Full text of DOI:10.1017/S0003598X00061378

Agricultural origins in the Korean
Peninsula
Gary W. Crawford and Gyoung-Ah Lee*
The authors report the first direct scientific evidence for the beginnings of agriculture in
the Korean peninsula.
Keywords: Korea, Chulmun, Mumun, archaeobotany, agriculture
Introduction
The development of agriculture in Korea, as elsewhere, is an important part of landscape
evolution and Holocene human adaptation and deserves study in its own right. Furthermore,
Korean archaeology can help to clarify the mechanisms and timing of the spread of agriculture
through Northeast Asia and the relative independence of developments in East Asia. In
particular, Yayoi origins in Japan and the extent to which early crop production in Japan
developed independently of mainland processes are open questions and dependent on a
critical assessment of events in Korea. However, understanding agricultural origins in Korea
has been handicapped by a reliance on circumstantial evidence and inadequate documentation.
Whereas archaeobotanical data attest to the development of agriculture in neighbouring
China and Japan, such evidence has been rarely collected in the Korean Peninsula (Korea).
In consequence, little is known about early Korean agricultural history, except that between
1000 and 500 BC there was intensive crop production focused on crops apparently introduced
from China (Choe 1982: 526; Nelson 1991: 106).
We report here the first contextually clear and directly dated sequence of early crop remains
recovered from Korea. Our research focuses on the Middle Chulmun through Middle Mumun
periods in South Gyeongsang Province, ranging in date from c 3000 to c 1000 BC. The term
‘Chulmun’ refers to the incised decoration on pottery common at the time, and this period
is usually considered the Neolithic of Korea. But in Northeast Asia, ‘Neolithic’ refers only to
the presence of pottery with stone technology and has no necessary connection with
agriculture. ‘Mumun’ refers to the plain ware that appears at the end of the Chulmun: it is a
period of significant change in Korea which includes the introduction of metallurgy.
The archaeobotanical data reported here have been recovered from the Tongsamdong shell
midden in Busan, the Daundong site in Ulsan, and several localities within a continuous
stretch of the Nam River in Jinju (Oun 1, Okbang 1, 2, 4, 6 and 9, Sangchon B) (Figure 1).
In the Nam River project, numerous pit-houses, exterior hearths, craft workshops, dolmens,
stone-cist burials, and dry fields surrounded by ditches and palisades have been exposed in a
*Department of Anthropology, University of Toronto at Mississauga, Mississauga, Ontario, Canada L5L 1
Received 19 December 2001; Revised and accepted 21 January 2003.
87

Gary W. Crawford and Gyoung-Ah Lee
400 ha area. In addition to documenting
early coastal adaptation in Korea and
contacts with Jomon cultures in the Japanese
archipelago, the Tongsamdong site has
played a key role in the development of the
South Korean prehistoric chronology (Jeong
1997). Its assemblage includes Chulmun
pottery, stone tools, Early to Final Jomon
pottery and obsidian from Japan, shell
ornaments, and composite bone fish hooks
(Sample 1974). Other sites help to continue
the sequence. Daundong is an early Middle
Mumun occupation and burial site. Okbang
is another Mumun community. Sangchon
B and Oun 1 have both Chulmun and
Mumun settlements. Flotation samples were
collected from all the sites, although we were
able to take only one sample each from
Tongsamdong and Daundong. Crop
Figure 1. Map of the Korean Peninsula
remains comprise nearly 70 percent or more of the seed assemblage from the Middle Chulmun
through Middle Mumun periods (Table 1).
Dry-field crops
The results show that dry-field crops preceded wet-field production in the Korean Peninsula
and continued to be important long after wet-field systems developed. The oldest examples
are foxtail millet and broomcorn millet from
Tongsamdong (for location see Figure 1),
in a stratified sequence with a series of
radiocarbon dates spanning the period from
about 4800 to 1700 cal. BC (Sample 1974).
Hoe-like stone tools appear in the Tudo
phase suggesting that agriculture had begun
at that time (Sample 1974: 118). TheTudo
phase, at one time dated to 1700 cal. BC,
is now dated to 3000 cal. BC (Middle
Chulmun) (Shin 1997). Until now, plant
remains had not been collected by flotation
at Tongsamdong to test the hypothesis of
Tudo phase agricultural origins. The single
flotation sample, from floor fill of a Middle
Chulmun semi-subterranean pit-house, has
a millet density of six grains per litre. An
Figure 2. A) Rice from Okbang 1; B) wheat from Oun 1; C)
foxtail millet from Tongsamdong and D) beefsteak plant from
Okbang 1.
AMS date of 3360 cal. BC on foxtail millet grains confirms their Middle Chulmun association
(Figures 2C and 3).
88

Agricultural origins in the Korean Peninsula
Table 1. Plant Taxa and Percent Representation in South Korean Flotation Samples
EARLY
MUMUN
MIDDLE
MUMUN
MIDDLE
MUMUN
CHULMUN CHULMUN
TONGSAM-
COMMON NAME
SCIENTIFIC NAME
DONG
NAM R.
NAM R. DAUNDONG NAM R.
WILD/WEEDS
fleshy fruit
azuki (probably)
0.4
1.9
2.3
0.8
0.1
1.2
Vigna angularis ssp.
nipponensis
Echinochloa crus-galli
Chenopodium sp.
Setaria sp.
Polygonum sp.
Panicum sp.
Paniceae
0.2
A
barnyard grass
chenopod
foxtail grass
knotweed
panic grass
panicoid grass
Other weeds
EAST ASIAN CROPS
azuki
broomcorn millet
foxtail millet
soybean
36.8
3.0
0.4
0.9
1.9
0.5
5.1
12.6
0.5
0.9
2.5
0.5
0.7
15.0
3.6
1.4
0.4
1.4
3.3
0.4
4.8
4.8
0.2
Vigna angularis
Panicum miliaceum
32.1
0.8
7.8
46.5
12.7
1.9
6.4
48.4
4.6
9.1
13.1
46.3
9.8
71.5
0.1
Setaria italica ssp. italica 35.5
Glycine max
rice
Oryza sativa
A
5.9
3.7
shiso/egoma
WEST ASIAN CROPS
barley
bread wheat
UKNOWN
Perilla frutescens
0.3
0.5
Hordeum vulgare
Triticum aestivum
0.4
1.2
2.8
0.4
2.1
5.6
A
0.1
4616
5.2
231
15.0
214
TOTAL NUMBER
2239
1072
A: less than 0.1 percent
Figure 3. Calibrated AMS dates: white box represents 2 sigma range; grey box represents 1 sigma range. F: foxtail millet, R:
rice, S: soybean, V: Vigna, W: wheat.
89

Gary W. Crawford and Gyoung-Ah Lee
Daundong and Okbang have produced the earliest substantial archaeological specimens of
soybean and Vigna sp. from early East Asia (Table 1 and Figure 3). Soybean recovered from
Okbang 1 dates to 1000 – 900 cal. BC and Vigna from Daundong to 760–600 cal. BC.
Domesticated soybean has high diversity due to the large number of land races (Hymowitz
& Singh 1987) so we are unable to link the archaeological remains to a specific soybean type
based on seed morphology.The wild ancestor of soybean, Glycine soja, is common in northeast
China, Korea, and Japan (Hymowitz & Singh 1987) but the Daundong Glycine (Figure 4A)
is not G. soja. Its morphology is within the range of domesticated soybean, G. max. and
indistinguishable from later soybean from Okbang 6. The soybean revolutionized Late Zhou
Chinese agriculture following its introduction from the northeast apparently in 664 BC (Ho
1977: 452). Archaeological soybean appears in Japan during the Yayoi period and post-dates
400 BC (Crawford 1992: 28). The Daundong sample is from floor fill collected from the
eastern portion of a pit-house of the Middle Mumun period, with a seed density of 660 per
litre. The high density of soybeans from a house floor substantiates the economic importance
and utilization of soybean during the Mumun. The AMS dates support Ping-Ti Ho’s
hypothesis that soybean was domesticated by, and economically important to, peoples in or
near the Manchurian Plain (Ho 1977: 452, 454). Our results indicate that the region of
early soybean importance includes Korea.
The role of Vigna in early Chinese agriculture is not known but it has an archaeological
record in Japan starting in the Early Jomon. Originally thought to be mung bean (V. radiata
Figure 4. The archaeological soybean specimens, ventral view (A1 and A2), exhibit wide variation from flat-ovate to ovate in
cross-section. Clearly visible on many specimens is the hypocotyl-radicle (embryonic stem and root) and hilum. The mean
dimensions of the soybeans are 7.3 mm long (n=323), 5.2 mm in breadth (n= 302), and 4.3 mm thick (n=186). B) is modern
reference azuki; C) is a charred reference mungbean; D) a modern Vigna radiata (mung bean) cotyledon E) is a Vigna
cotyledon from Daundong. The plumule-hypocotyl (embryonic leaves and stem) shape and size relative to the cotyledon appear
to distinguish East Asian azuki from the South Asian mung bean (Vigna radiata) (Yoshizaki 1992). The morphology of the
archaeological Vigna specimens resembles that of azuki rather than that of the mung bean.
90

Agricultural origins in the Korean Peninsula
var. radiata), the Jomon period Vigna is more likely azuki or wild azuki (Vigna angularis)
(Crawford 1992: 26, 30). The specific identification of the archaeological Vigna specimens is
not clear (Figures 4D and E) but they seem to be azuki whose ancestor (Vigna angularis ssp.
nipponensis) is indigenous to East Asia (Yamaguchi 1992). Domesticated azuki, like soybean,
has larger seeds than its wild ancestor. We hypothesize that selection for the large azuki beans
common among cultivars today had not yet occurred. The Daundong beans show a wide
range of size between specimens of wild and domesticated azuki (Figure 5) consistent with
being an early
d o m e s t i c a t e d
variety.
A form of dwarf
bread wheat with
seeds smaller than
those from any
other
compact
variety is present in
a r c h a e o l o g i c a l
collections in Japan
and early historic
Korea (Crawford
1992: 26). Wheat is
present no earlier
than the Yayoi
period in Japan
(Crawford 1992).
The oldest Korean
wheat is associated
with millet and rice
AMS dated to ca.
Figure 5. Plot of length vs. width of Vigna from Daundong. The size of the Daundong Vigna
specimens is extremely variable. None are as large as modern domesticated azuki. The Vigna
appears to be one species with highly variable cotyledon size. Mean length is 3.8 mm (n=653)
and mean width is 3.0 mm (n=652).
1000 cal. BC in Mumun contexts at Oun 1 and Okbang 1 (Table 1 and Figure 2). The wheat
grains in the Nam River project samples, averaging 3.9 by 2.5 by 1.9 mm, are the same
extremely compact bread wheat reported in Japanese assemblages (Crawford 1992).
Beefsteak plant (Table 1 and Figure 2) is another potential member of the Early/Middle
Mumun crop complex in the Nam River. Beefsteak plant is grown today for its aromatic
leaves, oil seeds and its medicinal properties in China, Japan, and Korea (Crawford 1992:
28). Although the main area of the plant’s diversity is China, no evidence for it has been
recovered from Chinese archaeological sites (Crawford 1992). In Japan, seeds of beefsteak
plant have been reported from Early through Final Jomon sites (Kasahara 1983; Matsutani
1983: 183).
Excavations in the Nam River research area have exposed 1.8 ha of dry fields. The fields
date to the Early and Middle Mumun as well as theThree Kingdom period based on associated
artefacts, structures, and stratigraphy. The fields consist of pairs of ridges and furrows 20 to
80 cm wide. Dry fields were located in the floodplains along the large rivers (Lee & Lee
2001). Foxtail and broomcorn millet, legumes, barley, wheat, and beefsteak plant remains
91

Gary W. Crawford and Gyoung-Ah Lee
have all been recovered from Early and Middle Mumun houses and hearths near the dry
fields.
Wet-field agriculture
Rice remains are reported from over 50 sites in Korea (Ahn 1996) and occur in many of the
sites in this study. Undated rice caryopses from non-cultural layers in peat in central Korea,
large Poaceae pollen grains, and Oryza-type phytoliths in archaeological contexts suggest rice
cultivation in Korea as early as 1500 to 3000 cal. BC (Im 1990; Kim & Lee 1997; Kim
1980; Kwak et al. 1994; Lee & Park 1997). Pollen is weaker evidence than charred grain for
the presence of domesticated rice because all grass pollen is morphologically similar
(monoporate), pollen is not yet directly dateable, and wild rice may have a long history in
Korea (Crawford & Shen 1998). Domesticated rice phytoliths are distinguishable from wild
or weedy rice only on the basis of statistical analysis of substantial samples of specific types of
phytoliths (Zhao et al. 1998); phytoliths cannot yet be directly dated. Rice (kernels and
separate husks) has been reported together with foxtail millet, barley, wheat, hemp, and a
legume from a pit house floor at the Middle Chulmun site at Daecheonri, North Chungcheong
province in central Korea (Han et al. 2002). The specimens were carbonized except for the
rice husks. Radiocarbon dates of 4240±110 BP, 4400±60 BP, 4490±40, and 4590±70 BP (ie
c 2500–2200 BC) have been obtained from associated charcoal, but direct AMS dating of
individual grains would be advisable and published photographs of the specimens suggest
that the identification of hemp and legume is problematic. The uncarbonized rice hulls
suggest these plant remains are more recent than the Middle Chulmun.
Two AMS dates of 1950 and 1000 cal. BC on rice grains from Oun 1 are the first direct
evidence for the antiquity of rice in Korea (Table 1 and Figure 3). The 1000 cal. BC date is
contextually consistent with the Early Mumun context in which it was found. The older date
is also from an Early Mumun pit-house but the date is consistent with Late Chulmun. Oun
1 is intensively occupied and rice grains found in Mumun pit-houses may be residual from
the earlier Chulmun contexts they were cut through. Indeed, Late Chulmun pottery has also
been found in this Early Mumun house. However, no rice has been recovered from unequivocal
Chulmun contexts at Oun 1.
Early and Middle Mumun rice appears to have been grown in paddy fields. Among the ten
archaeologically identified paddy fields found in Korea, the two oldest are the Early Mumun
Okhyun and Yaumdong sites in Ulsan (Figure 1; Lee & Lee 2001). The sites are in narrow
valley flats at the base of low hills (Kwak 2000). The Mumun inhabitants took advantage of
the high ground-water level in the valleys and running water from higher elevations. Artificial
reservoirs and irrigation canals appear during the Middle Mumun (Lee 2000). At the Majeonri
site, a reservoir was created near the bottom of a hill, but elevated higher than the field to
facilitate a gravity-fed irrigation system (Lee 2000). At the Gwanchangri site, five reservoirs
were constructed along small streams. None of these are associated with large floodplains.
Later historic rice fields have been found not only in the narrow valley flats but also in back-
swamp areas and floodplains along large rivers, requiring advanced water control techniques
(Lee 2000).
92

Agricultural origins in the Korean Peninsula
Nuts
A prevailing view is that nuts were the mainstay of Korean Chulmun, as well as the Japanese
Jomon diets. However, during the Jomon in south-western Hokkaido, Japan, nuts are common
in the archaeological record until about 4000 cal. BC and subsequently became less common,
while annual grasses and weedy grain plants including a probable local domesticate, barnyard
millet (Echinochloa utilis), came into use (Crawford 1983). Few nut remains have been
recovered in our Chulmun samples. With the exception of one sample at Obkang 4, Mumun
assemblages have not produced nut remains either. In contrast, charred nuts in addition to
small seeds are common at the waterlogged Early Chulmun Sejukri shell midden (5400 to
4600 cal. BC) (Kim, Yoon & Ahn 2002) currently undergoing study. The sequence is thus
similar to that in north-eastern Japan and may indicate the onset of agricultural production.
Discussion
The directly dated examples of early cultigens from Korea can be placed in context by briefly
noting their occurrence in the neighbouring territories of China and Japan. Foxtail and
broomcorn millet, and rice, all domesticated in China, were of some significance there by
7000–6000 cal. BC (Early Neolithic) (Crawford 1992; Crawford & Shen 1998). Late
Neolithic compact wheat from Donghuishan, Gansu Province has been directly dated to
2880 calendar years BC (BA92101: Gansu Provincial Institute of Cultural Relics 1998: 190),
and barley may appear at about the same time in China (Crawford 1992) although actual
remains have not yet been found. Rice was domesticated in south-central China (Crawford
& Shen 1998) while millets originated in the north (Crawford 1992).
Rice, foxtail and broomcorn millet are present in north-eastern Japan by 900 cal. BC
(D’Andrea et al. 1995), so connections with the mainland were occurring sometime before
that. Less is known about the sequence of events in Kyushu, but at least two sites, Nabatake
and Itazuke (Figure 1) have produced evidence of relatively intensive agriculture by 300 cal.
BC. Millets appear to have been present at Nabatake between 1000 and 400 BC (Crawford
1992).
The new Korean data are consistent with a time-transgressive model with the earliest barley,
broomcorn and foxtail millet, rice, barley, and wheat appearing first in China, then in Korea
and then in Japan. The Korean Middle Chulmun period is roughly contemporaneous with
the Chinese Late Neolithic so crops were certainly available to Korean populations at that
time. Beginning in the Middle Chulman complex diffusion processes probably resulted in
the introduction of these Chinese and West Asian crops to the Korean Peninsula. Our study
confirms that the two varieties of millet were established no later than 3400 cal. BC (Middle
Chulmun) in the southern Korean Peninsula. Millets were probably available throughout
Korea before then because Tongsamdong is at the south-eastern extreme of the Korean
Peninsula, over 700 km overland from the likely source of the crop in north-eastern China.
Furthermore, at least ten weed taxa including grasses appear in samples from the Middle
Chulmun and later (Table 1). The new data indicate that Middle Chulmun people had a
mixed economy that included gathering plants, fishing, and hunting. This appears to have
changed little through the Late Chulmun as evidenced by the Nam River data. We suggest
rice as well as millet was grown during the Late Chulmun, but the evidence needs
93

Gary W. Crawford and Gyoung-Ah Lee
strengthening. Middle and Late Chulmun cultures exhibit little, if any, significant alterations
suggesting crops helped maintain Chulmun lifeways rather than alter them. The earliest
crops in Archaic period North America and Jomon period Japan also seem to have helped
maintain existing economic systems while imparting no obvious immediate changes.
Significant culture change did not occur until the Mumun period when azuki, barley,
beefsteak plant, millet, soybean, wet-rice production, and wheat appear. Azuki and soybean
may have been secondary domesticates after the introduction of millet production, or they
may have been introduced from as yet unidentified locations. For now, the only substantial
archaeological evidence for early soybean and azuki is from Japan and South Gyeongsang
Province, South Korea. The Early Mumun, like the Yayoi in Japan, represents agricultural
intensification rather than origins.
The evidence fromTongsamdong shows that obsidian and pottery were exchanged between
Korea and Japan throughout the life of the settlement, so crops may have been included in
the exchange system as well. But we do not rule out indigenous plant management leading
to domestication of indigenous crops in either Korea or Japan. The case for barnyard millet
has already been made (Crawford 1983). Flotation samples from earlier Jomon contexts in
Kyushu are needed, and the contexts of azuki, beefsteak plant, and soybean need further
study.
How agriculture developed in Korea and Japan is still an open question, but it certainly
involved diffusion from mainland East Asia. Korea is thus a secondary region of agricultural
origins. The best examples of this process outside Asia are in northern Europe, the south-
western U.S. and north-eastern North America where agriculture developed at different rates
and in cultural settings with varying degrees of mobility (Smith 1998). In Korea and Japan,
factors underlying primary agricultural origins such as association with resource-richness,
proximity to aquatic resources and low residential mobility were all operating when crops
were first in use.
Acknowledgements
This research has been funded by the University of Toronto and the Kyungnam University Museum. We would
like to thank the many individuals and institutions who made this project come to fruition. Logistical support in
the Nam River Archaeological Project was provided by Lee Sang-kil and the Kyungnam University Museum; the
Dong-A University Museum; the Dong-Ui University Museum; the Jinju National Museum; and the Kyungnam
Archaeological Research Institute. The Changwon National University Museum made available the Daundong
site sample for study. The Sejukri sample was made available by the Dongguk University Institute of Cultural
Properties, while the Municipal Museum of Pusan at Pokcheon permitted us to examine the Tongsamdong
samples. The National Crop Experiment Station, Suwon provided seed reference material. Alison Dias provided
the soybean drawing. A.C. D’Andrea and Ahn Sung-mo provided helpful feedback on an earlier version of this
paper.
CRAWFORD, G.W. 1983. Paleoethnobotany of the Kameda
References
Peninsula Jomon. Ann Arbor, Michigan: Museum
AHN, S.M. 1996. Retrospect and prospect on research
of Anthropology, University of Michigan.
of prehistoric agriculture in Korea. Prehistory and
Ancient History (Seonsa-wa Godae) 7.
–1992. Prehistoric plant domestication in East Asia, in
C.W. Cowan & P.J. Watson (ed.), The Origins of
Agriculture: An International Perspective: 17–38.
Washington and London: Smithsonian Institution
Press.
CHOE, J.P. 1982. The diffusion route and chronology of
Korean plant domestication. Journal of Asian
Studies 41:19–29.
94

Agricultural origins in the Korean Peninsula
CRAWFORD, G.W. & C. SHEN. 1998. The origins of rice
KWAK, J.C. 2000. Prehistoric wet-field rice agriculture
in Korea, in National Museum of Korea (ed.),
Ancient rice agriculture in Korea (the National
Museum of Korea Symposium proceeding): 69–107.
Seoul.
agriculture: recent progress in East Asia, Antiquity
72: 858–66.
D’ANDREA, A.C., ET AL. 1995. Late Jomon cultigens in
north-eastern Japan, Antiquity 69: 146–52.
KWAK, J.C., ET AL. 1994. Rice plant-opal extracted from
Neolithic pottery, Hanguk Kogo Hakbo (Journal of
the Korean Archaeological Society) 32: 149–162.
GANSU PROVINCIAL INSTITUTE OF CULTURAL
RELICS.1998. Excavation Report of the Donghuishan
site in Min-le Excavation and study on the Cemetery
of Siba Culture (Min-le Tung-hui shan K’ao);
Science Press.
LEE, H.J. 2000. Early wet-rice agriculture in Korea,
Korean Journal of Agriculture and Industry 42: 1–
13.
HAN, C. G., G. W. KIM, I. Y. CHEON, AND J. J. GU.
2002. Preliminary results of excavations at the
Neolithic Daechonri site, Okcheon, Journal of the
Korean Neolihic Society 3: 55–77.
LEE, S.K. & G.A. LEE. 2001. Current Research on
Agricultural Sites in Korea. Paper presented at the
2nd Meeting of the Joint Research Task
Committee on International Exchange Between
Korea and Japan. Osaka, Japan.
HO, P.T. 1977. The indigenous origins of Chinese
agriculture, in C.A. Reed (ed.), The Origins of
Agriculture: 413–484. Chicago: Mouton.
LEE, Y.J. & T.S. PARK. 1997. New materials on
prehistoric rice cultivation in Korea from the
Kawaji site, Kyongki Province. Paper presented to
the 2nd ICAA, Nanchang, China, 1997.
HYMOWITZ, T. & R.J. SINGH. 1987. Taxonomy and
speciation, in J.R. Wilcox (ed.), Soybeans:
Improvement, production, and uses: 23–48.
Madison: American Society of Agronomy.
MATSUTANI, A. 1983. Egoma-Shiso, in S. Katoh, T.
Kobayashi & T. Fujimoto (eds.), Jomon bunka no
kenkyu, vo. 2. Seigyo (Research on Jomon culture, Vol.
2. Occupations): 50–62. Yuzankaku, Tokyo.
IM, H.J. 1990. Archaeological research on the Kimpo
Peninsula, Gyeonggi province, Annual Journal of
Seoul National University Museum 2: 7–13.
NELSON, S. 1991. Evidence for early agriculture in
JEONG, J.W. 1997. Main Neolithic sites in Korea, in
National history compilation Committee (ed.),
Korean history: The Palaeolithic and Neolithic
periods (Hanguksa 2: Guseokgi-Sidae, Sinseokgi-
Sidae): 360–424. Seoul.
Korea, Indo-Pacific Prehistory Assoc. Bull. 10: 102–
8.
SAMPLE, T. 1974. Tongsamdong: A contribution to
Korea Neolithic culture history, Arctic Anthropology
11: 1–125
KASAHARA, Y. 1983. Torihama Kaizuka (dai-6-ji
hakkutsu) no shokubutsu shushi no kenshutsu to
dotei ni tsuite [The search for and identification of
plant seeds from the Torihama Shell Mound, 6th
excavation] I. Okamoto (ed.), Torihama Kaizuka
[The Torihama Shell Mound] (ed.) I. Fukui: Fukui-
Ken Kyoiku Iinkai.
SHIN, S-J. 1997. Neolithic pottery. In National History
Compilation Committee (ed.), Korean History 2:
Paleolithic and Neolithic Culture: 437–465. Seol:
Tamgudang Publisher.
SMITH, B.D. 1998. The emergence of agriculture. New
York: Scientific American Library.
KIM, C.C., M.Y. YOON & J.H. AHN. 2001. Absolute
Dates at the Sejukri site, in Dongguk University
Institution of Cultural Properties (ed.), The
Neolithic Subsistence Patterns and Environment in
Korea: 197–206. Kyongju: Dongguk University at
Kyongju.
STUIVER, M. & P.J. REIMER. 1993. Extended 14C
database and revised CALIB radiocarbon
calibration program. Radiocarbon 35: 215–30.
YAMAGUCHI. 1992. Wild and weed azuki beans in
Japan. Economic Botany 46: 384–94.
YOSHIZAKI, M. 1992. Ancient cultigen distribution.
KIM, J.H. & Y.J. LEE. 1997. Rice agriculture in Korea
through plant-opal analysis of Neolithic pottery.
Paper presented to the 2nd ICAA, Nanchang,
China, 1997.
Kokogaku Janaru 355: 2–13.
ZHAO, Z., ET AL. 1998. Distinguishing rice (Oryza
sativa Poaceae) from wild Oryza species through
phytolith analysis, II: finalized method,. Economic
Botany 52: 134–45.
KIM, J.M. 1980. Environmental changes and origin of
agriculture in Korea, Korean Journal of Ecology
3:40–51.
95