Genetic variability and association among seed yield and yield related traits in Ethiopian mustard (Brassica carinata A. Braun)

Sixty Ethiopian mustard genotypes were evaluated in randomized complete block design for genetic variability and character associations at Kulumsa Agricultural Research Center. Data were collected on 16 characters. The analysis of variance showed highly significant (p0.01) difference for all characters, indicating the existence of variability and the potential for selection and improvement within characters. High phenotypic coefficient of variation (PCV) was recorded for seed yield per plot, oil yield per plot, biomass per plot, seed yield per plant and number of pods per plant. The magnitudes of PCV and genotypic coefficient of variation (GCV) were high for seed yield per plot, oil yield per plot, biomass per plot and secondary branches per plant. Heritability estimates were high for days to maturity, days to flowering, 100-seed weight, plant height, biomass per plot, seed yield per plot and oil yield per plot. Similarly, heritability values for primary branches per plant and oil content were also high. High heritability was coupled with high genetic advance as percent of mean for biomass per plot, seed yield per plot and oil yield per plot. This indicates the effectiveness of selection to improve these three characters. Correlation and path analysis were used to determine character associations. Seed yield per plot showed positive and significant associations with oil yield per plot, biomass per plot, harvest index per plot, plant height, harvest index per plant and seed yield per plant. This indicated that simultaneous improvement of seed yield per plot and these characters is possible. With regard to seed yield per plant, positive associations were observed with harvest index per plant, 1000-seed weight and biomass per plant. Among the characters considered, plant height and 1000-seed weight were positively correlated with oil content. Days to flowering and days to maturity exhibited negative correlations with seed yield per plot. Correlation and path coefficient analysis of seed yield per plot and plant revealed that biomass and harvest index had strong and positive correlations and also exerted favorable direct effects on seed yield at genotypic and phenotypic levels. Selection for biomass and harvest index, therefore, can be very useful for seed yield improvement. Furthermore, plant height, 1000-seed weight and number of pods per plant could be used for indirect selection of seed yield. Phenotypic path coefficient analysis of oil content as dependent trait revealed that biomass per plot, plant height, harvest index per plot and days to maturity had positive direct effects on oil content, whereas seed yield per plot and secondary branches exerted negative direct effects on oil content. Thousand seed weight has exerted positive direct effect on oil content at genotypic and phenotypic levels and, plant height and 1000-seed weight had positive correlations with oil content at both levels. Thus plant height and 1000-seed weight should be considered as selection criteria for oil content improvement in Ethiopian mustard. The genetic divergence in Ethiopian mustard was assessed using Mahalanobis square D statistics and the genotypes were grouped into seven clusters. The largest and the smallest clusters comprised about 28.3 and 1.7%, respectively, of the studied genotypes. The average inter-cluster distances were significant for all clusters, suggesting the possibility of getting suitable genotypes for hybridization program among the genotypes of Ethiopian mustard.