GENETIC DIVERSITY OF AROMATIC RICE IN BANGLADESH

Bangladesh The nature and magnitude of diversity in 53 aromatic rice genotypes was evaluated in rain fed condition at BRAC Agricultural Research and Development Centre, Gazipur in 2012. The Euclidian method of divergence analysis indicated the presence of appreciable amount of genetic diversity in the material. These aromatic rice genotypes were grouped into six clusters where cluster I was the largest. Inter-cluster distances were larger than the intra-cluster distance revealed that there situated considerable genetic diversity among the genotypes. Based on positive value of vector I and II days to 50% flowering, seed length, and grain yield per hill had maximum contribution towards genetic divergence. Maximum yield contributing traits were accumulated in cluster V and as a result higher grain yield (42.0 g/hill) was obtained in this cluster. The genotypes of cluster V can be used in hybridization program to produce higher yielding breeding materials with all other clusters. The genotypes Cluster IV and cluster V were found most suitable for the respective characters and can be used as potential donor for future breeding programs.


INTRODUCTION
Bangladesh is well known for its native wealth of rice genetic resources and among these, large number of aromatic varieties cultivated in different agro-climatic regions of country (Nayak et al. 2004;Pandey et al. 2011).Rice landraces of Bangladesh possess tremendous genetic variation.Evaluation of this large genetic diversity is very important for their rational use in diverse needs.Genetic diversity is pre requisite for any crop improvement program, as it helps in the development of superior recombinants by providing necessary gene sources (Naik et al., 2006).Success of hybridization and subsequent selection of desirable segregants depends largely on the selection of parents with high genetic variability for different characters.The more diverse the parents, within overall limits of fitness, greater are the chances of obtaining higher amount of heterotic expression in F1s and broad spectrum of variability in segregating generations.Statistical analysis quantifies the genetic distance among the selected genotype and reflects the relative contribution of specific traits towards the total divergence.The use of Mahalanobis D 2 statistic for estimating genetic divergence has been emphasized by Shukla et al. (2006) and Sarawgi and Rita Binse (2007).The crosses between parents with suitable genetic divergence are generally the most responsive for yielding the most promising segregants.The present study was, therefore, undertaken to assess the extent of genetic diversity in 53 aromatic rice genotypes which will help to select prospective parents to develop transgressive segregants.

MATERIALS AND METHODS
The present investigation was under taken during rain-fed season (Aman) in 2012 at the BRAC Agricultural Research and Development Centre (BARDC), Gazipur, Bangladedsh.Geographically BARDC is located at 23.975 o N latitude and 90.399 o E longitude and 14 meters above the mean sea level.The genotypes were grown in 5.4 m x 0.6 m plots following randomized complete block design with three replications.The spacing was 20 cm between rows and 15 cm between plants.The data were recorded for yield and yield attributing characters and quality characters of aromatic rice used in this study.Observations were recorded on five randomly selected plants in each replication from the centre row.Different productive and quality characters viz.days to Days to 50% flowering, plant height (cm), tillers per hill, effective tillers per hill, panicle length (cm), filled grains per panicle, unfilled grains per panicle, total grain, sterility percentage, spikelet density, 1000 seed wt (gm), seed length (mm), seed breadth (mm), kernel length (mm), kernel breadth (mm) and grain yield per hill were recorded.The data were analyzed following Mahalanobis's (1936) generalized distance (D 2 ) extended by Rao (1952).Intra and inter-luster distances were calculated by the methods of Singh and Chaudhary (1985).All statistical analyses were carried out using GenStat 5.5, STAR, version 2.0.1 2014 and PB Tools, version 1.4.2014 computer software.

RESULT AND DISCUSSIONS Genetic Distance
All the inter-cluster distances were larger than the intra-cluster distance, indicating presence of wider diversity among genotypes of different groups (Table 1).The germplasm are traditional but they showed high variability between them which was revealed from the results of intra and inter-cluster distances.Basher et al. (2007) reported that inter-cluster distances were larger than intra-cluster distances in a multivariate analysis in rice.Intra-cluster distances were low for all clusters indicated homogeneous nature of the genotypes within the clusters.The results were supported by the findings of Iftekharuddaula et al. (2002) in rice.Intra cluster distance is highest in cluster II.
Regarding inter-cluster distance, cluster V showed maximum genetic distance (12.004) from cluster I followed by 11.711 of same cluster V from cluster IV (104).It is obvious that in all the cases cluster V produced the highest inter-cluster distances with other clusters suggesting wide diversity of the genotypes within cluster V with the genotypes of other clusters and the genotypes in these clusters could be used as parents in hybridization program for getting transgressive segregants (Saini and Kaiker, 1987).Lowest inter cluster D 2 values was recorded between cluster I and III (3.349).Highly divergent genotypes would produce a broad spectrum of variability in the subsequent generation enabling further selection and improvement, which would facilitate successful breeding of rice.

Clustering of genotypes
The genotypes fall into six clusters (table 02).The distribution pattern indicated that cluster I, the largest cluster, comprised sixteen genotypes followed by cluster III (11 genotypes), cluster IV (10 genotypes), cluster II (9 genotypes), cluster VI (5 genotypes) and cluster V (2 genotypes).The clustering pattern revealed that the genotypes clustered together indicated there was no association between eco-geographical distribution of genotypes and genetic divergence.

Cluster Mean
The highest cluster means for grain yield, 1000 seed wt and panicle length were obtained from cluster V (table 03).Highest total grain number, sterility percentage, spikelet density found in cluster IV, whereas the lowest grain yield and number of filled grain per panicle found in cluster III.Mean performance of different clusters for the characters studied revealed that dwarf stature, lower tillers per hill were in cluster VI whereas highest 1000 seed weight, panicle length, tillers number per hill and high yielding genotype were in V. Maximum desirable characters were accumulated in cluster V and as a result higher grain yield (42.0 g/hill) was obtained in this cluster.It was interesting that considering cluster distances, the cluster V showed comparative higher estimates of inter cluster-values with all other clusters.

Canonical Vector Analysis
The canonical vector analysis revealed that the vectors (vector I and II) for Days to 50% flowering, seed length, and grain yield per hill were positive (Table 4).Such result indicated that these three characters contributed maximum towards divergence.The greater divergence in the present materials due to these three characters will offer a good scope for improvement of yield through rational selection of parents for producing heterotic rice hybrids.

Correlation
50% flowering time had significant positive correlation with sterility percentage, 1000 seed wt (gm), seed length (mm), kernel length (mm), and grain yield per hill while negatively associated with plant height (cm).Plant height showed some sorts of correlation for tillers per hill and effective tillers per hill, while significant negative correlation for seed length, kernel length (table 05).Nofouzil et al. (2008) reported significant and positive correlation between grain yield and plant height.Significant and positive correlation among number of productive tillers per plant and grain yield was also noticed by Ali et al. (2008) in wheat.This result is also in consistent with the findings of Dogan (2009) and Gashaw et al. (2007).They reported significant positive correlation between grain yield and 1000-grain weight.
The correlations of tillers/hill were significantly positive with effective tillers per hill and GYP on the other hand effective tillers per hill and panicle length have significantly positive relation with GYP.There is no negative correlation for tillers/hill, effective tillers per hill and panicle length, where total grain showed only negative correlation with 1000 seed wt, seed length, Seed breadth, Kernel length and Kernel breadth.TSW has positively correlated with all characters.Although sterility percentage and kernel breadth showed no significant correlation with any characters but seed length and seed breadth showed significant positive correlation.

CONCLUSION AND RECOMMENDATION
Evaluated 53 aromatic rice genotypes of the present study are good sources of valuable genes.There is a high degree of diversity existed among the genotypes which needs to be utilized in future varietal improvement and/or development programs.Specifically, the genotypes of clusters V and IV may be selected as parents for hybridization programs to develop high yielding rice varieties with desirable other good characters.Future breeding goal should concentrate on selecting genotypes with high grain yield, long panicle length and more tillers per hill.

Table 1 .
Intra and inter cluster average distances of 53 aromatic rice genotypes

Table 2 .
Clustering pattern of 53 aromatic rice genotypes

Table 3 .
Cluster means for 53 aromatic rice genotypes

Table 4 .
Contributions of the characters towards divergence in 53 aromatic rice genotypes