ArticlePDF Available

Prehistoric evolution of the dualistic structure mixed rice and millet farming in China


Abstract and Figures

Compared with the monistic structure of crop agriculture in Southwest Asia and Mesoamerica, agriculture in ancient China reflects the characteristics of a dualistic structure with millet in the north and rice in the south. It is argued that the rice and millet farming modes were mutually exchanged during their development and formed a vast region of mixed farming. However, the time and place of its origin, the routes of dissemination, and the development patterns and possible influence factors of mixed farming remain unclear. This study systematically collected information from 804 sites with millet and rice records and detailed floatation results from 78 mixed farming sites in prehistoric China. Three north?south communication corridors are identified between the upper, middle and lower Yellow and Yangtze River Valleys that began around 5500 BP, 8400 BP and 4600 BP, respectively. Cultural communication accompanied by human migration and the unique natural environment of loess and East Asia monsoons fa
Content may be subject to copyright.
The Holocene
1 –14
© The Author(s) 2017
Reprints and permissions:
DOI: 10.1177/0959683617708455
The origin and spread of agriculture provided a stable economic
base for the gradual progress of human society and the establish-
ment of ancient civilization (Chen et al., 2015a; Diamond, 2002;
Iriarte et al., 2004). China has long been accepted as one of the
three independent centres of origin for crop agriculture in the
world (Bellwood, 2005; Diamond and Bellwood, 2003; Harlan,
1971; Vavilov and Freier, 1951). In contrast to the single type of
crop agriculture in Southwest Asia (Lev-Yadun et al., 2000; Will-
cox, 2013) and Mesoamerica (Piperno et al., 2000, 2009), the
unique dualistic agricultural structure of millet in the north of
China and rice in the south was developed with the support of
various geographical climatic and land conditions, abundant
archaeological cultures and widely distributed wild ancestors of
these crops. However, the questions of when, where and how
these two types of farming modes interacted during the fluctuat-
ing climate in the Holocene remain controversial.
With regard to the relationship between millet and rice farm-
ing, the key dispute involves single or multiple centres of crop
origin. Many Chinese authors tend to consider millet and rice
agriculture as two types of agriculture that developed indepen-
dently (Li, 1970; Yan, 1997; Figure 1a). Additionally, a third cen-
tre of tropical agriculture has recently been proposed (Zhao,
2011b; Figure 1b). However, the view of single foci and farmer
dispersal is more prominent among international scholars. Based
on the earliest age of excavated crops, different scholars support
either the ‘rice-and-then-millet model’ (Bellwood, 2005; Cohen,
2011) or the ‘millet-and-then-rice model’ (Bar-Yosef, 2011). In
further research on the origin of millet and rice, the beginning of
millet and rice domestication has been traced to approximately
10,000–11,000 BP (Jiang and Liu, 2006; Lu et al., 2009a; Wu
et al., 2014; Yang et al., 2012a). Evidence increasingly supports
the view that millet and rice were domesticated separately and
connected with each other during their development and that a
vast region of mixed farming formed in Neolithic China (Fuller
et al., 2014; Qin, 2012).
Prehistoric evolution of the
dualistic structure mixed rice
and millet farming in China
Keyang He,1,2 Houyuan Lu,1,2,3 Jianping Zhang,1,3
Can Wang1,2 and Xiujia Huan1,2
Compared with the monistic structure of crop agriculture in Southwest Asia and Mesoamerica, agriculture in ancient China reflects the characteristics
of a dualistic structure with millet in the north and rice in the south. It is argued that the rice and millet farming modes were mutually exchanged during
their development and formed a vast region of mixed farming. However, the time and place of its origin, the routes of dissemination, and the development
patterns and possible influence factors of mixed farming remain unclear. This study systematically collected information from 804 sites with millet and
rice records and detailed floatation results from 78 mixed farming sites in prehistoric China. Three north–south communication corridors are identified
between the upper, middle and lower Yellow and Yangtze River Valleys that began around 5500 BP, 8400 BP and 4600 BP, respectively. Cultural
communication accompanied by human migration and the unique natural environment of loess and East Asia monsoons facilitated the interaction between
millet and rice farming through these corridors. As a comprehensive reflection of the interaction between millet and rice farming, the crop structure of
the four core mixed farming regions is in a continual process of adjustment, with the selection of foxtail millet in the southward spread of millet agriculture
and temperate Oryza japonica in the northern spread of rice agriculture.
communication corridors, crop selection, East Asian monsoon, loess, mixed farming, prehistoric China
Received 17 January 2017; revised manuscript accepted 8 April 2017
Key Laboratory of Cenozoic Geology and Environment, Institute of
Geology and Geophysics, Chinese Academy of Sciences, China
2University of Chinese Academy of Sciences, China
Center for Excellence in Tibetan Plateau Earth Science, Chinese
Academy of Sciences, China
Corresponding authors:
Keyang He, Key Laboratory of Cenozoic Geology and Environment,
Institute of Geology and Geophysics, Chinese Academy of Sciences,
Beijing 100029, China.
Houyuan Lu, Key Laboratory of Cenozoic Geology and Environment,
Institute of Geology and Geophysics, Chinese Academy of Sciences,
Beijing 100029, China.
708455HOL0010.1177/0959683617708455The HoloceneHe et al.
Research paper
2 The Holocene 00(0)
Mixed farming is not an isolated mode of agriculture but is the
result of the spread of millet and rice farming. Many studies have
been conducted to draw the route of the spread of millet across
Eurasia in prehistoric food globalization (Frachetti et al., 2010;
Hunt et al., 2008; Jones et al., 2011; Spengler et al., 2014; Stevens
et al., 2016) and the spread of rice in East Asia linked to hypo-
thetical language family dispersal models (Bellwood, 2011;
Fuller, 2011; Fuller et al., 2010; Zhang and Huang, 2010). How-
ever, the interaction between these two farming modes in China
has received little attention.
The conventional viewpoint is that millet and rice farming are
divided along the boundary of the Qinling Mountains and Huaihe
River (Yan, 1987; Zhao, 2011b), although some scholars have
discovered that they overlap in the Huaihe Valley (Yan, 1997;
Zhang et al., 1994; Figure 1). Because of advances in the methods
of archaeobotany, increasing details of mixed farming have
recently been revealed that show that it extended far beyond the
range of the Huaihe Valley. Modes of mixed farming have been
found in Shandong Province (Crawford et al., 2005; Jin et al.,
2014, 2016; Zhao, 2006), the middle and upper reaches of Huaihe
Valley (Yang et al., 2016b, 2016c; Zhang et al., 2012), Yiluo Val-
ley (Lee et al., 2007), Ying Valley (Zhang et al., 2010a), Guan-
zhong Basin (Zhang et al., 2010b), Hanshui Valley (Deng et al.,
2015; Nasu et al., 2011; Weisskopf et al., 2015a), Chengdu Plain
(Guedes, 2011; Guedes et al., 2013) and Yungui Plateau (Li et al.,
2016). In addition, accompanied by the spread of millet and rice
farming out of China simultaneously or asynchronously, mixed
farming existed not only in Mainland China but also in Taiwan
region (Hung and Carson, 2014), Korea (Crawford and Lee,
2003; Lee, 2011), Japan (Crawford, 2011; Nasu and Momohara,
2016) and Thailand (Castillo, 2011; Weber et al., 2010).
The study of mixed farming can be divided into two stages. In
the first stage, only scattered crop remains were found acciden-
tally in archaeological sites. Most of the early studies on mixed
farming focused on the Yellow and Huaihe River Valleys and pro-
vided descriptive summaries of scattered sites with both rice and
millet (Wang and Xu, 2003; Zhang et al., 1994). In the second
stage, with the introduction of new archaeobotanical methods
such as floatation and phytolith analysis, crop remains were sys-
tematically collected and quantitatively analysed. However, stud-
ies of single or several mixed farming sites are confined to
regional and discontinuous discussions (Guedes et al., 2013; Jin
et al., 2014; Nasu et al., 2011; Zhang et al., 2012). To date, spatio-
temporal systematization and quantification studies of the mixed
farming mode on a national scale are absent. Based on the collec-
tion of mixed farming archaeological sites and relevant published
floatation results, we attempt to systematically map the spatio-
temporal distribution of mixed farming sites and the path of prop-
agation, quantitatively reconstruct the evolution processes of crop
structure in the main mixed farming region and comprehensively
explore the factors influencing the mixed farming mode.
Archaeology background
‘Prehistory’ in this paper indicates the Neolithic Age and the
Bronze Age in China ranging from 12,000 BP to 2221 BP. Many
schemes have been proposed to divide the Neolithic Age and sys-
temically integrate different archaeological cultures (Chang,
1964; Su and Yin, 1981; Yan, 1987). The spatio-temporal frame-
work applied in this article is mainly based on the results of
Yan (1997) and the Neolithic Volume of Chinese Archaeology
(2010). The Neolithic chronology project is divided into five
stages, including the Pre-Peiligang Period (1200–9000 BP), the
Peiligang Period (9000–7000 BP), the Early Yangshao Period
(7000–6000 BP), the Late Yangshao Period (6000–5000 BP) and
the Longshan Period (5000–4000 BP). The spatial distribution of
Chinese Neolithic cultural communities is summarized into eight
regions, including the Yanliao Region, the Haidai Region, the
Jiangzhe Region, the Yanbei Region, the Central Plains Region,
the Lianghu Region, the Ganqing Region and the Bashu Region
(Figure 2a; Yan, 1987). Four of these eight regions were related to
mixed farming (Figure 2a, regions in yellow shadow) according
to previous studies. Their cultural sequences are shown in Figure
2b, which shows that the Hanshui Valley and the Chengdu Plain
are part of the Lianghu Region and the Bashu Region, respec-
tively. The Bronze Age in China is divided according to the latest
results of the Xia–Shang–Zhou Chronology Projects (Li, 2002);
the beginnings of the Xia, Shang and Zhou Dynasties are 2070,
1600 and 1046 BC, respectively. Prehistory ends with the union
of China by the Qin dynasty in 221 BC.
Materials and methods
The concept of mixed farming of rice and millet was proposed early
in the 1990s (Zhang et al., 1994), but the definition of mixed
farming remains unclear. It is not the same as mixed cropping in
agronomy, in which two crops are cultivated in the same field
simultaneously (Dong and Shen, 2000); rather, the two farming
Figure 1. Division and evolution of agriculture in Neolithic China according to the scheme proposed by (a) Yan (1997) and (b) Zhao (2011b).
He et al. 3
systems exist in the same site or region synchronously. The discov-
ery of fields is the most direct evidence of mixed farming, but only
a few paddy fields have been found in the middle and lower Yang-
tze River Valley (Fuller and Qin, 2009; Hu et al., 2013; Nasu et al.,
2011; Weisskopt et al., 2015b) and Shandong Province (Jin et al.,
2007), and no dry fields have been excavated because of limited
preservation conditions and excavation techniques. Therefore,
mixed farming is identified by the standard of crop materials exca-
vated, including charred seed (Crawford et al., 2005; Liu and Kong,
2004), phytolith and starch grains of crops (Lu et al., 1997, 2009b;
Yang et al., 2012b; Zhao et al., 1998) and stable isotopic analysis
of human bones (Barton et al., 2009; McGovern et al., 2004;
Pechenkina et al., 2005).
The materials used in this paper can be divided into two cate-
gories. First, archaeological sites with millet or rice or those with
millet and rice excavated simultaneously and reported in docu-
ments were gathered, for a total of 804 sites, including 462 sites
of millet, 482 sites of rice and 155 sites of millet and rice. Infor-
mation on the location, age, and cultural period for each of the
155 mixed farming sites was systematically collected from origi-
nal material published and modified into a uniform data format.
Second, to quantitatively investigate the changes in crop struc-
ture in different regions, two screening criteria were applied to the
database of mixed farming sites. First, only sites with complete
published floatation results were selected; this selection resulted
in 98 sites. Second, as most of the 98 floatation sites are distrib-
uted in four cultural regions (Haidai Region, Central Plains,
Hanshui Valley and Chengdu Plain), only the 78 sites within these
regions were selected for the analysis of mixed farming struc-
tures. Many cultural stages may be included in a single site and
are partitioned based on archaeological culture. Information on
sites with mixed farming and the floatation results of crops are
given in the Supplementary Table (available online).
Information on the location, age and cultural period of the 155
mixed farming sites was systematically collected on the basis of
original published materials. The location of each site is given in
a unified form of longitude and latitude. The age of each site is the
median of the cultural period considering the start and end dates
or the median considering the radiocarbon dating results, where
available. The cultural periods are described as the original pub-
lished material with little adjustment. The floatation results of 78
mixed farming sites are recounted with uniform standards to dif-
ferent sites. Pieces of crop seeds are not taken into account, and
the different parts of the crop seeds are added up to refer to the
concept of the ‘number of identified specialness taxon (NISP)’ in
zooarchaeology (Grayson, 1979, 2014). To eliminate the differ-
ence in crop seed numbers among sites, crop seed numbers are
converted to percentage first and then summed and averaged
rather than using the sum of the original crop seed numbers
among sites of the same cultural period when combining the floa-
tation results of different sites according to the cultural periods.
Spatio-temporal evolution of the distribution of mixed
farming sites
A vast zone of rice–millet mixed farming exists in prehistoric
China. The northmost–eastmost site is Gaolizhai (122°29′E,
39°32′N) in Dalian, Liaoning Province; the southmost–westmost
site is Shifodong (99°43′E, 23°37′N) in Gengma, Yunnan Prov-
ince; and the highest altitude sites are Jiaoridang in Xunhua,
Qinghai Province and Haimenkou in Jianchuan, Yunnan Province
(approximately 2200 m; Figure 4). Most of the sites are located
between the Yellow and Yangtze Rivers within the latitude of
30°N–37°N (Figure 3) and can be divided into three subareas
(Figure 4, regions divided by red dotted line) based on the natural
geographical distribution and archaeological cultural system. The
western subarea consists of the Sichuan Basin and the Yungui Pla-
teau in the upper Yangtze River; the middle subarea includes the
Guanzhong Basin and the Central Plains in the middle Yellow
River and the Hanshui Valley in the middle Yangtze River; and the
eastern subarea is the Haidai Region in the lower Yellow River.
Based on the distribution density of mixed farming sites and
the Neolithic cultural communities, four core mixed farming
regions have been highlighted (Figure 4, regions circled by black
dotted line). Mixed farming and single farming may coexist alter-
natively in the same region and period, and proportion of the mix-
ing agriculture sites to all farming sites in four core regions
exhibits diverse patterns during different periods. In the Haidai
Region, mixed farming sites account for 50% of all farming sites
during the Peiligang period, and the proportion declines to 16.7%
and 14.3% during the Early and Late Yangshao Period, respec-
tively, and then increases to 62.2% and 65% during the Longshan
Period and Bronze Age, respectively. In the Central Plains, mixed
farming sites account for 23% during the Peiligang Period and
then increase and keep at a relatively stable level of about 50%,
Figure 2. The spatial distribution and temporal sequence of archaeological communities in Neolithic China. (a) Eight archaeological
communities divided in Yan (1997); the four related to mixed farming are coloured in yellow. (b) The cultural sequence in four mixed farming
regions of Haidai Region, Central Plains, Hanshui Valley and Chengdu Plain.
4 The Holocene 00(0)
Figure 4. Distribution of mixed farming sites during different cultural periods and their partitioning into subareas and main regions. I, The west region;
II, the middle region; and III, the east region. A, Haidai Region; B, Central Plains; C, Hanshui Valley; and D, Chengdu Plain. Bar charts aside these regions
illustrate the proportion of mixed farming sites to all farming sites within these regions through the prehistory periods, respectively. Key sites of mixed
farming sites mentioned in the paper: 1, Gaolizhai (3364 ± 86 BP); 2, Xihe (8070–7900 BP); 3, Yuezhuang (8060–7750 BP); 4, Dongjiaqiao (2770–2255
BP); 5, Bianjiashan (4500–4300 BP); 6, Baligang (8500–8100 BP); 7, Chengtoushan (6497–6021 BP); 8, Jiaoridang (2547 ± 185 BP); 9, Guiyuanqiao
(4600–4300 BP); 10, Baodun (4500–3700 BP); 11, Haimenkou (3194 ± 121 BP); 12, Shifodong (3181 ± 69 BP); and 13, Nanguanli (4700–4200 BP).
Figure 3. Spatial distribution of rice, millet and mixed farming sites with a boundary of rice at the northernmost and millet at the
southernmost. Possible centres of agriculture: 1, Xinglongwa Culture; 2, Houli Culture; 3, Cishan Culture; 4, Peiligang Culture; 5, Laoguantai
Culture; 6, The Lower Yangtze River; and 7, The Middle Yangtze River.
He et al. 5
43.6% and 48.9% during the Early and Late Yangshao Period and
Longshan Period, respectively, and decreases to about 31.6% in
the Bronze Age. In Hanshui Valley, the ratio of mixed farming
sites to all farming sites keeps an ascending trend from 21.7% to
34.6% and then to 76.9% from the Late Yangshao Period to Long-
shan Period and then to Bronze Age, respectively. By contrast, the
ratio of mixed farming sites to all farming sites keeps a descend-
ing trend from 100% to 68.4% from the Longshan Period to
Bronze age in the Chengdu Plains, respectively.
The evolution of mixed farming in China can be divided into
five stages corresponding to the archaeological ages (Figure 4). The
mixed farming region first emerged in the middle and lower Yellow
River Valley during the Peiligang Period (9000–7000 BP) and was
distributed along the northern slope of Taiyi Mountain in the Haidai
Region and around Songshan Mountain and the Nanyang Basin of
the Central Plains. During the Early Yangshao Period (7000–6000
BP), the mixed farming region began to expand, entered the Guan-
zhong Region along the Weihe River and moved south into Liyang
Plain along the Hanshui and Yangtze Rivers. However, it shrunk
southwards into a small region in the southwest of the Shandong
Peninsula. The mixed farming region then expanded further during
the Late Yangshao Period (6000–5000 BP). It spread westwards
into the Hexi Corridor and boomed in the Central Plains, which
linked the Songshan Mountain, Nanyang Basin and Hanshui Valley
together, but remained limited to a small area in the Haidai Region.
Mixed farming flourished during the Longshan Period (5000–
4000 BP). It moved westwards into the Sichuan Basin along the
Yangtze River and expanded rapidly in the Haidai Region, spread-
ing northeast into the Shandong and Liaoning Peninsulas. Mixed
farming sites first appeared in the lower Yangtze River Valley and
on the west coast of Taiwan Island, which were isolated from the
main mixed farming region. During the Bronze Age (4000–2221
BP), mixed farming expanded again into a new domain: the Yungui
Plateau and Hengduan Mountain to the southwest and Qinghai
Region to the northwest. As a consequence of the dispersal of peo-
ple and agriculture, mixed farming gradually extended from 8400
BP and eventually flourished from 5000 BP in prehistoric China.
Corridors of exchange and spread between millet
and rice farming
Because of the restrictions on productivity and knowledge among
ancient people, agriculture tends to be disseminated along the
natural corridors of river valleys, basins or coasts. Based on an
analysis of geographical positions, archaeological ages and crop
structures of several key sites in the three subareas of mixed farm-
ing, three north–south corridors of agricultural dissemination are
recognized with different starting ages and directions of propaga-
tion (Figure 5).
The west corridor. The west corridor lies between the Ganqing
Region and the Chengdu Plain that was historically called the Zan-
gyi corridor (Chen et al., 2007), which belongs to an input corridor
of millet to the Chengdu Plain (Figure 5). The earliest archaeobo-
tanical materials discovered in the Sichuan Basin were at the
Haxiu site (5500–5000 BP) in the upper Daduhe River and the
Yingpanshan site (5300–5000 BP) in the Minjiang River (Zhao
and Chen, 2011; Figure 4), and crops found were all millet. The
earliest mixed farming sites excavated were Guiyuanqiao (4600–
4300 BP) and Baodun (4500–3700 BP) (Figure 4), which were all
rice-based agriculture (Guedes et al., 2013). Most scholars agree
that rice farming spread from the middle Yangtze River upstream
through the eastern Sichuan mountain area (Zhang and Huang,
2010). Archaeological results show that the western corridor is an
input corridor of millet farming from the Ganqing regions in the
north that began in 5500 BP, preceding rice farming imported from
the middle Yangtze River in the east around 4500 BP.
The central corridor. The central corridor lies between the Central
Plains and the Hanshui Valley through the Nanyang Basin, which
belongs to the original centre of agriculture and engaged in bilat-
eral exchange (Figure 5). The earliest mixed farming sites are
Baligang (8400–8100 BP) in Nanyang Basin (Deng et al., 2015;
Weisskopf et al., 2015a) and Chengtoushan (6497–6021 BP) in
Liyang Plain (Nasu et al., 2007, 2011) (Figure 4). Rice farming in
the south spread through the Hanshui River or the Suizao Corri-
dor into the Nanyang Basin and Songshan Mountain around 8400
BP and then to the Ganqing Region along the Weihe River in the
west (An et al., 2010; Li et al., 2007) and also to the Haidai
Region along the Yellow River in the east (Zhang and Huang,
2013a). Millet farming in the north spread southwards along the
Hanshui and Yangtze Rivers and reached the Hanshui Valley and
Liyang Plain around 6400 BP. Archaeological results show that
the central corridor was a bilateral corridor where rice spread
northwards starting in 8400 BP and millet spread southwards
around 6400 BP through the Nanyang Basin.
Figure 5. Three north–south communication corridors for crops from west to east in China. Red lines with arrows indicate the propagation
path of millet, whereas the black lines indicate the propagation path of rice. The source of rice in Haidai Region is uncertain and is indicated by
a dotted line with an arrow.
6 The Holocene 00(0)
The east corridor. The east corridor lies in the eastern alluvial
plain between the Haidai Region and the Jiangzhe Region along
the coastline of the Yellow Sea (Figure 5). The earliest mixed
farming sites are Xihe (8070–7900 BP) (Jin et al., 2014) and
Yuezhuang (8060–7750 BP) (Crawford et al., 2013) in the Haidai
Region (Figure 4). It remains unclear whether the rice was culti-
vated in site or spread from the upper Huaihe River (Zhang and
Huang, 2013a) or the lower Yangtze River (Figure 6). However,
rice farming was constrained to the southern parts of Haidai
Region during the following culture of Beixin and Dawenkou
(sites of rice in Haidai Region are given in Supplementary Table,
available online) and seemed to have little contact with the lower
Yangtze River until the Longshan culture.
The communion corridor was not defined until 4600 BP when
mixed farming sites exhibited explosive growth during the Long-
shan culture, and the extensive introduction of rice farming began
once again (Figure 5; Guedes et al., 2015). Few sites with foxtail
millet are found in the lower Yangtze River. The only two sites are
Bianjiashan (4500–4300 BP) of the late Liangzhu culture and
Dongjiaqiao (2770–2255 BP) of the Eastern Zhou Dynasty
(Figure 4) with merely one grain of foxtail millet floated out, indi-
cating that millet farming never arrived in the lower Yangtze
River. Although recently two grains of foxtail millet are floated
out in the late Neolithic Age of Shangshan site (6500 BP) (Zhao
and Jiang, 2016), no direct dating has been made of the seeds and
possibilities of later intrusion still exist (Yang et al., 2016a) which
need to be further confirmed.
Change in crop structure in the four core mixed
farming regions
A total of 98 of the 155 mixed farming sites have complete
published floatation results. Furthermore, 78 of 98 sites floated
cluster in four regions – Chengdu Plain, Central Plains, Hanshui
Valley and Haidai Region (Figure 4, regions circled by black dot-
ted line) – among the three subareas, whereas others were scat-
tered in the marginal zones of the mixed farming regions. To
investigate the main patterns of the mixed farming mode, these
four core regions of mixed farming are selected for comparison.
Haidai Region. The proportion of foxtail millet rises steadily and
gradually establishes dominance from the Longshan culture. Rice
accounts for a high proportion, approximately 50% during the
early and middle Neolithic periods, and then decreases progres-
sively. Although it recovers slightly during the Longshan culture,
rice decreases soon afterwards to 5% as a minor crop. The propor-
tion of broomcorn millet exceeds that of foxtail millet before the
Dawenkou culture but decreases rapidly afterwards. Wheat first
appears in the Longshan culture and increases to approximately
30% in the Shang and Zhou Dynasties, becoming the second main
crop after foxtail millet (Figure 6).
Central Plains. The proportion of foxtail millet shows a trend of a
fluctuating increase and generally exceeds 60% after the Long-
shan culture. The proportion of broomcorn millet is never higher
than that of foxtail millet; it decreases gradually to 20% after the
Yangshao culture and represents less than 5% in the Bronze Age.
Rice constitutes as much as 50% during the Peiligang culture and
then decreases progressively to 20%. Although rice returns to
reach 40% during the Qujialing and Shijiahe cultures, it still does
not become predominant (Figure 6).
Hanshui Valley. Mixed farming begins from the late Yangshao
culture, which is predominantly millet farming, and transforms
into predominantly rice farming during the Longshan culture. The
Figure 6. Change in crop structure in the four main mixed farming regions during different cultural periods, respectively.
He et al. 7
change in the millet–rice ratio exhibits an obvious negative rela-
tionship. The proportion of foxtail millet reaches as high as 85%
during the Yangshao culture and declines rapidly to 48%, ulti-
mately remaining steady at approximately 40%. Rice accounts for
only 10% during the Yangshao culture, rises rapidly to 45% and
eventually becomes dominant at approximately 60%. The propor-
tion of broomcorn is consistently less than 10% (Figure 6).
Chengdu Plain. Mixed farming emerges very late during the
Longshan Period, and rice consistently plays a leading role,
accounting for 80%. The proportion of foxtail millet rises slightly
from 10% to 20% through the late Neolithic to the Bronze Age,
whereas the proportion of broomcorn millet is consistently lower
than 5% (Figure 6).
Comparison of development patterns among four
core mixed farming regions
Among the four core mixed farming regions, the Haidai Region
and Central Plains lie in the middle and lower Yellow River Val-
ley, whereas the Chengdu Plain and Hanshui Valley are located in
the upper and middle Yangtze River Valley, which were tradition-
ally thought to be millet and rice farming regions, respectively.
The evolution of the crop structure in these four regions shows
diversified development patterns: the Haidai Region and Central
Plains in the north fluctuate sharply, whereas the Hanshui Valley
and Chengdu Plain develop steadily.
Ratios of foxtail millet to broomcorn millet and millet to rice
described below are merely based on the number of crop grains.
Although it is improper to take the grain of millet and rice in equal
station, the index of weight or food values should be more appro-
priate to exhibit the crop structure. But no effective formulas have
been proposed based on experimental observation at present
which concern the charred preservation and food value estimate
of each grain, and these conversions will also introduce an addi-
tional source of variation in interpreting these results. Therefore,
these methods are not applied now and more work needs to be
done in the future.
Ratio of foxtail millet to broomcorn millet. In the mixed farming
regions of the Yellow River, foxtail millet is the predominant crop
overall. The proportion of broomcorn millet is rather high during
the early and middle Neolithic periods but decreases rapidly after
the Longshan culture and is almost replaced by wheat during the
Shang and Zhou Dynasties. However, with regard to the mixed
farming regions in the Yangtze River, the proportion of broom-
corn millet is consistently low and is far less than that of foxtail
millet (Figure 7).
Ratio of millet to rice. Taking the ratio of millet as standard, the
four core mixed farming regions can be divided into three types of
development patterns (Figure 7). The ‘ascending type’ consists of
the Haidai Region and Central Plains, where the proportion of
millet is maintained at 50% and subsequently fluctuates upwards.
The ‘descending type’ is the Hanshui Valley, where the proportion
of millet farming is initially as high as 90% and declines continu-
ously afterwards. The ‘stable type’ is the Chengdu Plain, where
rice is primary and millet is supplementary throughout the mixed
farming mode.
Cultural communication between the Yellow and
Yangtze River Valleys
The spread of agriculture from its homelands involves two cases of
the mere acquisition of crops and technology and more cases of the
spread of farmers themselves and their culture (Diamond, 2002).
The Farming/Language Dispersal Hypothesis proposed by Bell-
wood postulates that the spread of early farming lifestyles is often
correlated with prehistoric episodes of human population and lan-
guage dispersal (Bellwood, 2009; Bellwood and Renfrew, 2002). In
fact, archaeological remains of pottery combinations have verified
the existence of communication corridors along the upper, middle
and lower valleys of the Yellow and Yangtze Rivers.
In the west corridor, archaeological evidence indicates that
the southwestward expansion of the Majiayao population in
Gansu and Qinghai along the Bailongjiang and Minjiang Rivers
around 5500 BP (Figure 6) led to the formation of the Yingpan-
shan culture in northwest Sichuan (Zhang and Huang, 2013b).
Accompanied by human migration, millet farming spread into
this region (Zhang and Huang, 2010), and this southward
migration was likely to be related to the cold event around 5500
BP (Bond, 1997).
In the central corridor, Nanyang Basin is a key link between
the middle Yellow and Yangtze Rivers, with the Fangcheng Pass
in the northeast and the Nanxiang Defile in the south (Li et al.,
2013). The prehistoric cultural configuration of the Nanyang
Basin is dominated by the Yangshao and Longshan cultures in
the Central Plains and profoundly affected by the Qujialing and
Shijiahe cultures in the Hanshui Valley (Li et al., 2015; Ma and
Yang, 2007), which have important effects on the spread of agri-
culture and the variation in crop structure. When the Yangshao
and Longshan cultures flourished, millet spread southwards and
accounted for a high proportion of crops in the Central Plains
and Hanshui Valley. On the contrary, when the Qujialing and
Shijiahe cultures thrived, rice spread northwards and accounted
for a high proportion in the Central Plains and the Hanshui Val-
ley (Figure 7, pink shadow).
In the eastern corridor, early communication between the
Haidai Region and the lower Yangtze River Valley can be traced
back to Beixin culture (7400–6200 BP) (Lu, 2009) and Dawenkou
culture (6200–4600 BP) (Luan, 1997) where typical potteries are
excavated in the opposite regions from the homeland. However,
study on the sites distributed along the abandoned Yellow and
Yangtze River deltas suggested little evidence for early cultural
migration between the northern and southern coasts until the Late
Liangzhu stage (5200–4000 BP), which may be blocked by the
sway of the Yellow River (Liu et al., 2007; Xue et al., 2004) or the
rise in sea level (Chen et al., 2008). Cultural communication
along the coast seems discontinuous until Longshan culture that
rice farming thrived again in Haidai Region.
Environmental background for the spread of millet
and rice farming
The expansion of crops tends to occur more rapidly along east–
west axes at the same latitude in areas that share similar climates
and habitats than along north–south axes (Diamond, 2002). How-
ever, the situation in China is quite different, with two types of
crops in the north and south. The valleys of the Yellow and Yang-
tze Rivers are the cradles of millet and rice farming, respectively,
along the east–west, with multiple centres of origin (Figure 3;
Bettinger et al., 2010; Silva et al., 2015). In fact, the east–west
propagation paths of crops within the rice and millet farming
region remain uncertain, but the south–north interaction between
the rice and millet farming regions is facilitated by the specific
natural environment of China.
Climate and soil are two basic elements of agriculture. The
unique East Asian monsoon and widespread loess in China have
had considerable effects on the formation of a dualistic agricul-
tural structure. One important feature of the East Asian monsoon
is the synchronization between rainfall and temperature in sum-
mer (Ding and Chan, 2005), which provides sufficient water and
8 The Holocene 00(0)
heat for the growth of rice in the eastern monsoon areas of China.
Quantitative temperature and precipitation reconstructed from
high-resolution climate proxies indicate that the temperature in
the early Holocene was high and decreased after 7000 BP, with a
sharp decrease around 4000 BP (Figure 7a, red curve) (Marcott
et al., 2013). Precipitation reached its maximum from 7800 BP to
5300 BP, with a rapid decline after 3300 BP (Figure 7b) (green
curve; Chen et al., 2015b). The Holocene Optimum from 8000 BP
to 4000 BP promoted the northward spread of rice in the central
and east corridors.
During the Peiligang Period (9000–7000 BP), mixed farming
first emerged in the Haidai Region and Central Plains, with rice
accounting for a high proportion. On one hand, this may be
because only a few sites were found; on the other hand, it may
reflect that rice propagation was facilitated by the beginning of
the Holocene Optimum (Figure 7, gold line) (An et al., 2000).
During the Bronze Age (4000–2221 BP), the proportion of rice
declined again in the Haidai Region and Central Plains, and millet
increased slightly in the Chengdu Plain. This shift may have been
affected by the cold event of 4000 BP (Figure 7, gold line) (Bond,
1997; Perry and Hsu, 2000), which had a significant influence on
the margin zone of rice farming.
Loess in China has long been related to the origin of millet agri-
culture. Ho (1969) notes that most millet sites are located in loess
terraces or mounds along various tributaries of the Yellow River
rather than in the great river valley itself. Liu supports this view that
the earliest millet sites are actually found in the foothills of the
neighbouring mountain chains along the margins of the Loess Pla-
teau and the Inner Mongolian Plateau (Liu et al., 2009). With char-
acteristics of friability and porosity (Liu, 1985), loess is appropriate
for ploughing, draining and growing roots for millet. Loess is
mainly distributed in north China but is also scattered in the Yang-
tze River Valley, including west of the Sichuan Basin, Dabieshan–
Wushan Mountain and Ningzhen Region (Figure 8, regions
between the black and red dotted lines). With detailed information
on 463 millet sites, we illustrate the striking feature that the distri-
bution of millet sites and loess regions in China is highly consistent
(Figure 8). Loess distributed in the slope of basins or mountains of
south China that are well drained provide the basis for the south-
ward spread of millet in the west and middle corridors.
Figure 7. A comparison of crop structure patterns among the four main mixed farming regions and the change in temperature and
precipitation in the Holocene. (a) Reconstructed temperature variation in Northern Hemisphere (30°–90°N) (Marcott etal., 2013).
(b) Pollen-based annual precipitation (PANN) reconstructed from Gonghai Lake (Chen etal., 2015b). (c)–(f) Different development
patterns of crop structures in four main mixed farming regions.
He et al. 9
Adaptation and adjustment of the crop structure to
the local habitat
Millet and rice are two main crops of native origin in China with
different growth habits and ecological niches. The distribution of
loess and the climate of East Asian monsoons make it possible for
these two crops to be cultivated in geographically opposite
regions. As a consequence of the southward spread of millet and
the northward spread of rice, mixed farming is not a simple com-
bination of indigenous and exotic crops but involves the con-
scious selection and alteration of crop structure to improve its
adaptive capacity to environmental change.
Selection of foxtail millet in the southward spread of millet agricul-
ture. Previous studies of millet farming in sites such as Dadiwan
(Liu et al., 2004), Cishan (Lu et al., 2009a) and Xinglonggou
(Zhao, 2011b) during the Peiligang Period have shown that
broomcorn millet was the principle crop. However, by the time of
the Longshan Period, foxtail millet became the most important
crop instead (Crawford et al., 2005; Zhao, 2011a). Dodson holds
the view that drier climates in the early Holocene (Feng et al.,
2006) may have been crucial in the adoption of broomcorn millet,
which can tolerate poorer soils and drought than foxtail millet
(Dodson and Dong, 2016). Crawford notes that foxtail millet is
traditionally more common in the wetter eastern areas, and
broomcorn millet is more common in the drier interior areas
(Crawford et al., 2005; Lee et al., 2007). However, whether there
is a spatio-temporal difference in the selection of millet remains
controversial in mixed farming regions.
The structure of millet shows diverse patterns in the four core
mixed farming regions. Broomcorn millet is approximately equal
in importance to foxtail millet before the Longshan Period but
declines markedly afterwards in the Haidai Region and Central
Plains. However, the proportion of broomcorn millet is consis-
tently small in the Hanshui Valley and Chengdu Plain after the
introduction of millet from the Central Plains and Ganqing
Region. Our results support the hypothesis that broomcorn millet
occupies an important position in the early and middle Neolithic
in north China. However, when millet agriculture developed in
the late Neolithic and spread into south China, foxtail millet
became predominant instead.
The ratio of foxtail millet to broomcorn millet may be linked
to their crop traits. Compared with foxtail millet, broomcorn mil-
let is more adaptable to adverse environments, has better resis-
tance to arid climates and saline–alkali soil and has a shorter
growing season; however, its yield is lower (Lin et al., 2002).
Broomcorn was preferable for cultivation in the early Neolithic
period (Lu et al., 2009a), when productivity was low and the cli-
mate was relatively dry. However, the advantage of foxtail millet
with regard to yield was noted with the progress of agricultural
techniques (Zhao, 2011a) and spread into moist regions in the
Yangtze Valley, while broomcorn millet ultimately degraded into
secondary crops prepared for emergencies and disasters.
Breeding of temperate Oryza japonica in the northward spread
of rice farming. Genetic studies have shown that rice clusters
into two separate complexes, Oryza indica and Oryza japonica,
Figure 8. Distribution of millet sites and loess in China. Yellow shadow represents the range of loess redrawn from Figure 2 of Liu (1985). Red
dots with black circles are millet sites. Black dotted lines represent the distribution of loess around 34° north, and red dotted lines represent
the south boundary of loess in China. 1, Ningbo city; 2, Nanjing city; 3, Dabieshan Mountain; 4, Wushan Mountain; and 5, West of Sichuan Basin.
10 The Holocene 00(0)
Figure 9. Grain morphology, genetics and thermal niche related to the cultivation of temperate Oryza japonica. (a) Sites studied for the grain
size of rice. (b) Scatter diagram of different types of rice grain around 8000 BP and 5000 BP. (c) Unrooted phylogenetic tree clearly illustrates
major division between the two varietal groups (indica and japonica), which are further subdivided into the five rice subpopulations: indica, aus,
tropical japonica, temperate japonica and aromatic (Garris etal., 2005). (d) Thermal niche of temperate and tropical O. japonica at a temperature
perturbation of 1°C in Shandong Province and southwest China (Guedes and Butler, 2014; Guedes etal., 2015).
He et al. 11
which were domesticated separately in China and India, respec-
tively (Huang et al., 2012; Kovach et al., 2007; Londo et al.,
2006). Oryza japonica varieties can be classified into temperate
and tropical ecotypes on the basis of their ecological, morpho-
logical and physiological characteristics (Garris et al., 2005; Si
et al., 2016). Genetic comparison within O. japonica supports
the hypothesis that temperate O. japonica is derived from the
tropical O. japonica and selected for the temperature during the
spread northwards (Garris et al., 2005; Khush, 1997; Wu et al.,
2013). With regard to the morphological traits of grains, typical
tropical O. japonica varieties have long, thin grains, whereas the
grains of temperate O. japonica are short and squat (Figure 9c)
(Liakat et al., 2011).
The morphological data of rice grains from 14 sites (Figure
9a) are used for discussion. Grains excavated in the sites around
8000 BP are all very thin, with a width below 2 mm. Grains at
the two sites of Bashidang and Kuahuqiao were as long as 5 mm
(Figure 9b). The characteristics of these grains cannot be con-
firmed; they may belong to wild, immature (Fuller et al., 2007)
or ancestral japonica similar to tropical japonica (Fuller and
Castillo, 2016). By contrast, grains found around 5000 BP that
had already been domesticated are obviously wider than 2.2 mm
and can be divided into two clusters by length. Grains found in
the Haidai Region and the Chengdu Plain are all shorter than
those found in the lower Yangtze River Valley (Figure 9b),
which suggests two types of rice, indicating temperate and tropi-
cal O. japonica, respectively. In addition, thermal niche models
constructed in the Haidai Region and Southwest China (Figure
9d) show that these two regions may be suitable for the cultiva-
tion of tropical O. japonica with 2900 GDD (growing degree
days) around 8000 BP that were approximately 1°C higher than
the present. However, given the temperature around 5000 BP
that was equal to or lower than the recent temperature, only tem-
perate O. japonica of 2500 GDD could be cultivated there
(Guedes and Butler, 2014; Guedes et al., 2015).
The earliest rice domesticated in the lower Yangtze River may
have been tropical O. sativa subs. japonica (Fuller and Castillo
2016), which spread to the Haidai Region and Central Plains by
the beginning of Holocene Optimum. During the Yangshao Period
(7000–5000 BP), the proportion of rice in the Haidai Region and
the Central Plains declined, which may have resulted from the
maladjustment of primitive tropical rice to the temperate climate
in the north as the temperature declined (Marcott et al., 2013).
During the Longshan Period (5000–4000 BP), rice farming
expanded again. Rice farming sites increased significantly in the
Haidai Region, with an increased proportion of rice in crop struc-
ture, and rice arrived in the Chengdu Plain, by occupying a domi-
nant position. Two waves of rice expansion have been proposed
before (Qin, 2012), and studies of this paper suggest that they
may be related to different types of rice. The first wave that begins
around 8000 BP is associated with the Holocene Optimum and
primitive tropical O. japonica, whereas the second wave that
occurred around 5000 BP was the consequence of success in
breeding temperate O. japonica.
From the discussions above, three conclusions can be drawn.
First, the earliest mixed farming arises in the Central Plains and
the Haidai Region around 8000 BP. Three north-southward corri-
dors of crop communion were developed between the upper,
middle and lower Yellow and Yangtze River Valleys, which began
around 5500 BP, 8400 BP and 4600 BP, respectively. Second,
cultural communication accompanied by human migration, the
unique natural environment of loess and the East Asia monsoon
facilitated the interaction between millet and rice farming and the
formation of mixed farming. Finally, the crop patterns of mixed
farming regions are not simple, stable compounds but involve
continual adjustments of the agricultural planting structure with
the selection of foxtail millet in the southward spread of millet
and of temperate O. japonica in the northward spread of rice.
This paper systematically investigated the interaction between
millet and rice farming and quantitatively revealed the evolution
of the crop structure in core mixed farming regions. However,
because of the limitations in the materials and the method used,
the ages of many sites are estimated by cultural period with few
little direct dating. Floatation results alone are used for the discus-
sion of crop structure, no phytolith or starch grain results are
included and only simple measures of proportion are applied to
the analysis of floatation results without consideration of density
and ubiquity. In future studies, more precise ages, archaeobotani-
cal data and analytical methods are needed to promote a better
understanding of mixed farming.
The authors are grateful to Zhenhua Deng, Yu Gao and Xiaoqu
Zheng for the acquirement of some floatation results; Qirui Sun
for the discussion of new chronology frame in Haidai Region; and
two anonymous reviewers for their helpful comments and sug-
gestions. They would also like to thank American Journal Experts
( for English language editing.
This work was jointly funded by the National Science and
Technology Major Project of China (Grant No. 2015CB953803),
the National Natural Science Foundation of China (Grant No.
41230104, 41430103) and Research Projects on the Environ-
mental Basis of the Formation and Evolution of Songshan
An CB, Ji DX, Chen FH et al. (2010) Evolution of prehistoric agri-
culture in central Gansu Province, China: A case study in Qin’an
and Li County. Chinese Science Bulletin 55: 1925–1930.
An ZS, Porter SC, Kutzbach JE et al. (2000) Asynchronous Holo-
cene optimum of the East Asian monsoon. Quaternary Sci-
ence Reviews 19: 743–762.
Barton L, Newsome SD, Chen FH et al. (2009) Agricultural ori-
gins and the isotopic identity of domestication in northern
China. Proceedings of the National Academy of Sciences of
the United States of America 106: 5523–5528.
Bar-Yosef O (2011) Climatic fluctuations and early farming in
West and East Asia. Current Anthropology 52: S175–S193.
Bellwood P (2005) First Farmers: The Origins of Agricultural
Societies. Malden, MA: Blackwell.
Bellwood P (2009) The dispersals of established food-producing
populations. Current Anthropology 50: 621–626.
Bellwood P (2011) The checkered prehistory of rice movement
southwards as a domesticated cereal – From the Yangzi to the
Equator. Rice 4: 93–103.
Bellwood P and Renfrew C (2002) Examining the Farming: Lan-
guage Dispersal Hypothesis. Cambridge: McDonald Institute
for Archaeological Research.
Bettinger RL, Barton L and Morgan C (2010) The origins of food
production in North China: A different kind of agricultural
revolution. Evolutionary Anthropology 19: 9–21.
Bond G (1997) A pervasive millennial-scale cycle in North Atlan-
tic Holocene and glacial climates. Science 278: 1257–1266.
Castillo CC (2011) Rice in Thailand: The archaeobotanical con-
tribution. Rice 4: 114–120.
Chang K-C (1964) The Archaeology of Ancient China. New
Haven, CT: Yale University.
12 The Holocene 00(0)
Chen FH, Dong GH, Zhang DJ et al. (2015a) Agriculture facili-
tated permanent human occupation of the Tibetan Plateau
after 3600 B.P. Science 347: 248–250.
Chen FH, Xu QH, Chen JH et al. (2015b) East Asian summer
monsoon precipitation variability since the last deglaciation.
Scientific Reports 5: 11186.
Chen SW, Hu QS, Xie Y et al. (2007) Origin of Tibeto-Burman
speakers: Evidence from HLA allele distribution in Lisu and
Nu inhabiting Yunnan of China. Human Immunology 68:
Chen ZY, Zong YQ, Wang ZH et al. (2008) Migration patterns of
Neolithic settlements on the abandoned Yellow and Yangtze
River deltas of China. Quaternary Research 70: 301–314.
Cohen DJ (2011) The beginnings of agriculture in China A multi-
regional view. Current Anthropology 52: S273–S293.
Crawford GW (2011) Advances in understanding early agricul-
ture in Japan. Current Anthropology 52: S331–S345.
Crawford GW and Lee G-A (2003) Agricultural origins in the
Korean Peninsula. Antiquity 77: 87–95.
Crawford GW, Chen X, Luan F et al. (2013) A preliminary analy-
sis on plant remains of the Yuezhuang site in Changqing Dis-
trict, Jinan city, Shandong Province. Jianghan Archaeology 2:
107–116 (in Chinese).
Crawford GW, Underhill AP, Zhao ZJ et al. (2005) Late Neo-
lithic plant remains from Northern China: Preliminary results
from Liangchengzhen, Shandong. Current Anthropology 46:
Deng ZH, Qin L, Gao Y et al. (2015) From early domesticated
rice of the middle Yangtze Basin to millet, rice and wheat
agriculture: Archaeobotanical macro-remains from Baligang,
Nanyang Basin, Central China (6700–500 BC). PLoS ONE
10: e0139885.
Diamond J (2002) Evolution, consequences and future of plant
and animal domestication. Nature 418: 700–707.
Diamond J and Bellwood P (2003) Farmers and their languages:
The first expansions. Science 300: 597–603.
Ding YH and Chan JCL (2005) The East Asian summer mon-
soon: An overview. Meteorology and Atmospheric Physics
89: 117–142.
Dodson JR and Dong GH (2016) What do we know about
domestication in eastern Asia? Quaternary International
426: 2–9.
Dong Z and Shen XY (2000) General Instruction to Crop Culti-
vation. Beijing: China Agriculture Press (in Chinese).
Feng ZD, An CB and Wang HB (2006) Holocene climatic and
environmental changes in the arid and semiarid regions of
China: A review. The Holocene 16: 119–130.
Frachetti MD, Spengler RN, Fritz GJ et al. (2010) Earliest direct
evidence for broomcorn millet and wheat in the central Eur-
asian steppe region. Antiquity 84: 993–1010.
Fuller DQ (2011) Pathways to Asian civilizations: Tracing the ori-
gins and spread of rice and rice cultures. Rice 4: 78–92.
Fuller DQ and Castillo CC (2016) Diversification and cultural
construction of a crop: The case of glutinous rice and waxy
cereals in the food cultures of Eastern Asia. In: Lee-Thorp
J and Anne Katzenberg M (eds) The Oxford Handbook of
the Archaeology of Diet. Oxford: Oxford University Press.
DOI: 10.1093/oxfordhb/9780199694013.013.8.
Fuller DQ and Qin L (2009) Water management and labour in the
origins and dispersal of Asian rice. World Archaeology 41:
Fuller DQ, Denham T, Arroyo-Kalin M et al. (2014) Convergent
evolution and parallelism in plant domestication revealed
by an expanding archaeological record. Proceedings of the
National Academy of Sciences of the United States of America
111: 6147–6152.
Fuller DQ, Harvey EL and Qin L (2007) Presumed domestica-
tion? Evidence for wild rice cultivation and domestication in
the fifth millennium BC of the Lower Yangtze region. Antiq-
uity 81: 316–331.
Fuller DQ, Sato Y-I, Castillo CC et al. (2010) Consilience of
genetics and archaeobotany in the entangled history of rice.
Archaeological and Anthropological Sciences 2: 115–131.
Garris AJ, Tai TH, Coburn J et al. (2005) Genetic structure and
diversity in Oryza sativa L. Genetics 169: 1631–1638.
Grayson DK (1979) On the quantification of vertebrate archaeo-
faunas. Advances in Archaeological Method and Theory 2:
Grayson DK (2014) Quantitative Zooarchaeology: Topics in the
Analysis of Archaeological Faunas. Amsterdam: Elsevier.
Guedes JDA (2011) Millets, rice, social complexity, and the
spread of agriculture to the Chengdu Plain and Southwest
China. Rice 4: 104–113.
Guedes JDA and Butler EE (2014) Modeling constraints on the
spread of agriculture to Southwest China with thermal niche
models. Quaternary International 349: 29–41.
Guedes JDA, Jiang M, He KY et al. (2013) Site of Baodun yields
earliest evidence for the spread of rice and foxtail millet agri-
culture to south-west China. Antiquity 87: 758–771.
Guedes JDA, Jin GY and Bocinsky RK (2015) The impact of
climate on the spread of rice to North-Eastern China: A new
look at the data from Shandong Province. PLoS ONE 10:
Harlan JR (1971) Agricultural origins: Centers and noncenters.
Science 174: 468–474.
Ho P-T (1969) The loess and the origin of Chinese agriculture.
The American Historical Review 75: 1–36.
Hu LC, Chao ZH, Gu M et al. (2013) Evidence for a Neolithic
Age fire-irrigation paddy cultivation system in the lower
Yangtze River Delta, China. Journal of Archaeological Sci-
ence 40: 72–78.
Huang XH, Kurata N, Wei XH et al. (2012) A map of rice genome
variation reveals the origin of cultivated rice. Nature 490:
Hung H-C and Carson MT (2014) Foragers, fishers and farmers:
Origins of the Taiwanese Neolithic. Antiquity 88: 1115–1131.
Hunt HV, Linden MV, Liu XY et al. (2008) Millets across Eur-
asia: Chronology and context of early records of the genera
Panicum and Setaria from archaeological sites in the Old
World. Vegetation History and Archaeobotany 17: S5–S18.
Institute of Archaeology, Chinese Academy of Social Sciences
(2010) Neolithic Volume of Chinese Archaeology. Beijing:
China Social Science Press (in Chinese).
Iriarte J, Holst I, Marozzi O et al. (2004) Evidence for cultivar
adoption and emerging complexity during the mid-Holocene
in the La Plata basin. Nature 432: 614–617.
Jiang LP and Liu L (2006) New evidence for the origins of sed-
entism and rice domestication in the Lower Yangzi River,
China. Antiquity 80: 355–361.
Jin GY, Wagner M, Tarasov PE et al. (2016) Archaeobotanical
records of Middle and Late Neolithic agriculture from Shan-
dong Province, East China, and a major change in regional
subsistence during the Dawenkou Culture. The Holocene 26:
Jin GY, Wu WW, Zhang KS et al. (2014) 8000-Year old rice
remains from the north edge of the Shandong Highlands, East
China. Journal of Archaeological Science 51: 34–42.
Jin GY, Yan SD, Udatsu T et al. (2007) Neolithic rice paddy from
the Zhaojiazhuang site, Shandong, China. Chinese Science
Bulletin 52: 3376–3384.
Jones M, Hunt H, Lightfoot E et al. (2011) Food globalization in
prehistory. World Archaeology 43: 665–675.
He et al. 13
Khush GS (1997) Origin, dispersal, cultivation and variation of
rice. In: Sasaki T and Moore G (eds) Oryza: From Molecule
to Plant. Dordrecht: Springer, pp. 25–34.
Kovach MJ, Sweeney MT and McCouch SR. (2007) New insights
into the history of rice domestication. Trends in Genetics 23:
Lee G-A (2011) The transition from foraging to farming in prehis-
toric Korea. Current Anthropology 52: S307–S329.
Lee G-A, Crawford GW, Liu L et al. (2007) Plants and people
from the early Neolithic to Shang periods in north China. Pro-
ceedings of the National Academy of Sciences of the United
States of America 104: 1087–1092.
Lev-Yadun S, Gopher A and Abbo S (2000) The cradle of agricul-
ture. Science 288: 1602–1603.
Li F, Wu L, Zhu C et al. (2013) Spatial-temporal distribution and
geographic context of Neolithic cultural sites in the Hanjiang
River Basin, Southern Shaanxi, China. Journal of Archaeo-
logical Science 40: 3141–3152.
Li H-L (1970) The origin of cultivated plants in Southeast Asia.
Economic Botany 24: 3–19.
Li HM, Zuo XX, Kang LH et al. (2016) Prehistoric agriculture
development in the Yunnan-Guizhou Plateau, southwest
China: Archaeobotanical evidence. Science China Earth Sci-
ences 59: 1562–1573.
Li XQ (2002) The Xia-Shang-Zhou chronology project: Meth-
odology and results. Journal of East Asian Archaeology 4:
Li XQ, Dodson J, Zhou XY et al. (2007) Early cultivated wheat
and broadening of agriculture in Neolithic China. The Holo-
cene 17: 555–560.
Li ZX, Zhu C, Yuan SY et al. (2015) Geographical distribution,
diffusion and subsistence variation of prehistoric cultures in
Nanyang Basin, Henan Province. Acta Geographica Sinica
70: 143–156 (in Chinese).
Liakat AM, McClung AM, Jia MH et al. (2011) A rice diversity
panel evaluated for genetic and agro-morphological diversity
between subpopulations and its geographic distribution. Crop
Science 51: 2021–2035.
Lin RF, Chai Y, Liao Q et al. (2002) Minor Grain Crops in China.
Beijing: China Agricultural Science and Technology Press (in
Liu CJ and Kong ZC (2004) Morphological comparison of foxtail
millet and broomcorn millet and its significance in archaeo-
logical identification. Archaeology 8: 76–83 (in Chinese).
Liu CJ, Kong ZC and Lang SD (2004) Discussion in the relation-
ship between the agricultural plant remains and human habitat
in Dadiwan site. Cultural Relics of Central China 4: 26–30
(in Chinese).
Liu J, Saito Y, Wang H et al. (2007) Sedimentary evolution of the
Holocene subaqueous clinoform off the Shandong Peninsula
in the Yellow Sea. Marine Geology 236: 165–187.
Liu TS (1985) Loess and the Environment. Beijing: China Ocean
Press (in Chinese).
Liu XY, Hunt HV and Jones MK (2009) River valleys and foot-
hills: Changing archaeological perceptions of North China’s
earliest farms. Antiquity 83: 82–95.
Londo JP, Chiang YC, Hung KH et al. (2006) Phylogeography of
Asian wild rice, Oryza rufipogon, reveals multiple indepen-
dent domestications of cultivated rice, Oryza sativa. Proceed-
ings of the National Academy of Sciences of the United States
of America 103: 9578–9583.
Lu HY, Wu NQ and Liu BZ (1997) Recognition of rice phyto-
liths. In: Pinilla A, Juan-Tresserras J and Machado MJ (eds)
The State-of-the-Art of Phytoliths in Soils and Plants. Madrid:
Monografias del Centro de Ciencias Medioambientales, pp.
Lu HY, Zhang JP, Liu KB et al. (2009a) Earliest domestication of
common millet (Panicum miliaceum) in East Asia extended
to 10,000 years ago. Proceedings of the National Academy
of Sciences of the United States of America 106: 7367–7372.
Lu HY, Zhang JP, Wu NQ et al. (2009b) Phytoliths analysis for
the discrimination of foxtail millet (Setaria italica) and com-
mon millet (Panicum miliaceum). PLoS ONE 4: e4448.
Lu JY (2009) On relations between Beixin culture and Majiabang
culture. Southeast Culture 6: 39–46 (in Chinese).
Luan FS (1997) Relationship between the Dawenkou culture and
Songze, Liangzhu culture. In: Luan FS (ed.) Archaeologi-
cal Research in Haidai Region. Jinan: Shandong University
Press, pp. 134–155 (in Chinese).
Ma BC and Yang L (2007) Tentative study on cultural com-
munication among Hubei, Henan and Shaanxi during the
Late Neolithic Age. Jianghan Archaeology 2: 42–51 (in
McGovern PE, Zhang JZ, Tang JG et al. (2004) Fermented bev-
erages of pre- and proto-historic China. Proceedings of the
National Academy of Sciences of the United States of America
101: 17593–17598.
Marcott SA, Shakun JD, Clark PU et al. (2013) A reconstruction
of regional and global temperature for the past 11,300 years.
Science 339: 1198–1201.
Nasu H and Momohara A (2016) The beginnings of rice and mil-
let agriculture in prehistoric Japan. Quaternary International
397: 504–512.
Nasu H, Gu HB, Momohara A et al. (2011) Land-use change for
rice and foxtail millet cultivation in the Chengtoushan site,
central China, reconstructed from weed seed assemblages.
Archaeological and Anthropological Sciences 4: 1–14.
Nasu H, Momohara A, Yasuda Y et al. (2007) The occurrence and
identification of Setaria italica (L.) P. Beauv. (foxtail millet)
grains from the Chengtoushan site (ca. 5800 cal BP) in cen-
tral China, with reference to the domestication centre in Asia.
Vegetation History and Archaeobotany 16: 481–494.
Pechenkina EA, Ambrose SH, Ma XL et al. (2005) Recon-
structing northern Chinese Neolithic subsistence practices
by isotopic analysis. Journal of Archaeological Science 32:
Perry CA and Hsu KJ (2000) Geophysical, archaeological, and
historical evidence support a solar-output model for climate
change. Proceedings of the National Academy of Sciences of
the United States of America 97: 12433–12438.
Piperno DR, Ranere AJ, Holst I et al. (2000) Starch grains reveal
early root crop horticulture in the Panamanian tropical forest.
Nature 407: 894–897.
Piperno DR, Ranere AJ, Holst I et al. (2009) Starch grain and phy-
tolith evidence for early ninth millennium B.P. maize from
the Central Balsas River Valley, Mexico. Proceedings of the
National Academy of Sciences of the United States of America
106: 5019–5024.
Qin L (2012) Archaeobotanical research and prospects on the
origin of agriculture in China. In: School of Archaeology
and Museology, Centers for the Study of Chinese Archaeol-
ogy (ed.) A Collections of Studies on Archaeology, vol. 9.
Beijing: Cultural Relics Press, pp. 260–315 (in Chinese).
Si LZ, Chen JY, Huang XH et al. (2016) OsSPL13 controls grain
size in cultivated rice. Nature Genetics 48: 447–456.
Silva F, Stevens CJ, Weisskopf A et al. (2015) Modelling the
geographical origin of rice cultivation in Asia using the rice
archaeological database. PLoS ONE 10(9): e0137024.
Spengler R, Frachetti M, Doumani P et al. (2014) Early agricul-
ture and crop transmission among Bronze Age mobile pasto-
ralists of Central Eurasia. Proceedings of the Royal Society B:
Biological Sciences 281: 20133382.
14 The Holocene 00(0)
Stevens CJ, Murphy C, Roberts R et al. (2016) Between China
and South Asia: A Middle Asian corridor of crop dispersal and
agricultural innovation in the Bronze Age. The Holocene 26:
Su BQ and Yin WZ (1981) Guanyu kaoguxue wenhua de quxi
leixing wenti [On the issue of the distribution and develop-
ment of regional cultures in Chinese archaeology]. Wenwu 5:
10–17 (in Chinese).
Vavilov NI and Freier F (1951) Studies on the Origin of Culti-
vated Plants. Buenos Aires: Acme Agency.
Wang XG and Xu X (2003) A discussion on the rice-millet
blended zone in the Neolithic Age. Agricultural History of
China 22: 3–9 (in Chinese).
Weber S, Lehman H, Barela T et al. (2010) Rice or millets: Early
farming strategies in prehistoric central Thailand. Archaeo-
logical and Anthropological Sciences 2: 79–88.
Weisskopf A, Deng ZH, Qin L et al. (2015a) The interplay of mil-
lets and rice in Neolithic central China: Integrating phytoliths
into the archaeobotany of Baligang. Archaeological Research
in Asia 4: 36–45.
Weisskopf A, Qin L, Ding JL et al. (2015b) Phytoliths and rice:
From wet to dry and back again in the Neolithic Lower Yang-
tze. Antiquity 89: 1051–1063.
Willcox G (2013) The roots of cultivation in southwestern Asia.
Science 341: 39–40.
Wu WX, Zheng XM, Lu GW et al. (2013) Association of func-
tional nucleotide polymorphisms at DTH2 with the north-
ward expansion of rice cultivation in Asia. Proceedings of the
National Academy of Sciences of the United States of America
110: 2775–2780.
Wu Y, Jiang LP, Zheng YF et al. (2014) Morphological trend anal-
ysis of rice phytolith during the early Neolithic in the Lower
Yangtze. Journal of Archaeological Science 49: 326–331.
Xue C, Zhou Y and Zhu X (2004) The Huanghe River course and
delta from end of Late Pleistocene to the 7th century BC. Acta
Oceanologica Sinica 26: 48–61.
Yan WM (1987) The unity and diversity of Chinese prehistoric
culture. Wenwu 3: 38–50 (in Chinese).
Yan WM (1997) The cradle of oriental civilization. In: Yan WM
(ed.) The Origins of Agriculture and the Rise of Civilization.
Beijing: Sciences Press, pp. 148–174 (in Chinese).
Yang XY, Wan ZW, Perry L et al. (2012a) Early millet use in
northern China. Proceedings of the National Academy of Sci-
ences of the United States of America 109: 3726–3730.
Yang XY, Zhang JP, Perry L et al. (2012b) From the modern to the
archaeological: Starch grains from millets and their wild rela-
tives in China. Journal of Archaeological Science 39: 247–254.
Yang XY, Zhuang YJ, Wang WW et al. (2016a) New radiocar-
bon evidence on early rice consumption and farming in South
China. The Holocene. Epub ahead of print 25 November.
DOI: 10.1177/0959683616678465.
Yang YZ, Cheng ZJ, Li WY et al. (2016b) The emergence, devel-
opment and regional differences of mixed farming of rice and
millet in the upper and middle Huai River Valley, China. Sci-
ence China Earth Sciences 59: 1779–1790.
Yang YZ, Li WY, Yao L et al. (2016c) Plant food sources and
stone tools’ function at the site of Shunshanji based on starch
grain analysis. Science China Earth Sciences 59: 1574–1582.
Zhang C and Huang H-C (2010) The emergence of agriculture in
southern China. Antiquity 84: 11–25.
Zhang C and Huang H-C (2013) Jiahu 1: Earliest farmers beyond
the Yangtze River. Antiquity 87: 46–63.
Zhang C and Huang H-c (2014) East Asia: Archaeology. In: Ness
I and Bellwood P (eds) Encyclopedia of Human Migration,
vol. 1: Prehistory. New York: Wiley-Blackwell, pp. 209–216.
Zhang H, Bevan A, Fuller DQ et al. (2010a) Archaeobotanical
and GIS-based approaches to prehistoric agriculture in the
upper Ying valley, Henan, China. Journal of Archaeological
Science 37: 1480–1489.
Zhang JP, Lu HY, Gu WF et al. (2012) Early mixed farming of
millet and rice 7800 years ago in the Middle Yellow River
Region, China. PLoS ONE 7(12): e52146.
Zhang JP, Lu HY, Wu NQ et al. (2010b) Phytolith evidence for
rice cultivation and spread in Mid-Late Neolithic archaeologi-
cal sites in central North China. Boreas 39: 592–602.
Zhang JZ, Kong ZC and Liu CJ (1994) Rice remains at prehistoric
sites in Wuyang County and prehistoric agriculture of Huang-
huai regions. Agriculture Archaeology 1: 68–75 (in Chinese).
Zhao ZJ (2006) A mixed farming system with both rice and mil-
let as the major crops during the Late Neolithic Period in the
southern Haidai Area. Orient Archaeology 3: 253–257 (in
Zhao ZJ (2011a) Characteristics of agricultural economy dur-
ing the formation of ancient Chinese civilization. Journal of
National Museum of China 1: 19–31 (in Chinese).
Zhao ZJ (2011b) New archaeobotanic data for the study of the
origins of agriculture in China. Current Anthropology 52:
Zhao ZJ and Chen J (2011) Results of the flotation carried out at
the site of Yingpanshan in Maoxian County, Sichuan. Nan-
fang Wenwu 3: 60–67 (in Chinese).
Zhao ZJ and Jiang LP (2016) Analysis of plant remains floated
from Shangshan site in Puyang, Zhejiang Province. Cultural
Relics of Southern China 3: 110–116 (in Chinese).
Zhao ZJ, Pearsall DM, Benfer RA et al. (1998) Distinguishing
rice (Oryza sativa Poaceae) from wild Oryza species through
phytolith analysis. II. Finalized method. Economic Botany 52:

Supplementary resource (1)

... Foxtail millet is common in the eastern regions of China throughout the Neolithic, however, it has not been frequently found in sites in Guanzhong until the Mid-Neolithic period [125]. Early evidence of mixed rice-millet farming has been discovered in Peiligang cultural sites in eastern and central China [53,126]. The westward spread of crops seems to have occurred along with certain types of pottery, by the gradual migration of small communities of farmers [126]. ...
... Early evidence of mixed rice-millet farming has been discovered in Peiligang cultural sites in eastern and central China [53,126]. The westward spread of crops seems to have occurred along with certain types of pottery, by the gradual migration of small communities of farmers [126]. ...
... Climatic deterioration had a significant impact on agriculture. Foxtail millet became the most important crop, although broomcorn millet and rice were also cultivated in some areas [126,139]. The reorganization of agricultural production may have led to an adjustment of subsistence strategies in different regions in response to various degrees of climatic deterioration [140,141]. ...
Full-text available
Transitions in animal exploitation patterns are caused by topographical and climatic varia- tions on both macro and micro scales. This paper presents temporally and spatially contextualized faunal profiles from 27 sites in the Guanzhong (关中) region of Shaanxi province (陕西省), PRC which date from the Early Neolithic to the Bronze Age (ca. 6000–1000 BCE). Climatic and environmental data was cross-referenced with archaeological, archaeobotanical and (where appropriate) historical sources to examine the reasons for the clear micro-transitions observed. Faunal profiles from sites in the Wei River plain (渭河盆地), loess plateau, and the transitional zone between them were ana- lyzed. Animal utilization was found to vary substantially between different zones during the period under analysis. The transition in praxis between the Wei River valley and the loess plateau was not gradual. The hilly transition zone was found to have its own distinct animal exploitation pattern. These spatio-temporal differences in animal exploitation were caused by changes in both the local microclimates and the topography of the landscape in which the communities were living. Some regions apparently reverted to ‘earlier’ animal exploitation patterns in response to climatic changes. These environmental factors were also augmented by internal social developments and interactions with neighboring communities.
... The Huai River region is located between the Yellow and Yangtze Rivers respectively [6][7][8][9]. The Huai River region is located between the Yellow and Yangtze Rivers and is considered a mixed rice-millet farming area [10,11]. Mixed rice-millet farming emerged in northern China no later than 7500 cal. ...
... Land 2022, 11, 831 ...
Neolithic rice remains were recovered from a mixed rice–millet farming area in China outside the original centers of rice farming. Whether the rice remains were the result of local cultivation or obtained through trade remains unclear. Rice paddy fields are direct evidence of local cultivation. In this study, phytolith samples from the Zhangwangzhuang site were analyzed. The discriminant function distinguished 17 of 30 samples in the suspected paddy field area as rice paddy fields with an average probability of 74%; The proportion of rice bulliform phytoliths with ≥9 scales indicated that rice (Oryza sativa) was still being domesticated and, moreover, six η-type phytoliths from broomcorn millet (Panicum miliaceum) were identified. These results suggested that the suspected paddy field at Zhangwangzhuang might be the earliest rice paddy field (ca. 6000 cal. BP) in northern China and that mixed farming was practiced here since the early Yangshao period. This study adopted discriminant analysis methods to discover ancient rice paddy fields, observed rice paddy fields outside the core rice origin area, and provided the earliest evidence regarding the development of mixed rice–millet farming in the upper Huai River region.
... Mixed cultivation of millet and rice emerged as early as the Peiligang culture period (7000-5000 BCE) in the Zhengluo region (He et al., 2017;Wang et al., 2017;Bestel et al., 2018;Zhang et al., 2012). However, due to the absence of bone stable isotope data, the effect of rice cultivation on human subsistence of Peiligang culture in the Zhengluo region cannot be elucidated. ...
Full-text available
Understanding the interplay between subsistence strategies and settlement patterns is fundamental for understanding past social-economic and cultural changes. Around the third millennium BC in the Zhengluo region of China, settlement hierarchies with two or three-tiered settlements were established. However, the detailed economic dynamics of settlement stratification in this area have not yet been clearly illustrated. In this paper, we report the new human isotopic data and archaeobotanical evidences with radiocarbon dates from archaeological sites of different scales in the Zhengluo region. The primary objective is to further assess the manner in which human populations among different scales of settlements in the Zhengluo region organized their subsistence strategies during the late Yangshao period (around 3500BC-2600BCE) by combining these new data with previous archaeological studies. Mixed crop cultivation including millet crops and non-millet crops (rice and/or soybean) was a stable agricultural pattern in center settlements, and pure millet agricultural only appeared in ordinary settlements. Meanwhile, human dietary patterns from center settlements were more diversified compared with the homogenous diets among inhabitants of the ordinary settlements. The distinct human dietary patterns at these two different types of sites results from and reflect their different responses of subsistence strategies to social dynamics. As settlements expanded and populations increased, stable millet cultivation guaranteed the essential food supply, and that the popularization of rice (a crop from the South) and the adoption of a new crop, soybeans , could diversify the agricultural patterns, expand the food supply and reduce the risks associated with food production in center settlements. Meanwhile, ordinary settlements still persisted with a more conservative subsistence strategy focused on millet agriculture. The different and successful coping strategies by late Yangshao inhabitants in Zhengluo region were the driving force in the emergence of the episode of peak social development in this area and likely important for populations and settlements in the following revolutionary periods "Longshan-Erlitou" in the same area.
... Recent studies have confirmed that foxtail millet (Setaria italica) and broomcorn millet (Panicum miliaceum) were domesticated in northern China and rice (Oryza sativa) in southern China Larson et al., 2014;Lu, 2017;Lu et al., 2009b;Yang et al., 2012). Subsequently, interactions between these two regions facilitated the formation of a dualistic farming tradition in a vast area of China (Deng et al., 2018a;He et al., 2017), which then dispersed southward into Southeast Asia and beyond, to profoundly influence subsequent social developments in these regions (Gao et al., 2020). Therefore, the pathway of crop domestication and the formation of mixed farming in China is central to the study of early agricultural practices. ...
Full-text available
Located in the transitional region between northern and southern China, the middle Han River valley is of pivotal significance for current debates on rice domestication and the formation of mixed farming in early China. Based on phytoliths and macro-plant remains from 10 sites, along with radiocarbon dating, this study has narrowed the 2000-year gap in early rice utilization by 700 years in the middle Han River valley, and demonstrated the continuous development of rice agriculture in this region. This new evidence also reveals that as early as 7000 cal. BP, mixed farming emerged in this region and facilitated the further southward dispersal of foxtail millet and broomcorn millet into southern China and beyond. In addition, a sample-by-sample comparison of our results with phytoliths and macro-plant remains highlighted the potential biases inherent in these methods. This work emphasized the importance of multi-index archaeobotanical analysis to reconstruct ancient crop assemblages.
... Foxtail millet and broomcorn millet were domesticated in Northern China around 10,000 BP (Lu et al., 2009;Yang et al., 2012;Zhao et al., 2020). Millet agriculture was widespread during the Neolithic to Bronze Age in the Chinese Loess Plateau area (He et al., 2017) and was also scattered throughout Taiwan (Tsang et al., 2017), Fujian (Fu et al., 2016;Zhou et al., 2017;Deng et al., 2018;Dai et al., 2021) and Jiangxi (Chen et al., 2015;Deng et al., 2020). In addition, wheat agriculture was also introduced into Central China at least 3,500 years ago (Zhao, 2011b;Chen, 2016). ...
Full-text available
As a global cooling event, many of the climatic and socio-cultural mechanisms that resulted in changes after the 2. 8 ka BP event remain unclear. In China, this period roughly corresponds with the Zhou Dynasty (1046-212 BC), a critical period when ancient Chinese civilization was experiencing significant cultural and technological changes, including the movement of people to modern-day Jiangsu Province, where they intensively used the natural resources found in this the coastal area. Recent archaeobotanical evidence, and two radiocarbon dates on wheat and foxtail millet, indicate that the Datongpu site, which dates around 2,600 cal a BP, was occupied during this period of transition around the 2.8 ka BP climate event. In total, our investigations recovered 3,399 carbonized seeds from seventy-four flotation samples, of which rice, foxtail millet, broomcorn millet, and wheat seeds where predominant along with 2,296 weed seeds. Additionally, we identified several rice spikelets and wheat rachises. The high number of carbonized rice grains indicates that rice farming was the primary crop in an otherwise mixed rice-dry farming system at Datongpu. In addition, we argue that the “2.8 ka BP cold event” probably influenced population growth and caused food shortages throughout Central China, leading people to migrate southeastward along the Huai River to the coastal areas of Jianghuai Region. We argue that this abrupt shift in the climate indirectly facilitated the exploitation and emergence of large-scale agriculture in this area. Our study provides an example for the indirect impact of climate change in areas with relatively favorable climate conditions.
... In this case, the agriculturalization of other regions was all realized through population expansion and agriculture dispersal from these two centers. With the progress in archaeological research over the past decades, it has become increasingly evident that the southward dispersal of agriculture in China was mainly along three main routes, among which the Zang-Yi Corridor is of critical significance in subsistence transformation and social development in southwest China, including the Chengdu Plain, the Yunnan-Guizhou Plateau, and the Tibetan Plateau (He et al., 2017;Deng et al., 2018;Gao et al., 2021). Further influence of this route could even have reached mainland Southeast Asia and joined the longdistance communications along the southern foothills of the Himalayan Mountains with south Asia and beyond (Gao et al., 2020). ...
Full-text available
The Zang-Yi Corridor is of pivotal significance for the interactions between northwest China, southwest China, and mainland Southeast Asia. It has been hypothesized that the formation of mixed farming in this region and its surrounding areas was based on multiple waves of crop dispersal, with foxtail millet and broomcorn millet arriving first from northwest China around 5,300 cal. BP and rice from middle Yangtze valley after 4,700 cal. BP. Based on the systematic sampling and direct dating conducted at the Guijiabao site, Sichuan Province, this study demonstrates that by no later than 5,000 cal. BP, mixed farming had already emerged in the south part of Zang-Yi corridor, which was much earlier than expected before. With this new evidence, it is argued that the transformation into farming in Southwest China was based on the dispersal of a crop package comprising foxtail millet, broomcorn millet, and rice instead of different waves of introduction. A further comparison of all archaeobotanical data in this region revealed that crop patterns varied significantly between different sites because of their diverse environmental conditions.
This study examines the archeological seed assemblage of Anle, a middle Neolithic site located in the Lower Yangtze region, China. The Lower Yangtze is thought to be the origin of domesticated rice and most studies of this region to date have focused on rice domestication and cultivation within its paleoenvironmental setting. In contrast, we highlight here diverse uses of non-rice plant resources. In addition to large quantities of rice remains (carbonized grains and spikelet bases), we identify both foxtail and broomcorn millet, both AMS radiocarbon dated earlier than 5750 cal BP, demonstrating the dispersal of millet cultivation to the Lower Yangtze in the middle Neolithic, earlier than previously securely documented. While most wild species identified in macrobotanical assemblages are traditionally categorized as weeds or incidental intrusions among food residues, many can be exploited for food and medicinal purposes. By analyzing the ecological and functional implications of identified plants, we infer ecological niches of cultivation, gathering, and possible propagation of wild plants as food and medicine. Analyses of diversity and seasonality of plant resources identified show that residents of Anle created a complex seasonal sequence of temporally compatible crops, constructing niches for two crops (rice and millet) and actively structuring opportunities to exploit available wild plant resources in their immediate environment.
Full-text available
This is just the bibliography
Full-text available
В монографии представлены материалы, отражающие процесс формирования неолита и земледелия в Китае. В первой главе рассматриваются особенности эпохи, переходной от палеолита к неолиту, много внимания уделено проблемам, связанным с установлением хронологии древнейшей керамики Китая. В главе, посвященной раннему неолиту, дается обзор раннеземледельческих культур Китая, их жилищ, поселений, погребальных и иных обрядов, керамической посуды и прочего инвентаря. В отдельной главе подробно рассматриваются обстоятельства и особенности формирования земледелия в Китае, анализируются текущие источники, а также сложившиеся подходы к их пониманию и интерпретации. Наконец, в последней главе разбираются возможные связи раннеземледельческих культур Китая с ранними земледельцами Западной Евразии и культурой дземон Японских островов.
Understanding the role of agriculture in the development of human societies around the world is an important field of study with many unanswered questions. As a step toward that greater understanding, we have studied the archeobotanical remains at the Jingshuidun site in the mountainous areas of the lower reaches of the Yangtze River in southern Anhui Province which are part of the core area of rice cultivation today. Our analyses of macrobotanical remains and phytoliths formed the basis for the reconstruction of the subsistence economy of ancient humans at the Jingshuidun site from the late Neolithic to early historical times. When our data are combined with that of previous archeobotanical work, we obtain a clearer picture of the development of rice and millet agriculture in the southern Anhui Province region, as well as the spread of millet cultivation. Macrobotanical remains and phytoliths of domesticated rice are present in layers at the Jingshuidun site dated to 4874–4820 cal. yr B.P. (middle-late Liangzhu Period) and 2667–2568 cal. yr B.P. (late Western Zhou Dynasty to the early Spring and Autumn Period). Moreover, macrobotanical remains and phytoliths from the site document the earliest remains of foxtail millet (Setaria italica) in southern Anhui Province, from a layer dating to the late Western Zhou Dynasty and the early Spring and Autumn Period (2667-2568 cal. yr B.P.). These results suggest that the people occupying the Jingshuidun site used single rice farming as far back as 4874–4820 cal. yr B.P., and they began to plant millet by at least 2667–2568 cal. yr B.P., documenting the spread of millet agriculture to the southern area by that time.
Full-text available
While the possibility of indigenous rice cultivation cannot be entirely ruled out, there is increasing evidence suggesting that rice farming was introduced to South China during the late-Holocene. However, determination of the exact timing of the spread of rice farming to South China is fraught with the lack of reliable radiocarbon dates. In this article, we present 15 new AMS ¹⁴C dates of charred plant remains recovered from late-Holocene sites of Shixia and Guye in Guangdong Province of South China. Our new AMS ¹⁴C dates suggest a later arrival of rice farming in the Pearl River Delta than previously thought. These new AMS ¹⁴C dates will shed new lights to an improved understanding of the environmental background and ecology of the southward spread of rice farming.
Full-text available
Mixed farming of rice and millet is one of the basic agricultural modes in the upper and middle Huai River Valley (HRV). According to the latest data, this agricultural mode appeared during the middle and late Peiligang Culture (7.8–7.0 ka BP) in the upper HRV, and then became a common subsistence economy in the end of the Neolithic (5.0–4.0 ka BP) in both the upper and middle HRV. However, it is still not clear how this mixed farming developed in the upper HRV after its occurrence, nor are the regional differences in the development of mixed farming between the upper and middle HRV during the Neolithic completely understood. In this paper, flotation and starch analyses were conducted on samples from eight archaeological sites in the upper and middle HRV. The results indicate that the mixed farming of rice and millet first appeared in the later phase of the middle Neolithic in the regions of the Peiligang Culture, then developed quite rapidly in the late Neolithic (6.8–5.0 ka BP), finally becoming the main subsistence economy at the end of the Neolithic in the upper HRV. However, there are obvious differences in the emergence and development of agriculture between the middle and upper HRV. Rice farming was the only agricultural system during the middle Neolithic, lasting until the end of the Neolithic, when mixed farming appeared in the middle HRV. Furthermore, although mixed farming appeared in both the upper and middle HRV during the end of the Neolithic, the roles of rice, foxtail millet and broomcorn millet in the subsistence economy were not the same. In general, millet was more widely cultivated than rice in the upper HRV, but rice occupied the same or a slightly more prominent position in the middle HRV at the end of the Neolithic. These results are significant for understanding the process of agricultural development and transformation, as well as human adaptation to climatic and cultural variability duringthe Neolithic.
Full-text available
The upper-middle Huai River valley (HRV) is located in the transition zone in the middle of eastern China. Previous studies have shown that the past populations in the upper HRV obtained plant food through gathering and cultivation of both rice and millet during the middle Neolithic. However, for the middle HRV it is not very clear what methods were used by humans to obtain plant food in that time. In this paper, starch grain analysis was carried out on 17 stone tools and 29 cauldron pottery sherds unearthed at Shunshanji (8.5–7.0 kyr BP), which is known as the earliest Neolithic site in the middle HRV excavated so far. Here, ancient starches from Coix lacryma-jobi, Triticeae, Oryza sativa, Trichosanthes kirilowii and one unidentified specie were recovered. This study contributes to the limited knowledge of food strategies as observed in the middle HRV. It demonstrates that both gathering and cultivation of rice took place during the middle-Neolithic. Among the identified species, Coix lacryma-jobi appears to have been the main plant food at Shunshanji. Moreover, it is clear that Oryza sativa was not consumed as much as other plant species according to the occurrence frequency of different starch grains. Starch grains were also found on the used surfaces of grinding stone implements as well as pestles, which means that these stone tools were used for food processing. In addition, axes from Shunshanji may also have been used for food processing as well as wood working because starches were also found on the edge of axes. Information about subsistence strategies and tool use at Shunshanji will also be helpful to understanding the utilization of plants and agricultural development in the middle HRV during the early-middle Neolithic period.
Full-text available
The period from the late third millennium BC to the start of the first millennium AD witnesses the first steps towards food globalization in which a significant number of important crops and animals, independently domesticated within China, India, Africa and West Asia, traversed Central Asia greatly increasing Eurasian agricultural diversity. This paper utilizes an archaeobotanical database (AsCAD), to explore evidence for these crop translocations along southern and northern routes of interaction between east and west. To begin, crop translocations from the Near East across India and Central Asia are examined for wheat (Triticum aestivum) and barley (Hordeum vulgare) from the eighth to the second millennia BC when they reach China. The case of pulses and flax (Linum usitatissimum) that only complete this journey in Han times (206 BC–AD 220), often never fully adopted, is also addressed. The discussion then turns to the Chinese millets, Panicum miliaceum and Setaria italica, peaches (Amygdalus persica) and apricots (Armeniaca vulgaris), tracing their movement from the fifth millennium to the second millennium BC when the Panicum miliaceum reaches Europe and Setaria italica Northern India, with peaches and apricots present in Kashmir and Swat. Finally, the translocation of japonica rice from China to India that gave rise to indica rice is considered, possibly dating to the second millennium BC. The routes these crops travelled include those to the north via the Inner Asia Mountain Corridor, across Middle Asia, where there is good evidence for wheat, barley and the Chinese millets. The case for japonica rice, apricots and peaches is less clear, and the northern route is contrasted with that through northeast India, Tibet and west China. Not all these journeys were synchronous, and this paper highlights the selective long-distance transport of crops as an alternative to demic-diffusion of farmers with a defined crop package.
The body of data based on new work on genetics and DNA, plus a growing number of radiocarbon ages which are independent of dates based on cultural associations has broadened our knowledge of domestication in eastern Asia. Here we review the situation for several plant and animal species that were domesticated locally or imported to east Asia. Major centres of plant domestication in China have been in the Yellow and Yangtze river basins, and in Yunnan. For animals it appears that the Yellow River region, the area around Xi'an and the south-east have been important centres. Many adopted domesticates have entered China through the north-west and later through ports such as Canton (Guangzhou). It appears that while there are outliers to extended ranges of wild plants and animals, sometimes not securely dated, widespread deliberate movement of plants and animals outside their natural ranges coincided with reduced hunting and gathering around 5–4 kyr in the Longshan cultural period. The adoption of agriculture has resulted in large scale landscape transformation; forests and woodlands have been replaced by crop and grazing lands and this is evident in many late Holocene sedimentary records. This transformation continues and the patterns are changing as diets are shifting and much grain is now used to feed chicken and beef, and in addition this has placed increased pressure on water resources.