Phenotypic relationships among growth, yield and black leaf streak disease responses of Musa genotypes
No hay miniatura disponible
Fecha
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Informa UK Limited
Resumen
Descripción
This study was carried out to assess the phenotypic correlations among phenological traits, yield, and black sigatoka disease (BSD) response of 14 Musa genotypes grown for three cropping cycles in three high-altitude sites that varied in temperature and annual rainfall. Pre-flowering plant growth had stronger relationship with bunch weight only in low yielding (LY) genotypes. In contrast, foliage traits and BSD resistance status of plants at flowering were more associated with bunch weight in high yielding (HY) genotypes. For example, plant size at flowering was positively associated with bunch weight in LY genotypes in all the MET sites (with r ranging between 0.58** and 0.89**), but for HY genotypes, the number of green leaves at harvest was significantly correlated with bunch weight in all sites (with r ranging between 0.75** and 0.88**). Also, bunch yield traits had higher inter-crop cycle r-values (ranging between 0.60** and 0.90**) than the foliage traits (r 0.52** 0.84**). The correlation coefficients between bunch weight and number of fruits per bunch in all sites and cropping cycles were significant and positive, ranging between 0.64** and 0.91**. Furthermore, principal component analysis (PCA) and stepwise multiple regression analysis retained the number of fruits per bunch (FNGS) and number of standing leaves at harvest (NSLH) as the most reliable predictors of bunch weight in this study. Thus, FNGS and NSLH could be used as indirect selection criteria for bunch weight in future breeding efforts.
Palabras clave
banana breeding, indirect selection, multiple regression, phenotypic correlation, principal component analysis, agronomic traits, germplasm