Abstract:
Although wheat has a long production history in Ethiopia, the mean national yield of the crop
is relatively low in contrast with the world average yield due to limited availability of
adaptable and stable varieties. Therefore, this study was conducted to estimate genetic
variations among 49 bread wheat genotypes and association among yield and related
characters to identify superior genotypes. Field experiment was conducted during 2018
cropping season at Kokate and Hossana, Southern Ethiopia. The experimental design used
was simple lattice. Data were collected on 12 quantitative and 6 qualitative characters and
all quantitative characsters were subjectd to analysis of variance using SAS statistical
analysis while qualitative characters were calculated using Shannon index. Analysis of
variance across locations revealed significant variations among location, genotype and
genotype x location interactions for most of the quantitative characters considered in the
study. Shannon index indicated appreciable diversity for most of the qualitative characters
studied. The phenotypic coefficient of variation ranged fron 4.78% for plant height to
26.26% for biological yield while the genotypic coefficient of variation ranged from 3.89%
for plant height to 23.69% for biological yield. Heritability ranged from 21.15% for number
of kernels per spike to 95.31% for days to heading. Heritability along with genetic advance as
percent mean for days to heading, biological yield and harvest index, respectively
were (95.31, 20.14), (81.40, 24.1) and (84.90, 38.31),showing presence of additive gene and
selection based on these characters would be ideal. The correlation analysis revealed number
of kernels per spike, biological yield, thousand seed weight and harvest indixe has positive
and significant association(P≤ 0.01) with grain yield. Path analysis revealed that biological
yield exerted positive direct effect on grain yield both at genotypic and phenotypic levels.
Cluster analysis grouped 49 genotypes into five clusters and two solitary groups. The highest
inter cluster distance occurred between clusters four and five while the lowest was between
clusters two and five. Principal component analysis revealed that five principal components
had accounted for 77.6% of the total variation. Generally, the study showed existence of
significant genetic variability among tested genotypes. Therefore, simple selection of
promising genotypes and crossing of highly divergent group to produce best heterotic
offspring could be recommended from the present study . For future breeding programs that
employ hybridization, parental material selection should be carried out between clusters
rather than within clusters.
