SELECTION INDEX FOR SELECTING PROMISING DOUBLED HAPLOID LINES OF BLACK RICE

Black rice is useful as source of carbohydrate as well as to prevent some diseases. Breeding on this rice is not as common as that of white rice. The objective of this research was to obtain potential doubled haploid (DH) black rice lines that have good agronomic performance based on selection index. The research was conducted during November 2017 to April 2018 using a randomized complete block design with three replications. The planting materials consisting of 54 DH lines produced from anther culture and three cultivars as check cultivars, i.e., Aek Sibundong, Inpari 24 and Ciherang were used in this study. Selection index was developed based on broad-sense heritability of traits, followed by various statistical techniques, such as correlation analysis, path analysis, and principal component analysis (PCA). The results of this research indicated that panicle density (PD), weight of 1000 grains (GWE) and the number of filled grains (NFG) showed significant and positive correlation with grain yield (GY), while days to 50% flowering (FL), flag leaf length (FLL) and number of unfilled grain (NUG) showed significant but negative correlation with grain yield. Path analysis suggested GWE, NFG and FL contributed directly to GY. Besides grain yield, selection of the right model based on the best PC with highest eigen vector value of grain yield gave three traits, i.e., FL, GWE and NFG, that can be used as selection criteria and constructed selection index model. The constructed model of selection index (I) was I = (3 * 0.39 GY) + (-0.42 FL) + (0.23 GWE) + (0.44 NFG). Based on the positive index values, 28 black rice DH lines were selected to be used in further yield trials.