Abstract:
Finger millet is an important cereal crop with high genetic potential in Ethiopia. However, its
production was challenged due to limited genetic information and availability of high- yielding
and adaptable varieties. The main objectives of this study were to assess the genetic variability,
heritability, genetic advance, and associations among yield and yield related traits including the
direct and indirect effects of yield-related traits on grain yield. Sixty four finger millet accessions
including the three released varieties were evaluated for thirteen traits in 2022 at Fadis
Agricultural Research Center using 8 x 8 simple lattice design. The results of analysis of
variance revealed highly significant differences among accessions for all traits. The grain yield
ranged from 687.5 kg ha-1
(Acc#215880) to 3791.6 kg ha-1
(Acc#215950) with a mean of 2889 kg
ha-1
. The yield of the highest yielding check (Tassema) was 3541.66 kg ha-1
. The genotypic
coefficient of variation (GCV) values ranged from 6.88% for days to heading to 33.51% for
harvest index while PCV values ranged from 7.59% for days to heading to 41.7% for harvest
index. Broad sense heritability values ranged from 41.33 % for number of effective tillers per
plant to 98.99% for days to maturity. High heritability coupled with high genetic advance was
recorded for days to maturity, finger length, finger width, biomass yield and harvest index;
indicating that these characters are controlled by additive gene action and phenotypic selection
for these characters will be effective. Grain yield had a highly significant positive correlation
with thousand seed weight and harvest index at both genotypic and phenotypic levels. Ear weight
and number of fingers per ear had positive correlation at phenotypic level and finger width at
genotypic level. Among these traits, harvest index, finger width and finger length had positive
direct effects on grain yield at genotypic level. Number of fingers per ear, thousand seed weight,
and harvest index had positive direct effects on grain yield at phenotypic level. This suggested
that direct and simultaneous selection of genotypes for grain yield of these traits is possible. The
first four principal components with Eigenvalues greater than one cumulatively explained about
72.9% of the total variation among the 64 finger millet accessions with respect to all the 13
traits. The genetic distance for all possible pairs of 64 finger millet accessions ranged from 0.29
to 29.3. The 64 finger millet accessions were grouped into six distinct clusters, of which cluster I
consisted of 15, 10 accessions in cluster II , 23 accessions in cluster III, 9 accessions in cluster
IV, and 3 accessions in cluster V and 4 accessions in cluster VI. This study showed the presence
of ample genetic variations among the finger millet accessions for yield and other traits studied
which could be potentially exploited in future breeding programs. The accession Acc#215950
and Acc#215852 were identified to be high yielders as compared to the standard checks and also
there is high genetic variability among them that used for future breeding programs. However,
further evaluation is required at least for one more year across multiple-location to draw
conclusive recommendations.