More than 95% of the rainfed lowland rice is found in South and South East Asia, of which nearly 17 m ha is located in India. Rainfed lowland areas are characterized by frequent floods and droughts that delay planting. Varieties that are popular in these areas, such as Mahsuri and its derivatives, are not suitable for delayed planting as they are sensitive to extreme temperatures and often result in low yields. They lodge under high input management and are often susceptible to major diseases and pests prevalent in these areas. IRRI's shuttle breeding program has made available many breeding lines with improved traits. These lines have been evaluated in rainfed lowland area for direct utility or for their use in further breeding. The objective of the present study is to assess the genetic diversity at molecular level using a set of RAPD markers and to select the genetically diverse lines with desirable traits for use in further breeding. The usefulness of RAPD markers for studying diversity in rice has been noted by several workers (Yu and Nguyen, 1994; Virk et al., 1995).

In the present study, we used 47 RAPD primers to detect the polymorphism and genetic diversity among 21 rainfed lowland genotypes (Table 1). Extraction of genomic DNA using modified CTAB and use of the polymerase chain reaction followed standard procedures used for RAPD analysis. The DNA products from PCR amplification were separated on 1.5% agarose gels in 1 x TBE and visualised by ethidium bromide staining and photographed under UV light. For each primer, the PCR products were sequentially designated as a, b, c, d and so on, as described by Yu and Nguyen (1994). Data were scored on the presence or absence of amplification products. If the product was present in a genotype it was designated as '1' and if absent it was designated as '0'. Pair-wise comparisons of genotypes based on both unique and shared polymorphic products were used to generate similarity coefficients. (Jaccard, 1908). The similarity coefficients were then used to construct a dendrogram by Unweighted Pair Group Method with Arithmetical averages (UPGMA) using a computer program NTSYS verson 2-02.

Products ranging from 0.2 to 1.6 kilobase pairs were amplified by the 47 primers of arbitrary nucleotode sequence. For each primer evaluated, the number of DNA segments amplified for any given genotype ranged from 1 to 6. A total of 302 DNA fragments were amplified and 91% (275) of them showed some degree of polymorphism across the studied genotypes. Figure 1 shows the primer J-16 amplification products of 21 genotypes resolved by agarose gel electrophoresis. Associations among 21 rainfed lowland genotypes as revealed by UPGMA cluster analysis are presented in Figure 2. The 21 genotypes were classified into two major clusters. Cluster I contained all 19 genotypes which have been selected for one or more of the key traits, while, cluster II contained two commonly cultivated varieties, Mahsuri and its derivative Rajshree both of which lack similar traits for their successful cultivation in rainfed lowland areas. The varieties Rajshree and Mahsuri appeared to be distinct from others as they produced unique bands across 16 and 13 primers respectively (Table 1). These were followed by genotypes KMJ-1-17-1, CR333-10 and Jal-lahari with 11,11 and 10 unique bands, respectively. These three genotypes behave similarly as far as their sensitivity to drought is concerned. Further, it is interesting to note that the last genotypes that is added to each sub cluster is very different from others in terms of number of unique bands it produced (normally more). This indicates a positive relationship between the diversity of the genotype and number of unique bands it produced. The genotypes KMJ-1-17-1 and KMJ-1-19-1 although shared common parentage appeared in two distinct subclusters. The genotype KMJ-1-17-1 produced 11 unique bands as against 7 unique bands of KMJ-1-19-1. These two genotypes differ for their key traits (Table 1). This clearly indicated the efficiency of RAPD primers to differentiate the genotypes. The present study indicated that the genotype RAU83-82-4 is genetically very different from Mahsuri and Rajshree and could be used for improving the commonly grown

genotypes, as it possessed all the desirable key traits.

References

Jaccard, P., 1908. Nouvelles recherches sur la distribution florale. Bull. Soc. Vaudoise Sci. Nat. 44: 223-270.

Yu, L.X. and H.T. Nguyen, 1994. Genetic variation detected with RAPD markers among upland and lowland rice cultivars (Oryza sativa L.). Theor. Appl. Genet. 87: 668-672.

Virk, P.S., B.V. Ford-Lloyd, M.T. Jackson and H. Johnnewbury, 1995. Use of RAPD for the study of diversity within plant germplasm collections. Heredity 74:170-179.