Prediction of genetic variability and character contribution using path analysis in Curcuma longa L. germplasm

Seventeen genetic stocks of Curcuma longa L. were evaluated for nature and amount of genetic variability, associations and path analysis for six economic traits, namely plant height (cm), oil content in leaf (%), oil content in fresh rhizome (%), fresh rhizome weight (kg/plot), 1,8 cineole and p-cymene-8-ol in the essential oil of leaves. Phenotypic coefficient of variance (PCV) was higher than genotypic coefficient of variance (GCV) for all the traits in the study indicating high influence of environmental factors for these traits. Path coefficient study showed the largest direct contribution to rhizome yield was plant height, while other traits were negative; but their direct contribution via oil content in leaf and rhizome was remarkable. All traits expressed medium to high heritability in broad sense in percent (ĥ2 (BS)%), low to medium genetic advance and positive correlations. The correlation coefficients between traits expressed that rhizome oil content and 1,8 cineole were highly significant and positively correlated with plant height and 1,8 cineole at genotypic and phenotypic level, respectively. Overall, based on mean performance and other elite genetic/selection parameters, the accession ST-1 (UNK) (51.30 kg/plot) followed by ST-3 (9) (36.00 kg/plot) and ST-2 (4) (29.80 kg/plot) were the best for fresh rhizome yield; accession AMT-3 (1.50%) for essential oil content in leaf followed by ST-2 (4) (1.40%) and ST-3 (9) (1.30%); for essential oil content in rhizome, the accession TUR-1 UNK (0.73%) followed by AM-2 and TUR-13 (0.66%) were found very promising. Therefore, these accessions may be exploited for commercial production and/or exploitation.