge Shigu yizhi fajue
baogao [Excavation report of the Shigu site in Changge]. Huaxia Kaogu 1:
3-125.
--1999. Wuyang Jiahu. Beijing: Kexue Press.
Henan Provincial Museum. 1981. Henan Mixian Egou Beigang xinshiqi
shidai yizhi [The Neolithic site at Egou Beigang in Mixian, Henan].
Kaoguxuejigan 1: 1-26, 48.
HIGHAM, C. 2005. East Asian agriculture and its impact, in C.
Scarre (ed.) The human past: 234-63. London: Thames & Hudson.
KAMINGA, J. 1979. The nature of use-polish and abrasive smoothing,
in B. Hayden (ed.) Lithic usewear analysis: 143-57. London: Academic.
LEE, G.-A., G.W. CRAWFORD, L. LIU & X. CHEN. 2007. Plants and
people from the early Neolithic to Shang periods in north China.
Proceedings of the National Academy of Sciences of the United States of
America 104(3): 1087-92.
LIU, C. 2006. Dadiwan yizhi zhiwu yicun jianding baogao
[Identification of plant remains from the Dadiwan site], in Gansu
Institute of Cultural Relics and Archaeology (ed.) Qin'an Dadiwan:
914-16. Beijing: Wenwu Press.
LIU, L. 2004. The Chinese Neolithic: trajectories to early states.
Cambridge: Cambridge University Press.
--2008. Zhongguo shiqian de nianmo shiqi, jianguo caiji, dingju ji
nongye qiyuan [Grinding stones, nut collection and origins of sedentism
and agriculture in prehistoric China], in Editorial Board of
'Essays for the Celebration of Ping-ti Ho's Ninetieth
Birthday' (ed.) He Ping-ti Xiansheng Jiushi Shouqing Wenji: 105-32.
Xi'an: Sanqin Press.
LIU, L., J. FIELD, A. WEISSKOPF, J. WEBB, L. JIANG, H.
WANG & X. CHEN. In press. The exploitation of acorn and rice in
early Holocene Lower Yangzi River, China. Acta Anthropologica Sinica.
LU, T.L.-D. 1999. The transition from foraging to farming and the
origin of agriculture in China (British Archaeological Report
International series 774) Oxford: J. & E. Hedges.
LUO, Y. 2007. Zhongguo Gudai Jiazhu Yanjiu [Domesticated pigs in
ancient China]. Unpublished PhD dissertation, Chinese Academy of Social
Sciences.
MANDIKOS, M.N. 1998. Polyvinyl siloxane impression materials: an
update on clinical use. Australian Dental Journal 43(6): 428-34.
MASON, S.L.R. 1992. Acorns in human subsistence. Unpublished PhD
dissertation, University College London.
--1996. Acomutopia? Determining the role of acorns in past human
subsistence, in J. Wilkins, D. Harvey & M. Dobson (ed.) Food in
antiquity: 12-24. Exeter: University of Exeter Press.
MATSUI, A. & M. KANEHARA. 2006. The question of prehistoric
plant husbandry during the Jomon period in Japan. World Archaeology
38(2): 259-73.
MEURERS-BALKE, J. & J. LONING. 1999. Some aspects and
experiments concerning the processing of glume wheats, in P. Anderson
(ed.) Prehistory of agriculture: new experimental and ethnographic
approaches (Monograph of the Institute of Archaeology, University of
California 40): 238-53. Los Angeles (CA): Institute of Archaeology,
University of California.
SHI, X. 1992. The discovery of the pre-Yangshao culture and its
significance, in C.M. Aikens & S.N. Rhee (ed.) Pacific north-east
Asia in prehistory: hunter-fisher-gatherers, farmers, and sociopolitical
elites: 125-32. Pullman (WA): Washington State University Press.
SMITH, B.D. 1995. The emergence of agriculture. New York:
Scientific American Library.
SMITH, M.A. 1988. Central Australian seed grinding implements and
Pleistocene grindstones, in B. Meehan & R. Jones (ed.) Archaeology
with ethnography: an Australian perspective (Occasional Papers in
Prehistory 15): 94-108. Canberra: Department of Prehistory, Research
School of Pacific Studies, The Australian National University.
SONG, Z. 1997. Shiqian shiwu de jiagong jishu--lun moju yu chujiu
de qiyuan [Food processing techniques in prehistory--on the origins of
grinding tools, mortars and pestles]. Nongye Kaogu 3: 187-95.
UNDERHILL, A. 1997. Current issues in Chinese Neolithic
archaeology. Journal of WorM Prehistory 11: 103-29.
WANG, Q. 2008. Haidai Diqu Shiqian Shiqi Mopan, Mobang Yanjiu
[Functional analyses of prehistoric stone slabs and mullers in the
Haidai region]. Unpublished PhD dissertation, Shandong University.
Wu, J'A. 1986. Shiqi shidai de shimopan [Stone grinding slabs
in the Neolithic period]. Shiqian Yanjiu 1(2): 46-54.
YAN, W. 1992. Origins of agriculture and animal husbandry in China,
in C.M. Aikens & S.N. Rhee (ed.) Pacific north-east Asia in
prehistory: hunter-fisher-gatherers, farmers, and sociopolitical elites:
113-24. Pullman (WA): Washington State University Press.
YANG, X., J. YU, H. LU, T. CUI, J. GUO, X. DIAO, Z. KONG, C. LIU
& Q. GE. 2009. Starch grain analysis reveals function of grinding
stone tools at Shangzhai site, Beijing. Science in China Series D: Earth
Sciences 52(8): 103-222.
YOUNG, H.R. & E.L. SYMS. 1980. The use of acetate peels in
lithic analysis. Archaeometry 22(2): 205-208.
ZHAO, S. 2005. Shimopan, shimobang shi guwu jiagong gongju ma?
[Were stone slabs and mullers used for processing cereals?]. Nongye
Kaogu 3: 134-47.
ZHAO, Z. 2004. Cong Xinglonggou yizhi fuxuan jieguo tan Zhongguo
beifang hanzuo nongye qiyuan wenti [The origins of dry-land agriculture
in north China based on the flotation results from the Xinglonggou
site], in Nanjing Normal University (ed.) Dongya Kaogu: 188-99. Beijing:
Wenwu Press.
Li Liu (1), Judith Field (2), Richard Fullagar (3), Sheahan Bestel
(4), Xingcan Chen (5) & Xiaolin Ma (6)
(1) Archaeology Program, School of Historical and European Studies,
La Trobe University, Melbourne, VIC 3086, Australia
(2) Australian Centre for Microscopy & Microanalysis, The
University of Sydney, Sydney, NSW 2006, Australia
(3) Scarp Archaeology, P.O. Box 7241, South Sydney Hub, NSW 2015,
Australia; Centre for Archaeological Science (CAS), School of Earth and
Environmental Sciences, University of Wollongong, Wollongong, NSW 2522,
Australia
(4) School of Geography and Environmental Science, Monash
University, Clayton, VIC 3800, Australia
(5) Institute of Archaeology, Chinese Academy of Social Sciences,
27 Wangfujing Dajie, Beijing, P.R. China 100710
(6) Henan Provincial Institute of Cultural Relics and Archaeology,
No. 9 Longhaibeisan Street, Zhengzhou, P.R. China 450000
Table 1. Descriptions of grinding stones that were examined in
this study from the sites of Egou and Shigu.
Artefact/ Starch PVS Tool
sample no. sample no. sample no. type
EGT4M26/ GS31 GS31A GS31-1 slab
GS31B GS31-2
EGT18M40/ GS36 GS36A (top) -- foot from a
GS36b (side) slab
GS36c (foot)
EGM55:6/ R25 R25 R25 handstone
SG215/ GS28 GS28A GS28-1 slab
GS28B GS28-2
SGAT62M86:2/ GS29 GS29A GS29-1 slab
GS29B GS29-2
SGT46:1/ R32 R32 R32A R32B handstone
Artefact/ Stone raw Wt Use surface
sample no. material (kg) observation with
naked eye
EGT4M26/ GS31 medium-coarse 10.13 pitting and
grained brown striations
sandstone
EGT18M40/ GS36 medium-coarse -- --
grained grey
sandstone
EGM55:6/ R25 medium-grained grey -- polishing and
sandstone striations
SG215/ GS28 medium-coarse gray 10.18 pitting and
sandstone striations
SGAT62M86:2/ GS29 medium-coarse gray 10.09 rough surface,
sandstone little pitting
SGT46:1/ R32 medium-grained gray -- polishing and light
sandstone striations
Artefact/ No. used
sample no. surfaces & shape
EGT4M26/ GS31 1, flat-concave
EGT18M40/ GS36 1, flat
EGM55:6/ R25 multiple, flat-sphere
SG215/ GS28 1, flat-concave
SGAT62M86:2/ GS29 1, flat-concave
SGT46:1/ R32 multiple, round in
cross-section
Table 2. Measurement data for comparative reference material
used in this study.
Species N Mean SD Median
Lithocarpus glaber 189 12.93 3.25 12.36
L. fenestratus 135 13.05 5.24 11.54
Quercus acutissima 101 10.9 4.74 9.94
Q. aliena 123 10.85 4.34 10.27
Cyclobalanopsis fleuryi 117 15.57 4.05 15.82
C nubium 130 11.16 3.03 10.52
Setaria italica 124 10.58 2.73 10.01
Panicum miliaceum 131 7.90 1.76 7.73
Trap a sp. 117 25.07 5.11 26.00
Fagopyrum sp. 109 7.72 2.47 7.40
Vigna sp. 102 20.56 8.85 25.27
Species Range ([micro]m)
Lithocarpus glaber 6.77-24.13
L. fenestratus 5.67-38.63
Quercus acutissima 3.31-27.53
Q. aliena 2.88-22.11
Cyclobalanopsis fleuryi 7.90-25.14
C. nubium 5.92-19.15
Setaria italica 5.83-20.54
Panicum miliaceum 3.93-13.07
Trap a sp. 11.18-40.66
Fagopyrum sp. 2.70-14.48
Vigna sp. 5.27-42.85
Table 3. Results of starch analysis of samples from grinding
stones from Shigu and Egou, showing mean and maximum lengths
of grain.
Site and sample no. Tool type N Mean SD Median
Shigu
GS28 (A,B) slab 9 26.04 10.46 23.7
GS29A slab 0 -- -- --
GS29B slab 11 15.57 4.43 19.66
R32A handstone 2 15.72 -- --
R32B handstone 0 -- -- --
Egou
GS31A slab 42 10.88 3.89 10.08
GS31B slab 97 13.06 6.29 11.87
GS36A slab 0 0 -- --
GS36B slab 17 21.00 7.75 21.24
GS36C slab 138 16.00 5.59 14.81
R25 handstone 24 12.55 2.88 11.86
Site and sample no. Range (pm)
Shigu
GS28 (A,B) 13.14-45.13
GS29A --
GS29B 9.82-22.81
R32A 12.8-18.64
R32B --
Egou
GS31A 5.82-20.94
GS31B 2.47-38.68
GS36A --
GS36B 11.2-35.78
GS36C 5.65-32.30
R25 7.24-18.47
