Phenotypic Characterization Of Cowpea.

Abstract

Cowpea is indigenous to the African continent and is usually grown for its leaves and grain in different countries of the world. The objective of this study was to evaluate cowpea accessions using morphological descriptors and phytochemical analysis to identify cowpea accessions with a more specific traits that could be used by breeders. Phenotypic characterization of twenty-seven (27) cowpea accessions which consist of twenty-four (24) from the Department of Plant and Environmental Biology as test material and three (3) others, namely ‘Asontem’, ‘Wang Kae’ and ‘Kirkhouse’ which were used as checks, was undertaken. The cowpea population was evaluated based on agro-morphological traits (quantitative and qualitative) and phytochemical traits. STATA version 15.0 was used to carry out descriptive, multivariate analyses whiles ‘R’ software was used for genetic component analysis to determine the extent of variation. Phytochemical analysis was carried out to study the polyphenolic and amino acid contents of the cowpea seeds using standard methods and protocols. Polyphenolic compounds were determined using a spectrometer and amino acid contents were determined with assistance from Ghana Standard Authority. In the morphological qualitative traits studied, erect cowpea plants showed the highest distribution for growth habit with frequencies of 91% in the test material and 84.17% in the entire population. The pale tan pod colour recorded the highest frequency of 62.58% in the entire population, similarly, the pale tan with frequencies 65.48% and 62.58% were the highest frequency in both control and test material respectively. The white seed coat colour had the highest frequency of 66.11%in the entire population as well. The chi-square test of association between qualitative traits showed 86 significant associations in the test materials, with growth habit having the

highest signicant associations with leaf size (χ² = 220.84, P = 0.000), whereas the control showed 78 significant associations. The overall population exhibited 86 significant associations, with growth habit having the highest positive significant association with terminal leaflet shape (χ² = 243.87, P = 0.000). The phenolic compound had higher mean concentration in the controls than the test material, however higher mean concentration of amino acid was obatined in the test materials than in the control. A total of 145, 113 and 123 significant pairwise correlations were observed in the control, test materials, and entire population respectively for all quantitative traits. In the test material vanillic acid was highly correlated with gallic acid (r = 1.00) and p-coumaric acid (r = 1.00). Principal component analysis showed that, the first seven principal components in the morphological traits accounted for a total variability of 84.00% in the control, while nine principal components explained 81.00% of the total variation in the test material. On the other hand the first five principal components in the phytochemical traits accounted for 100%, 100% and 83% of the total variation in the control, test materials and entire population respectively. The biplot showed that the relationships among accessions and morphological, phytochemical and all traits explained 33% and 51% of the total variance respectively. The cluster analysis for the morphological traits revealed five distinct clusters. Genotypes named Asontem, T3, T5, T22 were grouped in a single cluster and were characterized by least days to germination (3.58), highest mean number of branches (4.93), least mean number of leaves (41.26), highest mean number of nodes (9.83), least mean leaflet width (3.69), the highest mean days to first flowering, least mean number of pods per peduncle (2.81), least average pod width (0.80), highest mean number of pods per plant (49.50), the highest average number of seed per pod (15.97), highest mean days to first matured pod (42.01), highest mean

number of seeds per plants (795.75) compared to other clusters. These genotypes could be used as parents in genetic improvement programs aiming to increase yield. Biometrical analysis revealed that the phenotypic variance (σ2 p) was higher than the genotypic variance (σ2 g) regarding all morphological traits evaluated. The number of seeds per plant reported the highest genotypic (80841.20) and phenotypic (107162.90) variance while the lowest genotypic (0.09) and phenotypic (0.69) value was recorded by the days to germination and average seed thickness respectively. Low (≤20%) heritability broad sense values were observed in all the morphological traits and ranged from 0.33% for days to germination to 0.99% for seed weight.