SUPPLEMENT TO THE "THREE CRITERIA" FOR EVALUATING THE ORIGINAL AREA OF CHINESE CULTIVATED RICE -- FURTHER DELIBERATION OF "CHINESE CULTIVATED RICE ORIGINATING ON THE MID-YANGTZE RIVER"     

At the Conference session Chinese Cultivated Rice Culture, Guanmiaoshan, Zhijiang at Hubei, China, in October, 1994, I submitted a paper "Re-examining Chinese cultivated rice origin, proposing a mid-Yangtze River origin" and outlining three criteria for origin, as follows:

(1). Neolithic geographic and climatic conditions must be suitable for cultivating rice, with wide wild rice distribution.

(2). Regional archaeological sites must have wide chronological and cultural continuity, especially Late Palaeolithic and Early Neolithic.

(3). Ancient rice remains include the earliest dated in China, plus later and more recent ones.

Thus, I reviewed the mid-Yangtze area for: a) Middle Neolithic ecology; b) Palaeolithic and Early Neolithic site distribution; and c) ancient rice finds; concluding:

1. Middle Neolithic mid-Yangtze area and climate were suited to rice cultivation, with wild rice widely distributed.

2. Palaeolithic and Early/Middle/Late Neolithic cultures chronologically evolved.

3. Regional rice remains include the earliest dated and common later-dated ones.

I conclude the mid-Yangtze Valley is the origin of Chinese cultivated rice, especially W Hubei and NW Hunan. This article in Journal of History (Shixue Huikan)(7), Nov. 1995, was published by Taiwan Zhonghua Xueshuyuan Zhonghua Shixue Xiehui, Zhongguo Wenhua Daxue Shixue Yanjiusuo and Department of Historiography; and Chinese Agricultural History (Zhongguo Nongshi), 1996(3). Later, I wrote "Macro-perspectives of Chinese ancient cultivated rice culture (Zhongguo Shiqian Daozuo Wenhua Di Hongguan Toushi)" in Agricultural Archaeology (Nongye Kaogu) 1995(1). There I repeated the "Three Criteria", with the mid-Yangtze Valley origin of Chinese cultivated rice.

Several years resulted in many new cultivated rice finds; e.g.s, phytoliths in Xianrendong and Diaotonghuan sites, Wannian, Jiangxi Province (1); phytoliths, two gray-yellow rice grains and partial phytolith in Yuchanyan site, Daoxian, Hunan Province (2); thousands of Pengtoushan culture rice grains in Bashidang site, Lixian, Hunan (3); world's earliest-dated paddies in Chengtoushan site, Lixian, Hunan (4); etc. As more mid-Yangtze River old cities add new but unclear material for finding rice origin, I insist "this area is the rice origin", basing my proof on "three criteria", archaeology and personal opinion on new rice finds.

I. Repercussions and supplement about " three criteria" in academic circles

　

Reasons for "three criteria". Undoubtedly, rice origin is ecotonal and human related, with rice-planting technology producing much archaeological data. I wish to use the "three criteria" to delimit rice origin and as criterion to measure the level and growth rate of ancient rice culture in different areas. Many researchers accept the "three criteria", some giving supplements. Wang Haiming's "Research and archaeological excavation of Chinese rice origin" says "upholding the 'three criteria' is not dogma to be applied or copied mechanically. We must note: [1] the main body of rice activity is cultural adaptability, allocating and altering nature; i.e., where people domesticated wild rice; and [2] effects of overpopulation, environment, etc., on cultivation" (5). Sun Shengru thought the "three criteria" were a great advance after rice origin research entered the third stage (6). Xiang Anqiang, founder of the "mid-Yangtze River" theory, agreed with the "three criteria", supplementing them with "[1]the earliest rice culture remains in most concentrated sites". Like mid-Yangtze Pengtoushan culture, all sites commonly have rice remains; but Yellow and Huai River Peiligang culture depended on millet, with rice only in Jiahu site, hence irregular. Accidental finds relate to excavation, with 'statistics and carbonized rice dating in similar ecotones allowing some conclusions' (Lin Huadong, Chinese rice origin and spread to Japan, Agricultural Archaeology (Nongye Kaogu) 1992:1). 'The many sites with their earliest rice remains reflect more than archaeology' (Yan Wenming, New Chinese Ancient Rice Finds, Jianghan Kaogu 1990:3).[2]They must link N-S Late Palaeolithic & Early Neolithic cultural exchange and spread. This middle area was critical to ancient cultural growth, agricultural origin and spread [3] and would be a 'millet-rice area' prehistorically (esp. Early Neolithic)"(7).

About Xiang Anqiang's three supplements to the criteria, I think his first supplement resembles the third criterion and his third supplement is unnecessary". Rice origin could not have been in the "millet-rice" area because both plants have different physiology and climatic and environmental needs, limiting production and spread. Xiang Anqiang's second supplement is needed but his description is too theoretical and vague in saying "mid-China's mid-Yangtze is the most important N-S cultural link. Its old petroglyphs (Liangshan, upper Han River, Xianan S of Shaanxi, Dongting Lake, Hubei) show different Palaeolithic N-S traits, exchange and influence. Lijiacun culture, upper Han River, not only had important N-S Early Neolithic links, but show Zhongyuan culture related directly to Yellow and Yangtze Rivers. Mid-Yangtze Pengtoushan and Chengbeixi cultures also have many common traits compared to Zhongyuan Cishan and Peiligang cultures, showing ancient cultural unity, regular thought and creativity and frequent N-S contraction and mutual-influence. For all ancient rice culture, mid-Yangtze sites are more numerous and concentrated, with surrounding sites few and scattered. This is intentional because it shows the effects of mid-Yangtze rice origin and spread. Its independent growth affected surrounding cultures. Pengtoushan culture Bashidang site, Lixian, Hunan, exposed the earliest (7000-8000 years) ditches, plus thousands of rice grains (village ditches in this period are also in Liyang Pingyuan). Chengtoushan is now China's oldest city (6000 year-old early Daxi culture), with 6500 year-old large ditches and paddies under its walls much bigger than Banpo site. Many precious 6-7,000 year-old artifacts show well-developed mid-Yangtze cultural rice spread; e.g.s, elegant wood oars, 3 m wood scull, etc."(8).

Although Xiang Anqiang's opinion is absolutely right I deleted some content and summarized his supplement 2 as criterion 4: "Ancient cities here are the oldest, largest and most numerous, with the highest level of ancient economy."

Criterion 4 divides rice origin directly and clearly, the city a sign of civilization and group activity. Unlike a modern city, it marks high agricultural, political, economic and cultural growth, "independent enough to influence its surroundings" (Xiang Anqiang), especially where surplus labor and food were needed. Food storage shows big harvests or expanded area. If this area was not the origin, there could be no rice-dependent cities. But cities appear first in groups of heavy population. Scientists think agricultural origin resulted from natural and cultural links, especially under higher millet and rice harvests. Hunter-gatherers were forced to depend on agriculture in certain areas due to population pressure or environmental change, and cities reflect specialization and population density. 9000-4000 year-old mid-Yangtze population growth was estimated in 1200 Neolithic sites, with an early growth rate of 0.12%, like the Neolithic world. 0.21% Daxi culture growth shows significant agricultural improvement over later lower growth rate, the peak being the first 2000 years and spread from Pengtoushan to Daxi culture (9). This is why I added criterion 4, the "three criteria" being fundamental, the fourth a significant mark for judging rice origin.

Daxi and Qujialing culture cities exempify the central role of the mid-Yangtze River.

Mid-Yangtze cities:

1. Chengtoushan S of Xujiagang and NW of Liyang at the Pingyuan-Li River junction is at 111°40' E Long. & 29°42' N Lat. Its 314-324 m diameter 5-6 m high walls enclosing 80,000 sq.m is surrounded by a 35-50 m wide, 4 m deep river with access in all 4 directions. City protection came from a 2.5 m ditch, its soil used to build its oldest wall level 12 of 2 m, with a 1.5 high gap from base to ditch. Level 11 accumulated from the outer slope covers both ditch banks. Level 10 came from inner slope city cultural debris. Potsherds and C14 data show level 10 is Daxi phase 2; its lower layer an earlier 2^nd culture proving its upper layer was >6000 year-old Daxi phase 1 and the oldest wall. As Chengtoushan limits stabilized then, it should be the oldest city (10).

2. Majiayuan, Jingmenshi, Hubei Province, is in mid-Yangtze Wulizhen, NW of Jianghan Plain between Jing Plain and Mountain, 28 km S of Jinancheng (Chu's old capital), Jiangling and 4 km W of Jingxiang ancient rice (Route 207). The Donggang near the W city wall flows into the Han River and passed N to S by Baohe and Changhu. On a 2-3 m elevation enclosed by wild paddies, its 640 x 300-400 m N-S ladder-shape encloses 340,000 sq.m. The E section of its S wall is 14-30 m thick (av. 20), its 5-6 m high S wall is 440 m long (33 m bottom, 8 m top). The N wall is 1.3 m high, 250m long (30 m bottom); 3 m high E wall is 640 m long (30 m bottom); and 4-6 m high W wall is 740 m long (35 m at bottom, 8 m top). The protected inner slope is 5 m wide; the outer one steep and straight. Overall, the city is in good condition, except 30-50 m parts S and N. A 4-6 m wide natural and artificial conduit connects to the river, winding inside from NW to SE gates. A gate is in E, N & S walls, plus river access in both W and E walls. Majiayuang was an important Qujialing culture city (11).

3. Yinxiang near Jingzhou city, Hubei Province at 112°011'11'' E Long. & 30°30'58'' N Lat., is 41-42 m AMSL with nearby land at 38 m. A rounded square, its E, W and S walls are in good condition with protruding SE corner, but the N wall was destroyed by lakewater. The remaining site is 580 m E-W and 350 m N-S, enclosing 200,000 sq.m. Walls are 10-25 m wide; the widest in E wall bottom at 46 m. Total length is about 900 m, 1-2 m higher than inner ground and 5-6 m higher than 30-40 m wide outer ditches. E and W walls are wide and high; the S wall narrow and short. The highest point 44.4m on the E wall; the lowest 37.9m 6.5 m away. Yinxiang built in Early Qujialing culture is a very large Daxi culture site of academic value as an important mid-Yangtze city (12).

4. Zoumaling by Shishoushi City is at the juncture of Tunzishan Village, Huajiadang District, and Zoumaling Village, Jiaoshanhe District, Hubei Province. Its 370 E-W x 300 m N-S irregular elliptical area encloses 78,000 sq.m. Its wall is 4-5m high and 20-27 m wide, with several gaps that may be doors. Round soil marks beside some gaps may represent door bases of additional defensive buildings. Soil from a 25-30 m wide external ditch was used for city building. Zoumaling is Early Qujialing period, used in Early and Middle period and abandoned in Late period (13).

5. Jiming is on a narrow small plain in Longchuanzui Village, Shuanglou, Shizikou District, Hubei Province. It is NE of tiny mild slopes 2 km S of Weicangshui River & 30 km NE of Gonganxian City at 111°59'03.8'' E Long. & 29°55'41.6'' N Lat. Its 500 NE-SW x 400 m E-W irregular elliptical area encloses 150,000 sq.m, with rounded SE and SW corners. Its 1140 m wall is 15 m wide at the top, 30 m at bottom & 2-3 m above ground, esp. the NW side which is 1 m higher. Except the vague E part, its surrounding river is 1300 x 20-30 m and 1-2 m deep, with no known gates. Excavators think the city was built in Qujialing period (14).

6. Jijiao near Lixian City, Hunan, is a 400 E-W x 500 m N-S rounded rectangle enclosing 200,000 sq.m, with 2-4 m high remaining wall with access to river. It is Qujialing culture (15).

7. Shijiahe in Shihezhen Village, Tianmen City, Hubei Province, is a 1200 N-S x 1100 m E-W rounded rectangle enclosing 1,200,000 sq.m. Its 2000 m E, W & S remaining wall is 8-10 m wide at top, >50 at bottom and 6 m at its highest. An enclosing 4800 x 80-100 m ditch built in a natural conduit was 6 m below the top of the wall. 30 Qujialing and Shijiahe culture sites are in the adjacent 8 sq.km; most commonly Early and Middle Shijiahe periods. It is no earlier than Middle Qujialing culture, prospering in Early and Middle Shijiahe periods (16).

In Taihu area near the mountains beside the lower Yangtze Plain, the largest Liangzhu cultural grouping and its "capital" is at Liangzhu, Anxi, Changming & Pingyao cities, W Yuhang, Zhejiang Province. >50 sites concentrated in a 5 km N-S and 10 km E-W range have areas: 29 @ <10,000 sq.m, 17 @ 10,000-50,000, 5 @ 50,000-100,000, one @ 150,000 and another @ 300,000. Some believe they represent "a city without protective walls", an intentional phenomenon tied to local rice cultural growth (17). Why were large cities like Chengtoushan on mid-Yangtze River in 1^st Daxi culture? Unlike many contemporaneous Qujialing culture sites on mid-Yangtze, why are there no city walls in lower Yangtze Liangzhu culture? It relates to regional undeveloped agriculture depending on rice planting and management because it affected both unit and total production. Planting and management directly relate to the length of planting history. The longer the history the more developed the culture, the easier rice planting and management improved. As cities reflect the growth of security, economy and culture, more cities may yet be found on the mid-Yangtze River.

In conclusion, seven mid-Yangtze cities are the oldest, largest and most numerous, reflecting the highest growth level of ancient rice society.

Cities came from agriculturally-based population growth, as seen in Chengtoushan, the world's oldest, an expansion of 8000 year-old mid-Yangtze Pengtoushan culture's Bashidang site, before city walls began.

Most artifacts in Bashidang on lower Li River NE of Liyang Plain, NW Hunan (Xiangxibei) are 7000-8000 year-old Early Neolithic Pengtoushan culture. Hunan Archaeological Research Institute (Wenwu Kaogu Yanjiusuo) excavated 1200 sq.m over six seasons in 1993-7. Bashidang has early (>30,000 sq.m), middle (most developed with wall & ditch) & late (30,000 sq.m) periods in a maximum N-S distance of 200m and E-W of 160m. 1-2 m high E, W & S walls (5m wide at top; 2m wide at bottom) enclose the living area. The N wall and 2m deep ditch (4m wide at top; 2m wide at bottom) connect to the river. A W wall door has river cobble stairs. Semi-subterranean and surface structures, fences and platforms are higher in the NW and NE, surrounded by hundreds of postholes. Bashidang also has many plant seeds, animal and poultry bones, bone & bamboo tools, cotton textiles, etc. (18). The first recorded city of Chengtoushan resulted from Pengtoushan through Daxi cultural growth, while Bashidang has no wall. Likewise, the sites of Majiayuan,Yinxiang, Zoumaling, Jiming, Jijiao and Shijiahe also grew from Daxi through Qujialing cultures.

Bashidang also had 15,000 rice grains, then the world's oldest, their number exceeding all other Chinese sites. In perfect form, they remain fresh, some with 1cm long awn. A Chinese Agriculture University (Zhongguo Nongye Daxue) rice specialist says it is both diverse and complex, the L/W ratio of the biggest 3x its smallest. It resembles indica and japonica, but their totally different phytoliths suggested it be called "Bashidang ancient rice (19).

Mid-Yangtze 8000 year-old Pengtoushan culture relied on rice, but Chengbeixi site, Yidu, Hubei has pre-Pengtoushan rice (20). The abundant "Bashidang ancient rice" is the oldest and represents incipient and evolving rice culture, while the oldest rice paddy fields are in Chengtoushan. Data is listed as follows (21):

In testing its E wall in winter, 1996, sticky gray-green paddy clay was below city wall 1 and culture level and on the surface. Cracks show this clay dried, while its rice stalks and root hair are identical to their modern counterpart. Carbonized rice and phytoliths were in the upper, middle and lower levels, while 3 rice leaf phytoliths were in the Qujialing culture wall soil. Cultural levels had 13, 5, 5, 2 & 0 leaf phytoliths and 47, 39, 46, 20 & 55 rice phytoliths, while grassy soil levels had 5, 21, 6 & 4 leaf phytoliths and 215, 234, 223 & 207 rice phytoliths, respectively. The field soil had 11 leaf and 180 rice phytoliths, while the bottom level had 2 leaf and 18 rice phytoliths. >95% leaf phytoliths are japonica, the rest a few indica and indica-japonica.

In winter, 1997, 3 NW/SE slopes occurred in the E wall. From W to E, slope 1 falls to the lower original field, its excavated length 19.5 m, with actual length exceeding excavation. 4.6-5x13m of slope 2 was excavated, its N part covered by the inside slopes of first and second period city walls. To save the wall it and its top weren't excavated, so actual length is unknown. Its S part was cut or destroyed by Daxi culture ditches. Slope 3 is 2.5m from slope 2 with only 5m excavated. It was covered by city walls 1 & 2, while ditches destroyed its N part.

These three slopes form two rectangles. These two were excavated from the lower original field, while preserving the original slopes W of the excavation. Where fields were as high as original slopes, people added soil to heighten them, while slope 1 rose naturally without added soil. The added section of slope 2 didn't connect to the original but further E, the reason unclear. As slopes have two periods, paddies do also, although it is hard to differentiate field section colour and quality, but Hongon University accelerator-dated two charcoal specimens in the top field at 4320-4055 & 4230-3985 B.C. (corrected with 95% probability). They used thermoluminescence to date T1030 and T1080 in the lower field at 6629±896 years, showing Chengtoushan paddies are the world's oldest.

In winter, 1998, at the SE corner of newly excavated T3131, slope 1 turned E. As its distance to the E side of the excavation is several dozen cm, its NW was not found. As slope 1 was 30m long and its N length unknown due to excavation, field 1 area at 1E is 132 sq.m (2 fen), much larger than the Majiabang culture field in Caoxieshan site, Wuxian, Jiangsu Province. Primitive irrigation includes a cistern and conduit in the W original soil level higher than fields. Three cisterns are: 1^st in the middle of T1028 (1.2m diameter, 1.3m deep, round-bottom); half of 2^nd beyond excavation (1.2m diameter, unexcavated bottom); quarter of 3^rd in NE corner of T3028 (1.5m diameter, thin bottom mud). In 1992, shallow conduits occurred in original soil in T3028 and some sections to the S, the border of occupation. The 1^st paddy was found in winter, 1997, in T1028, with two SW>NE conduits traceable to T3028's W side. Another conduit was in T3028 fill. As conduits connect with paddies, they are accessories, so conduit potsherds identify paddies as 6500-6300 year-old Tangjiagang culture, results like that of Hong Kong University.

The tie between "Bashidang rice" and "Chengtoushan paddies" is like a tree and forest, with similar origin, growth, cause and result. Chengtoushan's paddy irrigation reflects ancient mid-Yangtze rice cultivation and management, but more downriver and more developed than Caoxieshan's later irrigation. Caoxieshan's 20 variable size elliptical or rounded rectangular paddies are in two N-S connected lines, their mean area 3-5 sq.m, with several ones only 1 sq.m and larger ones 9 sq.m. They are low-lying, surrounded by slopes and connected with conduits and wells in the E and N. Cisterns are also at the end of conduits (22).

Caoxieshan's paddies are small and irregular, with ancient farmers expanding from lowland. They could not supply sufficient food (23), but Chengtoushan's rice irrigation and management were advanced, providing enough rice from paddies. More labor to build the city would have meant abandoning agriculture, which was impossible. In short, we think Chengtoushan is China's oldest city so far; its paddies not only the world's oldest, but show mid-Yangtze rice origin.

From mid-Sept. to mid-Nov., 1995, the Archaeology Faculty of Peking University, Jiangxi Archaeology Institute (Kaogu Yanjiusuo) and the American Andover Archaeology Foundation excavated Xianrendong and Diaotonghuan sites. Cultural Relic Newspaper (Zhongguo Wenwu Bao) published Xianrendong and Diaotonghuan Excavation on its front page in January 28, 1996, its subtitle saying "one of China's oldest pottery sites, transitional from Late Palaeolithic to Early Neolithic, with importance on rice origin in S China (Huanan)". It also said upper Early Neolithic layers in these two sites are 9000-14000 years-old; lower Late Palaeolithic or Mesolithic layers are 15000-20000 years-old, with clearest evidence in the transitional Palaeolithic level. Sporo-pollen analysis shows grass rose in upper layer, with pollen grains enlarging like rice. Upper layer phytoliths are fan-shaped like rice, all providing important clues on rice origin.

Chinese Cultural Relic Newspaper published New Finds in Yuchanyan about Rice Origin in March 3, 1996, saying in November, 1995, Hunan Archaeology Institute found phytoliths again (2 gray-yellow, one complete & one-fourth) in Yuchanyan. Comparing 1993 phytoliths, they redated rice to 10000 years ago. Levels show they postdate 1993 rice. All three 1993 layers have rice phytoliths, showing rice existed in Yuchanyan, while its electron microscopy show it is CWR, but intimately influenced by people, while 1995 rice was cultivated, with wild, japonica and indica traits showing the transition from wild to cultivated. For Yuchanyan, this dates 10000 years ago, using Sanjiaoyan, Hunan artifacts, with a C14 of 12060±120 years ago, plus similar cultural traits as Yuchanyan.

Xianrendong, Diaotonghuan and Yuchanyan support a mid-Yangtze rice origin, including Hubei, Hunan and Jiangxi Provinces; Hanzhong, Ankang and Shangluo regions in Shaanxi Province and Nanyang area in Henan Province. It includes Jianghan, Dongting and Poyang Plains; Nanyang Basin; Exi, Nanling, Xiangxi, Xiangdong,Gannan, Gandong, Ganxi and Shaannan mountain regions; Edong, Xiangzhong and Ebei hills (24), a total range between 106°50' & 118° E Long. and 24°20' & 30°20' N Lat.(25).

Xianrendong and Diaotonghuan near Lake Poyang, NE Jiangxi Province (117°13' E Long. & 28°44' N Lat.), Yuchanyan site near Xiao River, a Yangtze tributary, one of four rivers of Dongting Lake (111°30' E Long. & 25°30' N Lat.), are all part of the mid-Yangtze Basin.

Xianrendong and Diaotonghuan upper layers are 9000-14000 years old. The C14 Laboratory, Archaeology Institute, Chinese Academy of Social Sciences (Zhongguo Shehui Kexueyuan Kaogu Yanjiusuo), dated Xianrendong upper level clam shell provided by Jiangxi Museum to 10870±240 & 10560±240 (26); i.e., it and Diaotonghuan's farmers cultivated rice 10000 years ago. The lab also dated three animal bones in Yuchanyan to 7042-6039, 8327-7409 and 7911-6414 years, making Yuchanyan occupation 8327-6059 years ago (27). Yuchanyan's earliest 2cm thick potsherd is brown-black, poorly-fired and soft textured with coarse sand. Bifacial decoration resembles cord-wrapping but shows weaving with clear warp and weft. It predates 8000-9000 year-old Pengtoushan, with Sanjiaoyan, Hunan artifacts and C14 date of 12060±120 years, placing Yuchanyan at 10,000 years ago, as aforementioned.

Xianrendong, Diaotonghuan and Yuchanyan phytoliths provide strong evidence for a mid-Yangtze rice origin.

References:

1. Liu Shizhong, Important progress in Xianrendong and Diaotonghuan sites, Chinese Cultural Relics Newspaper, January 28, 1996:1.
2. Yuan Jiarong, Important material evidence of rice origin in Yuchangyan, Chinese Cultural Relics Newspaper, March 3, 1996:1.
3. Pei Anping, Many valuable remains in Bashidang site, Chinese Cultural Relics Newspaper, February 8,1998:1.
4. Hunan Archaeology Institute (Hunan Wenwu Kaogu Yanjiusuo), Chengtoushan 1997-8 excavation report, Cultural Relics 1999(6).
5. Wang Haiming, Chinese rice agricultural origin research and excavation, Agricultural Archaeology 1998(1):28.
6. Sun Shengru, Thinking about rice origin research, Agricultural Archaeology 1998(1):103.
7. Xiang Anqiang, mid-Yangtze River is the origin of Chinese rice culture, Agricultural Archaeology 1998(1):215,217.
8. see 7.
9. Wang Jiange, Population pressure and ancient agricultural growth, Agricultural Archaeology 1997(3):65.
10. see 4.
11. Hubei Jingzhou Museum, Investigating Qujialing culture site in Majiayuan, Cultural Relics 1997(7):49-53.
12. Jingzhou Museum, Excavation report about E city wall in Yinxiangcheng site, Archaeology 1997(5). Jingzhou Museum, Yinxiangcheng excavation reports in 1995, Archaeology 1998(1).
13. Jingzhou Museum, Excavation report of Zhoumaling Neolithic site in Shishou city, Archaeology 1998(4).
14. Jia Hanqing, Jimingcheng site investigation in Gongan, Cultural Relics, 1998(6).
15. Ren Shinan, Investigating Chinese ancient cities, Archaeology 1998(1).
16. Archaeology Faculty of Beijing University et al. Investigation report of Shijiahe site group, South China Archaeology, 1992 (5); Shijiahe Archaeology Group, Excavation of Shijiahe site group in 1987, Archaeology 1990(8).
17. Ren Shinan, Investigating Chinese ancient cities, Archaeology 1998(1).
18. see 3.
19. see 18.
20. Lin Chun et al. Chengbiexi, Pengtoushan culture and Chinese early rice agriculture, Agricultural Archaeology, 1993(3):118.
21. see 4.
22. First find of 6000 year-old paddy remains in Caoxieshan site, Chinese Cultural Relics Newspaper, June 18, 1995:1.
23. Udatsu Tetsuro, Tang Linghua et al. Investigation of paddy structure, Agricultural Archaeology 1998(1):143.
24. Wei Si, Second discussion on rice original area, Journal of History (17) (Shixue Huikan, Taiwan), November 1995.
25. see 24.
26. Archaeology Institute, Chinese Academy of Social Sciences (Zhongguo Shehui Kexueyuan Kaogu Yanjiusuo), C14 time data collection in Archaeology (1965-1991), Cultural Relics Publication 1991:127.
27. Archaeology Experimental Research Center, Chinese Academy of Social Sciences, Radiocarbon dating report (24), Archaeology 1997(7):38.

农业考古2000.1期