Chuanqing SUN, Xiangkun WANG and Zichao LI, Chinese Agricultural University, Beijing, PR CHINA

(Agricultural Archaeology 1998:21-29. Translated by Jiwu Wang; ed. by B. Gordon)


        RFLP (Restricted Fragment Length Polymorphism) has been an effective method of marking molecules and been widely used in analyzing finger prints, determining genes and selecting certain molecules. Some research has been reported about genetic relations between rice species and subspecies and other species by RFLP. Wang, etc.,39, 40 used this method to analyze wild and cultivated rice, the result showing no difference between RFLP and traditional methods. Nakazono et al.26 and Doi et al.11 used RFLP to analyze O. sativa, O. rufipogon & O. meridionalis and classified 7 species in the AA gene group of rice genera. Zhang et al.41 did similar research on 12 indica and 14 japonica species.

        The method of analyzing limited inscribed mtDNA enzyme has been successfully used in some higher plant research on wheat37, corn36, cocoa20 and potato20. Iwahashi et al.18 illustrated plates of rice mitochondrial gene group from the double clone of cox II, atp9, atpA, rrn26, rrn18, nad3, rsp12, cob, coxI, atp6 and 54 other clones. Ishii et al. (1993)15 compared rice mitochondrial DNA in samples from 4 indica, 2 japonica, 2 Java and 2 African cultivated rices in inscribed enzyme analysis and 5 inscribed enzymes and 4 probes in RFLP analysis. Results show Java rice has the same genes as japonica, with indica having much more remote affinities with African cultivated rice. Nakazono et al.26 also found a chloroplast DNA array in mtDNA.

        One research method on plant origin is studying the splitting and genetics of the chloroplast group. In rice research, Hirai et al. (1985)13 built the first plate collection of chloroplast DNA and measured chloroplast DNA length at 130 kb. Later, Hiratsuka et al. (1989)14 proved this length is 134.525 bp by analyzing chloroplast DNA array. Dally et al. (1990)9 used 247 samples typically representing genera in a limited inscribed cpDNA enzyme cp module to show indica and japonica differ, with most Chinese O. rufipogon having japonica cp modules. Ishii, etc. (1986, 1988)16,17 found the same cp modules in Asian cultivated and perennial common wild rice, proving cultivated rice rose from perennial common wild rice. They also show 3 types of cultivated rice cpDNA, and different cpDNA’s of indica (cpDNA III) and japonica (cpDNA I), with indica 0.1 kb shorter in length than japonica, the same difference as in wild rice. They concluded the cpDNA split happened before rice was cultivated and that indica and japonica cpDNA’s originated independently from respective Asian perennial common wild rice.

        Kanno et al., analyzing the exact position and accurate length of deficiencies in the P12 fragment of Pst I restricted enzyme of indica choroplast DNA, show deficiencies are 69bp long in ORF100. They designed two guiding materials based on the array of two ends of the deficiencies, processing them by PCR to test their existence. The deficiencies are therefore used symbolically to classify indica and japonica choroplast DNA’s because the cp DNA of most indica species has 69bp deficiencies, but japonica does not. Using this character, Chen (1993, 1994)7, 8 analyzed Yunnan dryland, wild and cultivated rice, finding cultivated rice species basically similar using the standard of 69bp deficiencies in ORF100 and Sato’s function and isodynamic enzyme.

        The gene group of advance life forms have many repeat arrays. Although we do not know clearly their origin and function, much research shows they appear as genetic symbols of a species or gene group that reveals evolutionary relationships between different species; e.g., the nuclear rice gene group has about 50% repeat DNA’s, many only existing in chromosomes of rice species. Wu et al. (1987)40 published the AA gene group’s peculiar repeat array, while Zhao (1989)42 and Aswildinnoor et al. (1991)6 reported those of EE, CC, FF and BBCC and EE, respectively.

        While almost all current researchers recognize Asian cultivated rice rising from common wild rice, they question if common wild rice splits into indica and japonica. Oka and Morishima’s early research (1961, 1962) suggested its possibility, but Shilu Zhou was the first (1948)1 to say in China is the Original Home of Rice that japonica came from Chinese japonica wild rice (Chaohu wild rice) and indica from India’s wild rice. Using isodynamic enzyme, Second (1982)33, 34 also found the indica and japonica split. Later, Morishima (1987)24 found most common wild rice is between typical indica and japonica using isodynamic enzyme and shape; i.e., Chinese common wild rice is closer to japonica, while that from Indonesia and the Philipines is closer to indica. Sano et al. (1990)32 found Dongxiang wild rice is closer to japonica in their comparative study of rDNA. The indica and japonica split from common wild rice was shown by Hongwei Cai et al. (1993)2 with isodynamic enzyme analysis; Chuanqing Sun, etc., by RAPD research and Wang, etc., and Nakano, etc., by RFLP research.

        The above shows change in indica and japonica split relating to geography, but did it happen before cultivation or result from cultivated rice genes? Is there any common wild rice without indica and japonica split? What is the relationship between degree of split and geographical factors? Meanwhile, further study remains on the cell nucleus, mitochondria, chloroplast, common character of three genetic systems in evolution and split, gene group repeat array and common wild rice indica and japonica split.

        This paper summarizes our research results on common wild rice nuclear DNA, mitochondrial DNA, chloroplast DNA, DNA repeat array and common wild rice indica and japonica split. It also studies the original centre of Asian cultivated rice and its evolutionary path.


1. Indica and japonica split in Common Wild Rice Nuclear DNA


1.1 Materials and Methods

        1.1.1 Materials. We selected 122 common wild rice samples from 10 Asian countries (China 39, India 27, Sri Lanka 7, Bangladesh 6, Thailand 17, Burma 12, Cambodia 5, Malaysia 6, Indonesia 2 & Philipines 1), plus 76 cultivated rice samples from 11 Asian countries. We chose a single DNA from every sample for RFLP analysis.

        1.1.2 We used Rogers and Bendich’s CTAB method (Cetyl triethylammonium bromide) for EFLP analysis, taking 3m g DNA from a sample and digesting it with Dra I. After electrophoresis we evaluated it using Saito’s method32 and used non-radioactive enhanced chemoluminescence as a RFLP probe to test sample material.39 The 48 probes we chose are from atlases of Saito’s RFLP (1991),31 Kurata21 and Kyushu University’s Faculty of Agriculture IR24 genes, a digest of Hind III. We defined every observed strip as value "1", no strips as value "0". We then used Nei’s method (1975)27 to count average gene diversity in test material, and calculated standard genetic distance from the difference of two test materials using Nei’s method (1987).28 Based on obtained genetic distance, we used UPGMA (Unweighted Pair-Group Method with Arithmetic mean)35 to analyze sample material and draw an illustrated tree.


1.2 Results and Analysis

        Analysis showed all test material comprised 4 groups: indica and common wild group inclining to indica (Group 1), japonica and common wild group inclining to japonica (Group 2), Chinese original common wild rice (Group 3) and south Asian and SE Asian original common wild rice (Group 4).

        Group 1 includes 74 wild and 38 indica species, plus 74 common wild species from India, Burma, Thailand, Sri Lanka, Bangladesh, Indonesia, Cambodia, Malaysia and China. 12 common wild rice species include Guangxi (3), Guangdong (1) and unknown (8). Chinese japonica species compared very distantly with those from India, Burma, Sri Lanka, Malaysia and other countries.

        Group 2 has 20 common wild (China 18, Cambodia & Burma 1 each) and 36 japonica species. China’s 18 include Guangxi (4), Yuanjiang (2) and unknown (7).

        Group 3 includes 8 common Chinese wild species: Jiangxi province’s Dongxiang (5 closely tied), Hunan’s Chaling (1) & Yuanjiang (2 closely tied), but Yuanjiang is quite separate from Dongxiang, while Hunan is neither too close nor too far. All 8 have distinct shape with 5-7mm long anthers; visible purple modules & leaf sheath; long red awns, black shells, red grain and all grow along the ground. As Xiangkun Wang4, Hanhua Pang, etc., say they depict original Chinese common wild species traits,5 we call this group the original Chinese common wild group. As they have distinct DNA finger prints; 10 probes can be used to test their traits.

        Group 4 has 19 common wild species from India (8), Sri Lanka (4), Burma (4), Philipines (1) and Malaysia (2). While atypical compared to Chinese, they are the most original south and SE Asian species we used (long anthers & purple modules), suggesting we call this group the south and SE Asian original common wild group. Of the 48 probes we used, 12 reflect traits of south and SE Asian common wild species in RFLP.

        From the above result, it can be found that both Chinese common wild rice and south and SE Asian common wild rice have either close or distant relations to cultivated rice; e.g., 3rd & 4th groups have a distance relative to cultivated rice. For the first time, this study differentiates common wild rice that has a distant relative to the cultivated one; i.e., original common wild rice is from another common wild rice species. At this point, our result differs from Wang et al.’s. Wang’s RFLP illustrative figure cannot divide sub-indica and sub-japonica species; and subspecies cannot be differentiated from common wild species either.

        Our results show Chinese and south and SE Asian common wild rice are two different species with independent origins and different ties to cultivated rice. Japonica and related species are all Chinese except one each from Burma and Cambodia. Of indica and related species, Chinese common wild species are more distantly tied than Indian and south Asian rice, but Guangxi and Guangdong common wild species are in this group.

        Thus, nuclear DNA differentiation shows Chinese common wild species divide into those inclined to indica and japonica; with south and SE Asian common wild species inclined to indica, but the possibility of japonica-inclined in the latter. Chinese geography shows different genetic splits in common wild species; e.g.s, those in Jiangxi province’s Dongxiang, Hunan province’s Chaling and Yunnan are neither indica nor japonica but independent because their shape is so unique they suggest original ancient species. Guangdong and Guangxi wild rice has more traits of either indica or japonica. The split and genetics of common wild rice nuclear DNA suggest China and south Asia (India is central) are two original centers of cultivated rice evolution; japonica starting from China, and indica from China and India.


2 Split between indica and japonica in Common Wild Rice DNA


2.1 Materials and Methods

        Except only 4 Chinese common wild rice samples, materials and DNA collecting and RFLP analysis are identical to 1.1. Our probes are ATPA (ATPass alpha subunit from peas), cox I (cytochrome oxidase, subunit I from rice), cox II (cytochrome oxidase, subunit II from rice), nad3 (NADH dehydrogenase subunit 3, from rice), atp A (ATPase alpha subunit from rice) and rrn 18 + 5 (18s + 5s ribosomal RNA from wheat). ATPA and rrn (18 + 5) were provided by Kobe University, Japan; other mtDNA probes from Tokyo University. Five limited inscribed enzymes; Dra I, Hind III, EcoR V, Pst I and Sal I, were chosen in the experiment.

        To verify if the multi-shaped nature of probed objects are from mitochondrial genes, we also chose Jin Nan Feng/DV85’s regrouped self-pollenating system’s (RI system) F7 group and IR24/K503 F2 group.


2.2 Analysis and Results

        After using UPGMA method to conduct grouping analysis of 118 common wild and 76 cultivated rice samples, we divided them into the following five groups, based on the multi-shaped nature of mitochondrial DNA:

        Group One: Common wild group that is indica or inclined to it, containing 121 samples (86 common wild & 35 cultivated), the former as 26 China, 16 India, 6 Thailand, 14 Malaysia, 6 Sri Lanka, 4 Bangladesh, 3 Cambodia, 9 Burma and 2 Indonesia. Chinese samples are 1 Dongxiang, 1 Hunan, 8 Guangxi, 4 Guangdong and 12 unknown. The traditional classification shows most of the 35 cultivated rice samples are indica, like Guanglu short #4, IR24 and Kasalath.

        Goup Two: Common wild group is japonica or inclined to it, with 7 wild and 40 cultivated (local Jiaodong Green, Red Beard japonica, Red Beard #1 & Balila; some improved varieties like Japanese Qing, Fall Light & Nonglin #8, and other breeds like Miyang #23). Wild samples are from China (3), India (2), Cambodia (1) and Philipines (1). In terms of inner-group variation this group is smaller than the first.

        Group Three: Jiangxi Dongxiang common wild group only includes two common wild rices (WA113 & WA114) with distinct traits using 4 inscribed enzyme probes and 8 multi-shaped pieces. Traits are stolon, long anther (6.5mm), black shell, red long beard, red rice and lower fruit-bearing rate while self-pollenating. All are original multi-annual common wild rice species.

        Group Four: India, Thailand, South & SE Asia and Yunnan common wild group of 22 samples: India (8), Thailand (3), Bangladesh (1), Cambodia (1), Indonesia (1) and Yuanjiang (4).

        Group Five: Two samples of Bangladesh common wild and cultivated species.

        Using 7 probes and 17 inscribed enzyme probes together to analyze 118 common wild mitochondria (mtDNA) by RFLP method, we found most are indica (86 samples) and a few japonica (7 samples). Of others, 1 sample is undetermined and 24 have no trace of indica-japonica split. In cultivated samples, the mtDNA ratio between indica and japonica is almost identical. Cultivated rices are from wild; and japonica mtDNA in common wild rice is uncommon. Where does japonica mtDNA come from? Three ways exist: from indica or japonica common wild rice or directly from original wild rice.


3 Indica-japonica Split in Common Wild Rice Green Leaf DNA


3.1 Material and Method

        3.1.1 Material. 245 samples include 93 cultivated and 151 common wild samples: the latter as 67 China, 27 India, 17 Thailand, 12 Burma, 6 Maylasia, 5 Cambodia, 2 Indonesia, 8 Sri Lanka, 6 Bangadesh and 1 Philipines.

        To verify if PCR products from the common DNA module truly reflect cpDNA multi-shaped character, we used 100 units of Nanfeng/DV85’s RI F7 and 100 units of IR24/K503F2 for PCR analysis.

        3.1.2 Method. We collected common DNA after the section 1.1 method according to Chen (1993)7 to combine introduced varieties (P1:5’-GGCCATCATTTTCTTTAG-3’ & P2:5’-AGTCCACTCAGCCATCTCTC-3’) using common DNA as module, with 25m l reaction liguid (10ng module DNA, 10m MTris-HC1 pH 9.0), 50mM KC1, 1.5mM MgC12, 0.1% Triton X-100, a pair of fuses (1m l for each), 1 unit Taq promega, plus a drop of mineral oil as cover material. This was placed in a Iwaki Thermal Sequencer (TSR-300) at 94°C for 1 min., 51°C for 1 min. and 72°C for 2 min. After 40 cycles, it was kept at 72°C for 10 mins. Later, we put 4m l PCR material into 1.4% liquid agarose containing 0.5m g/ml bromide amylum B and electrophoresced at 80V for 5 hours, then took pictures and observed results.


3.2 Results and Analysis

        As the cpDNA forms of Jinnanfeng/DV85’s RI system resemble Jinnanfeng in terms of their ORF100, without 69bp defects, they belong to japonica. But, the cpDNA in IR24/K-503F2 group has defects like IR24 and belongs to indica. Obviously, when using common DNA as module and placing PCR through P1 and P2 probes, we got different products; the difference really resulting from chloroplast gene group difference.

        Of 93 cultivated rice samples, 32 are defective (indica) and 61 are not (japonica). Traditional classification suggests indica, as they have its cpDNA traits; i.e., 69bp defects on ORF100. While having some typical indica like IR36, Shengli indica, Guichao 2, Shuanggui 36, Longgui, Guanglu 4, Luzhenzao, etc., others think they are japonica; e.g.s, Luhui 422, 02428, Fubao, Keqing 3, Han 9, Jigeng 44, Liaogeng 5, Qiuguang, Balila, etc., because their cpDNA has japonica traits; i.e., there are no cpDNA defects on ORF100. Others say they are indica like Miyang 23, but their cpDNA is japonica.

        As they belong to different indica and japonica, common wild species in different countries differ in their cpDNA defects on ORF100. More indica than japonica occurs in the cpDNA split in 67 Chinese common wild species (excluding 16 samples of unknown origin where more japonica than indica occurs). But indica and japonica cpDNA traits differ from common wild rice; e.g., of 10 Jiangxi Dongxiang common wild samples, 6 are indica & 4 japonica; while some earlier research said Dongxiang common wild rice is more likely japonica.12 While Hunan Chaling common wild rice was thought to be quite pure, 6 samples contain 1 indica and 5 japonica; japonica>indica. As Guangxi has most common wild species (7 indica & 15 japonica in 22 samples), japonica obviously dominates. Guangdong common wild species is also japonica dominant (6 of 9 samples). Of 3 Fujian common wild rices, 1 is indica and 2 are japonica, while all 4 Yunnan common wild samples are indica.

        Indica and japonica cpDNA split in India and other Asian countries also has its own trait. Of 27 Indian common wild samples, 11 are indica (40%), 16 japonica (60%); japonica>indica. Of 17 Thai samples, 93% (16) are indica. Of 12 Burma samples, 75% (9) are japonica. Of 6 Bangladesh samples, 5 are japonica. Samples from other Asian countries like Malaysia, Cambodia, Sri Lanka and Indonesia are mainly indica.

        We suggest indica and japonica orginated from two different sources from their indica and japonica split in common wild rice cpDNA.


4. Genetics and Differentiation in Chinese Common Wild Rice Repeat Array DNA’s

        We used BamH I enzyme to split 29 Chinese common wild rice DNA samples (Guangdong 6, Hainan 2, Guangxi 5, Jiangxi 3, Hunan 6, Fujian 1, Yunnan 2 and Taiwan 2) and 43 cultivated rice samples. We used repeat array (pOs139) cloned from indica’s narrow green leaves as probes to conduct Southern hybridization. In 24 japonica samples, hybridization strips of all except "Xiangnuomi" and "Qiuguang" are in the area with high molecular weight, with hybridization strip number few. While "Haoangu" has fewer hybridization strips in 19 indica samples, their number in 18 other samples exceed japonica samples (av. >10 strips). Of 29 Chinese common wild samples, 1 from Guangxi has fewer strips; others have more, showing similar hybridization result with indica samples. Therefore, Chinese common wild rice is closer to indica than japonica on repeat array. It also looks more like indica.


5. Genetics and Differentiation in Chinese Common Wild Rice Natural Group


        5.1. Indica:japonica split in Chloroplast DNA of Chinese Common Wild Rice Natural Group

        Following the method used in 3.1, we made a PCR analysis of Dongxiang group, Guilin group and Fusui group’s cpDNA to see if the cpDNA’s ORF100 has 67bp defects. Results show that of 32 Dongxiang common wild samples, 25 are indica (78%) and 7 are japonica (22%). Of 36 Fusui samples, 7 are indica (11%), 29 are japonica. Obviously, Fusui is more likely to be japonica and Dongxiang group’s cpDNA inclines to indica.

        5.2 Indica:japonica split in Chinese Common Wild Rice Core DNA

        We selected 28 samples respectively from Dongxiang, Guilin and Fusui groups, and used two kinds of inscribed enzymes, 14 core gene group probes, 14 probes/inscribed enzyme to analyze these three common wild groups’ DNA indica:japonica split. Results show the genetic multi-character of common wild groups mainly comes from inside (ca. 70% of all multi-character genes produced in groups and 30% produced among groups). Moreover, of 14 location points, 5 differentiate indica and japonica. In these three groups points are mainly japonica (>50%), with indica low. As total mean japonica frequency is 0.772 and indica is 0.008, groups tend to be japonica in terms of core gene groups.

        Of the above three groups, Dongxiang is well differentiated from Guilin, but their core DNA (esp. Dongxiang) have different chloroplast DNA split; i.e. the core gene group and group evolution of the chloroplast gene is highly independent.


6. From indica:japonica split in Common Wild Rice DNA to the Origin and Development of Asian Cultivated Rice


6.1 Multi-Shapes of the Indica:japonica split

        Whether individual or group level; core or gene gene group, single copy or repeat DNA array, or gene group analysis of common wild rice core, grain and chloroplast, all show high indica:japonica split in common wild rice. This split is both regional and multi-shaped: (1) on core DNA, south Asia and SE Asia common wild rice are indica or original; and there are some Chinese common wild rice species that are indica or original, but mainly japonica; (2) core cells, mitochondria and chloroplast have multi-shaped genetic systems; e.g., only 41% of core cells and mitochondria in Chinese common wild rice have similar character - indica or japonica. 59% appear different. 54% of core cells and chloroplast are alike, 46% differ, while 21% of mitochondria and chloroplast are alike. South Asian and SE Asian common wild rice also show multi-trait genetic systems. (3) comparing repeat array of a single or lower copy with a multi-copy, splits vary; e.g., Chinese common wild rice core DNA and single and lower copy probes show japonica, indica or original when hybridized. But when multi-copy repeat array probes are hydridized, they show more indica.


6.2 Original Centre of Cultivated Rice and its Route of Spread

        After observing the genetics and differentiation of common wild rice core DNA and mitochondrial DNA, we believe China and South Asia (India as centre) are two original centres of cultivated rice; with SE Asia a middle region. Japonica may originate in China, and indica from China and India because China and South Asia have respective unsplit original rice species. Common wild rice core gene appears as japonica mainly in China, while indica common wild rice occurs in China, South Asia and SE Asia.

        Chloroplast DNA evolution is very conservative, with indica and japonica types in common wild rice. Indica and japonica co-existed according to their core DNA, but their mitochondrial DNA also co-existed, although more indica than japonica occurred. In the indica:japonica evolutionary process, the following may have occurred: common wild rice first became indica-like and japonica-like; the former becoming indica and the latter becoming japonica. As China and South Asian common wild rice both have small groups without indica:japonica split in core and mitochondrial DNA, it is not impossible that cultivated rice derived from original common wild rice.

        Rice mitochondrial DNA evolution is very conservative. Common wild rice mtDNA is mainly indica (86 samples), less japonica (7 samples); but the indica:japonica ratio in cultivated rice mtDNA is nearly half and half. In the repeat array, common wild rice appears likely to be indica while cultivated rice is nearly equal. Where does japonica mtDNA in cultivated rice come from? What does the repeat array in japonica consist of? While they may arise via indica evolution, it is also possible they came directly from original wild rice. Therefore, in the process of indica:japonica evolution, indica common wild rice may evolve to japonica. In a word, indica and japonica evolve various ways.




1 Zhou Shilu. "China is the Original Place of Rice" Chinese Rice, 1984, 7(5):53-4.

2 Cao Hongwei, Wang Xiangkun & Pang Hanhua. "An Isoenzyme Study of indica and japonica Split in Chinese Common Wild Rice," Collection of Works of Agricultural Science, vol. 1, Beijing: Agriculture Publishing House, 106-110.

3 Sun Chunqing, Mao Long & Wang Zhenshan. "An Analysis of Multi-Shaped DNA (RAPD) of Gene Groups of Chinese Common Wild and Cultivated Rice" Chinese Rice Science, 1995, 9(1):1-6.

4 Wang Xiangkun. "A Preliminary Exploration of the indica and japonica Split and the Origin of Chinese Common Wild Rice," Chinese Rice Science, 1994, 8 (4):205-210.

5 Pang Zhonghua, Cai Hongwei & Wang Xiangkun. "A Classified Study of Chinese Common Wild Rice Shape," Journal of Crop Studies, 1995, 21 (2):17-24.

6 Asindinnaor H.R., J. Nelson, J.F. Dallos et al. "Cloning, Characterisation and Repetitive DNA from Genomes of Oryza Minutae and Oryza Australiemsis," Genome, 1991, 34:790-8.

7 Chen, W.B., I. Nakamura, Y.I. Sato et al. (1993) "Distribution of Deletion Type in cpDNA of Cultivated and Wild Rice." Japanese Journal of Genetics, 68:597-603.

8 Chen, W.B., I. Nakamura, Y.I. Sato et al. (1994) "Indica-japonica Differentiation in Chinese Rice Landraces," Euphytica 74:195-201.

9 Dally A.M. & G. Second. "Chloroplast DNA Diversity in Wild and Cultivated Species of Rice (Genus Oryza Section Oryza). Cladistic-Mutation and Genetic-Distance Analysis," Theory of Applied Genetics, 1990, 80:209-222.

10 Dhar M.S., M.M. Dabak, V.A. Gupta et al. "Organization and Properties of Repeated DNA Sequences in Rice Genome," Plant Science, 1988, 55:43-52.

11 Doi K., A. Yoshimura, M. Nakano et al. "Phylogenetic Study of a Genome Species of Genus Oryza using Nuclear RFLP," Rice Genetics Newsletter, 1995, 12:160-162.

12 Flavell R. "The Molecular Characterization and Organization of Plant Chromosomal DNA Sequences," Annual Review of Plant Physiology, 1980, 31:569-596.

13 Hirai A., T. Isshibashi, A. Morikami et al. "Rice Chloroplast DNA: Physical Map and Location of Genes for the Large Subunit of Ribulose 1,5 - Bisphosphate Carboxylase and the 32KD Photosystem II Reaction Center Protein," Theory of Applied Genetics, 1985, 70:117-122.

14 Hiratsuka J., H. Shimade, R Whittier et al. "The Complete Sequence of the Rice (Oryza sativa) Chloroplast Genome: Intermolecular Recombination between Distinct tRNA Genes Accounts for a Major Plastid DNA inversion during the Evolution of the Cereals," Mol. Gen. Genet. 1989, 217:185-194.

15 Ishii T., T. Terachi, N. Mori et al. "Comparative Study on Chloroplast, Mitochondrial and Nuclear Genome Differentiation in Two Cultivated Rice Specids," Oryza sativa and Oryza glaberrima, by RFLP Analysis," Theory of Applied Genetics, 1993, 86:88-96.

16 Ishii T., T. Terachi & K. Tsunewaki. "Restriction Endonuclease Analysis of Chloropiast DNA from Cultivated Rice Species, Oryza sativa & O. Glaberrima," Japanese Journal of Genetics, 1986, 61:537-541.

17 Ishii T., T. Terachi & K. Tsunewaki. "Restriction Endonuclease Analysis of Chloroplast DNA from a Genome Diploid Species of Rice," Japanese Journal of Genetics, 1988, 63:523-536.

18 Iwahashi M., M. Nakazono, A. Kanno et al. "Genetic and Physical Maps and a Clone Bank of Mitochondrial DNA from Rice," Theory of Applied Genetics. 1992, 84:275-279.

19 Kanno A., N. Watanabe, I. Nakamura et al. "Variation in Chloroplast DNA from Rice (Oryza sativa): Differences between Deletions Mediated by Short Direct-Repeat Squences within a Single Specy," Theory of Applied Genetics, 1993, 86:579-584.

20 Kemble R. J. & J. F. Shepard. "Crytopasmic DNA Variation in a Potato Protoclonal Population," Theory of Applied Genetics, 1984, 69:211-6.

21 Kurata N., Y. Nagamura, K. Yamamoto et al. "A 300 Kilobase Interval Genetic Map of Rice Including 833 Expressed Sequences," Nature Genetics, 1994, 8:365- 372.

22 Laurent V., A.M. Risterucci & C. Lanaud. "Chloroplast and Mitochondrial DNA Diversity in Theobroma Cacao," Theory of Applied Genetics, 1993, 87:81-8.

23 Luo H., B.V. Copenolle, M. Seguin et al. "Mitochondrial DNA Phylogenetic Relationships in Hevea Brasiliensis," Molecular Breeding, 1995, 1:51-63.

24 Morishima H. & L.U. Gadrinab (1987) "Does Asian Common Wild Rice Differentiate into indica and japonica Types?" in S. C. Hxieh, ed. Crop Exploration and Utilization of Genetics Resources (Taichung District Agricultural Improvement Station, Changhua, Taiwan, 1987, 11-20).

25 Nakano M., Yoshimura A. & N. Iwata. "Phylogenetic Study of Cultivated Rice and Its Wild Relatives by RFLP," Rice Genetics Newsletter, 1992, 9:132-134.

26 Nakazono M. & A. Hirai. "Identification of the Entire Set of Transferred Chloroplast DNA Sequences in the Mitochondrial Genome of Rice," Molecular Genetics, 1993, 236:341-346.

27 Nei M., Molecular Population Genetics and Evolution. Amsterdam: North Holland, 1975.

28 Nei M., Molecular Evolutionary Genetics. New York: Columbia University Press, 1987, 190-1.

29 Pollard D., C.A. Read, M.J. Downes et al. "Non-radioactive Nucleic Acid Detection by Enhanced Chemoluminescenec Using Probes Directly Labeled with Horseradish Peroxidase," Analytical Biochemistry, 1990, 185:84-89.

30 Rogers O.S. & A.J. Bendich. "Extraction of DNA from Plant Tissures," Plant Molecular Biology Manual, 1988, A6:1-10.

31 Saito A., M. Yano, N. Kishimoto, M. Nakagahra et al. "Linkage Map of Restriction Fragment Length Polymorphism Loci in Rice," Japan Journal of Breeding, 1991, 41:665-670.

32 Sano Y. & R. Sano. "Variation of the Intergenic Spacer Region of Ribosomal DNA in Cultivated and Wild Rice Species," Genome, 1990, 33:209-218.

33 Second G. "Evolutionary Relationships in the Sativa Group of Oryza Based on Isozyme Data," Genet. Sel. Evol. 1985, 17 (1):89-114.

34 Second G. "A New Insight into the Genome Differentiation in Oryza sativa L. through Isozymic Studies," Advances in Chromosome and Cell Genetics (New Delhi: Oxford & L IBH Publishing House (Sharma A.R & Sharma A., eds., 1985), p. 75.

35 Sokal R.R. & C.D. Michener. "A Statistical Method for Evaluating Systematic Relationship," Science Bulletin, University of Kansas, 1958, 28:1409-39.

36 Timothy D.H., C.S. Levings, D.R. Pring et al. "Organelle DNA Variation and Systematic Relationships in the Genus Zea, Teosinte". Proceedings of the National Academy of Science, U.S.A., 1979, 76:4220-4224.

37 Vedel F., F. Quetier, F. Dosba et al. "Study of Wheat Phylogeny by EcoR I Analysis of Chloroplast and Mitochondria," Plant Science Letter, 1978, 113:97-102.

38 Wang Z.Y. et al. "Polymorphism and Phylogenetic Relationships among Species in Genus Oryza as Determined by Analysis of Nuclear RFLPs," Theory of Applied Genetics, 1992, 83:565-581.

39 Wang Z.Y. & S.D. Tanksley. "Restriction Fragment Length Polymorphism in Oryza sativa L. ," Genome, 1989, 32:113-118.

40 Wu T.Y. & R. Wu. "New Rice Repetitive DNA Shows Sequence Homology to both 5S RNA and tRNA," Nucleic Acids Research, 1987, 15 (15):5913-5923.

41 Zhang Q.F., M.A. Saghai, T.Y. Lu et al. "Genetic Diversity and Differentiation of indica and japonica Rice Detected by RFLP Analysis," Theory of Applied Genetics, 1992, 83:495-9.

42 Zhao X., T. Wu, Y. Xie et al. "Genome Specific Repetitive Sequences in the Genus Oryza," Theory of Applied Genetics, 1989, 78:201-209.