Main Article Content

Genetic variability in cowpea (<i>Vigna unguiculata</i> (L.) Walp.) genotypes


Abe S Gerrano
Patrick O Adebola
Willem S Jansen van Rensburg
Sunette M Laurie

Abstract

Information on genetic variability among the existing cowpea genotypes will increase the efficiency of the cowpea improvement. Field experiments were conducted at the Agricultural Research Council–Roodeplaat Vegetable and Ornamental Plant Institute in South Africa, in 2011 and 2012, to estimate the level of phenotypic variability among a collection of 25 cowpea genotypes. The experiment was laid out in a randomised complete block design with three replications. Sixteen phenotypic markers were recorded. Analysis of variance for the phenotypic traits revealed that differences among genotypes were highly significant for all traits. This indicated the high level of genetic variability among the cowpea genotypes studied. Genetic and phenotypic coefficient of variation, and broad-sense heritability were estimated for all phenotypic traits. The first five principal components showed 79.30% of the total variability among the genotypes. Pod length, leaf area, leaf area index and number of seeds per plant contributed mainly to PC1 and leaf number, plant height, dry biomass and fresh biomass contributed mainly to PC2. Cluster analysis of the phenotypic traits resulted in five distinct groups of genotypes. The phenotypic traits therefore provide a useful measure of genetic distances among the cowpea genotypes and will enable the identification of potential parental materials for future breeding efforts. Genotypes IT93K129-4, Fahari, Glenda and Veg cowpea Dakama Cream were associated with desirable grain yield characteristics and are recommended as suitable parental lines for improvement of grain production. Genotypes 5431, Tatro mix, Kisumu mix and Okalulenu were identified to possess good vegetative traits and are also recommended for use as suitable parents when breeding for leafy vegetable or for fodder production.

Keywords: heritability, phenotype, principal component analysis, variance


Journal Identifiers


eISSN: 2167-034X
print ISSN: 0257-1862