Abstract

Ndiso J.B, Mugo S, Kibe A.M and Pathaka R.S

Crop genetic diversity is the basis of our food supply and survival. Determination of the phenotypic diversity of Kenya local coastal maize landraces (LCML) was important to understand the dynamics of maize genetic resources for the improvement and sustenance of maize productivity in the coastal region. The experiment with 30 genotypes was laid out in a randomized complete block design with three replications. Data collected included days to anthesis, days to silking, number of leaves, ear height, plant height, ears per plant, grain yield and anthesis-silking interval. Analysis of variance of morphological traits was performed using a general linear model (GLM) of SAS computer version 9.1. Pattern analysis for the relationship among genotypes was achieved through cluster analysis; dendogram were developed using a hierarchical agglomerical clustering method. Association among genotypes identified by principal component analysis were portrayed by proximity plots. Two major groups were found; Group 1- 1 consisted of the 28 coastal germplasm while group1 – 2 consisted of the two check entries being OPVs from outside the coast region (Entries 26 and 28). Group 2 - 1 had 9 entries, which include entries 7, 18, 9, 10, 17, 3, 11, 12 and 14. This group is dominated by entries from Kilifi and Lamu Counties. Group 2 – 2 had 19 entries, which include entries from Kilifi and Kwale districts. Entries from Taita Taveta and Lamu appeared only in G 2 - 1. Other checks such as PH 4, CLS-3 and KDV-3 were also in Group 2 – 2. This may indicate that these checks were developed from the local landraces from Kilifi and Kwale Counties. This study showed that LCML displayed large amounts of variation for morphological traits. The pattern of forming clusters may have some geographical implication since some clusters were formed with entries from neighboring Counties. The broad trait diversity evident among the local coastal maize landraces suggests ample opportunity for genetic improvement of the crop through selection directly from the accession and/or the development of inbred lines for future hybrid programs.