Premium
A Combined Univariate and Multivariate Approach for Selecting High Performing Genotypes of Vicia faba L.
Author(s) -
Pace C.,
Filippetti A.,
Ricciardi L.
Publication year - 1988
Publication title -
plant breeding
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.583
H-Index - 71
eISSN - 1439-0523
pISSN - 0179-9541
DOI - 10.1111/j.1439-0523.1988.tb00241.x
Subject(s) - univariate , selection (genetic algorithm) , diallel cross , biology , selfing , multivariate statistics , statistics , multivariate analysis , vicia faba , multivariate analysis of variance , mathematics , horticulture , hybrid , computer science , population , machine learning , demography , sociology
High performing V. faba genotypes have been successfully selected using an approach that combines the univariate and multivariate biometrical analyses of F|S from two complete 7×7 and 9×9 diallel cross experiments. The method used in the analyses included canonical analysis and cluster analysis of the phenotypic and genetic variance‐covariance matrices. All analyses have been applied, to the yield and yield components. Results of analyses provided information on the unit of selection, the selection criteria, and the selection procedure. The choice of an array of F 1 S seems to be more efficient than the choice of a single F 2 . Among the yield components studied, 100‐seed weight is the most important selection criterion which can improve yield most efficiently. The detected importance of the nonadditive genetic effects is that selection after intercrossing random plants from the mixed F 1 progenies of the same array is expected to be more effective than selection after Selfing.