Research Article

Novel selection strategy for half-sib families of sour passion fruit Passiflora edulis (Passifloraceae) under recurrent selection

Published: June 27, 2019
Genet. Mol. Res. 18(2): GMR18305 DOI:
Cite this Article:
(2019). Novel selection strategy for half-sib families of sour passion fruit Passiflora edulis (Passifloraceae) under recurrent selection. Genet. Mol. Res. 18(2): GMR18305.


Several strategies have been employed in the breeding of passion fruit with a view to the generation of superior progeny. In an effort to develop more precise methods in breeding, we compared the efficiency of the Post-Hoc Blocking Row-Col technique, which is an a posteriori technique that consists of the overlapping of a block structure on the original-field design, with a randomized-block design and compared different selection strategies within and among half-sib families, using the REML/BLUP mixed-model methodology. Twenty-three half-sib families from the third cycle of recurrent selection of the breeding program of Universidade Estadual do Norte Fluminense Darcy Ribeiro - UENF were evaluated. The trial took place in the experimental unit of UENF, in Itaocara - RJ, Brazil. Plants were trained on vertical stakes, with four replicates and three plants per plot. They were assessed individually for the traits number of fruits per plant, fruit mass per plant, fruit length, fruit diameter, peel thickness, total soluble solids, pH, pulp percentage, and production per plant. No significant difference was found in the test of efficiency of the designs for any of the evaluated traits. Within-family heritability (h2ad) had a similar magnitude to individual heritability (h2a), indicating that even in the 4th cycle of recurrent selection, genetic variability still exists within the evaluated progeny. Selection within half-sib families provided superior gains when compared with selection among families for the traits number of fruits; production; fruit mass, length, and diameter; total soluble solids; pH; and pulp percentage. The best selection strategy was within families, as it generated higher selection-gain estimates than those obtained with selection between families and the direct-selection and index-selection approach.