(go
to KOMUGI Home) (go
to WIS List) (go
to NO.83 Contents)
Gene action
The joint scaling tests indicated that both the additive-dominance
model and digenic epistatic model were inadequate to explain the
nature of the gene action for all the characters in either of the
environments (Table 2). This may be due to
influence of the higher order interactions and/or linkage among the
genes governing the inheritance of these traits, but further testing
of data was not feasible due to the limited number of generations.
Both additive and dominance gene effects played an important role in
determining inheritance of majority of the traits under both
environments. The magnitudes of dominance gene effects prevailed over
their respective additive gene effects for all the characters under
irrigated conditions (E1), and indicated that the
dominance gene action was more important under E1 for aU the
characters. But under rainfed conditions (E2) the
dominance gene action was more pronounced only for grain yield/plant,
tillers/ plant, days to heading and maturity, and biological
yield,while additive for the remaining traits. This implied that
selection in later segregating generations would be effective
especially under irrigated conditions. These results are also in
agreement with those of Tripathi et al. (1983) and Redhu (1988). As
the results are based on mean performance and may be influenced by
cancellation effects, the information generated through variance
approach should also be considered.
Considering the interactions, it was revealed that the majority of
the characters were largely influenced by dominance x dominance (l)
type of gene effects for all the characters. A greater portion of
genetic variability can be attributed to dominant genetic effects
under both the environments. The signs of h and l components were
screened for all characters where both the components were
significant. It was noted that the h and l components possessed
opposite signs for majority of the characters irrespective of the
environments, thereby suggesting that difficulty would be encountered
in selecting for these characters. The complementary type of
interaction for grain yield and days to heading under E2
also indicated the chances of direct selection for these characters.
Gene dispersion was also verified by comparing the magrutude of h and
d, and the higher estimates of h than d for majority of the
characters indicated that the parents were in dispersion phase and
there was an accumulation of dominant parental genes in the
hybrids.
The results of components of variances, heritability in narrow sense
and genetic advance are shown in Table 3. All the components of
variances were significant under both the environments. In addition,
the magnitude of variances was, in general, higher under
E1 than under E2. The lower magnitude of
variation under E2 may be due to suppressed expression of
genotypes under stress conditions (Ludlow and Muchow 1990). Degree of
dominance indicated that additive gene effects appeared to be the
important factor contributing to the genetic control of majority of
the characters under both the environments. This does not agrees with
the results obtained on the basis of generation mean analysis which
may due to cancellation of positive and negative gene effects
responsible for dominance at most of the loci. The relative magnitude
of degree dominance under E1 and E2 revealed
that, in general, expression of dominance component suffered more
than the additive component under stressed soil. This also agrees
with the results obtained from gene effects (Table
2). The estimates of various effects are valid under the
assumptions: (i) diploid segregation, (ii) homozygous parents, (iii)
absence of multiple alleles, (iv) absence of linkage and (v) no
genotype- environment interaction. The first two assumptions are
fairly met in wheat population. Multiple alleles arise as a result of
mutation. If the individual chosen to be the parents do not exhibit
multiple allelism, there is a very remote possibility that multiple
alleles will bias the estimates. In addition, there is no available
report on multiple alleles in the literature for the characters under
study. The remaining assumptions could not be tested. The failure of
any assumption may cause bias in estimates. The estimates of effects
are expected to be biased due to linkage in the presence of epistasis
(Kempthorne 1957).
Heritability estimates were high to moderately high for all the
characters except grain yield and tillers/plant under E1,
under E2, moderately high estimates of heritability were
observed for days to heading, moderate for harvest index and
biological yield/plant, and low for the remaining traits. The values
of expected genetic advance show possible gain from selection as per
cent increase in the F3 over the F2 mean when
most desirable 5% (K=2.06) of the F2 plants are selected.
The combined estimates of heritability and genetic advance indicated
the scope of selection for biological yield, harvest index,
grains/spike and tillers/plant under both the environments. Although,
100-grain weight, days to heading and days to maturity had moderate
to high heritability values under both the environments, yet the poor
variation in F2 limited their scope of selection. As
majority of these characters are interrelated, therefore, the
correlations among the characters should also be considered in the
process of selection.
<--Back | --Next
(
go to KOMUGI Home) (
go to WIS List ) (
go to NO.83 Contents )