M.H. RANI, M. FARUQUEE, M.S.R. KHANOM, and S.N. BEGUM
Thirty rice landraces were evaluated during the 2020 wet season for the estimation of the genetic variability of six grain physical properties, viz. grain length (GL), grain breadth (GB), milled grain length (MGL), milled grain breadth (MGB), milled grain length breadth ratio (MGL/MGB), and 1000-grain weight (TGW), at the Bangladesh Institute of Nuclear Agriculture Substation, Sunamganj, Bangladesh. The relative contribution of these traits to variability was estimated by using principal component analysis (PCA), and the landraces were clustered by using Mahalanobis distance (D2)statistics. The TGW and MGL/MGB ratio exhibited high estimates of the phenotypic coefficient of variation and genotypic coefficient of variation. The high broad-sense heritability and genetic advance of all the traits indicated that the environmental effect had a weak involvement in the expression of these traits. PCA revealed six principal components, among which two were significant and contributed up to 96.9% of the total variance cumulatively. GL, GB, MGL, and TGW contributed to PC1 to create the variation among the landraces, whereas MGL/MGB ratio, GL, and MGL contributed to PC2. The landraces were grouped into six clusters. Cluster analysis revealed that the maximum and minimum intracluster distances were found in cluster III (235.11) and cluster VI (0.00), respectively. The longest intercluster distance was found between clusters IV and VI, and the shortest distance was found between clusters I and III. The maximum mean values for GL and TGW were observed in cluster VI. The mean values for GB and MGB were highest in cluster V, whereas the MGL/MGB ratio and MGL were highest in cluster II. ‘Madhumala’/‘Sada Madhumala’ and ‘Pankhuraj’ could be used in hybridization programs to exploit maximum heterosis for rice grain size and shape and for the direct selection of superior quality traits because these traits are less affected by the environment than other traits.
Date published: March 2022
Keywords: Rice landrace, heritability, principal component analysis, clustering
DOI: http://doi.org/10.54910/sabrao2022.54.1.1