Evaluation of cassava ('Manihot esculenta' Crantz) genotypes reveals great genetic variability and potential selection gain

• Published in: Australian Journal of Crop Science Vol. 9; no. 10; pp. 940 - 947
• Main Authors: Matoso Francisco Avijala, Leonardo Lopes Bhering, Leonardo de Azevedo Peixoto, Cosme Damiao Cruz, Pedro Crescencio Souza Carneiro, Constantino Estevao Cuambe, Anabela Zacarias
• Format: Journal Article
• Language: English
• Published: Lismore, N.S.W Southern Cross Publishers 01.10.2015
• Subjects: Analysis of variance; Breeding; Cassava; Crop yields; Plant breeding
• ISSN: 1835-2693

The aim of this study was to estimate genetic and phenotypic parameters of cassava in order to contribute in the orientation of selection strategies of breeding programs in cassava. In this sense, genetic and phenotypic variances, heritability coefficient, coefficients of genetic and phenotypic correlation, and expected gains from selection were estimated. Twenty-one genotypes of cassava collected from Mozambique were evaluated in a complete randomized block design (CRBD) with three replications for one year. The following traits were evaluated: plant height (PH); first branch height (FBH); shoot biomass weight (SBW); number of roots per plant (NRP); tuberous root yield (TRY); production of commercial roots (PCR); harvest index (HI) and dry matter content (DMC). Statistical analyses were carried out using the GENES program. Genetic and phenotypic parameters were estimated among the evaluated traits. The coefficient of genetic variation ranged from 8.86 to 54.74%, and the highest CVg values were for shoot biomass weight (54.74%), and mean number of roots per plant (47.71%). The ratio of genetic and environmental variation coefficients was higher than 1 for 6 of the 8 evaluated traits. These same traits presented high values for heritability, indicating that most of the observed variation is of genetic nature. The genotypic correlations were higher than the phenotypic correlations in all cases. It was possible to identify positive and highly significant genetic correlations among SBW vs. TRY (rg = 0.85), and NRP vs. TRY (rg = 0.94). The graphical dispersion analysis, based on the first two canonical variables allows the formation of six clusters. Selection based on shoot biomass weight had indirect gain greater than direct gain for root yield. Therefore, based on these results, it was found that shoot biomass weight and number of roots per plant can be used as auxiliary criteria for selection of more productive cassava genotypes.