Calculated t was tested against table value of ‘t’ at error degrees of
freedom, where ve = error mean square obtained from R.B.D. analysis, F1 = mean of F1, BP = mean of the better parent, MP = mean
of the two parents, S.E. = standard error.
2.4.3. Estimation of dominance effect
The dominance estimates (D.E.) also referred to as “potence
ratio” was computed using the following formula as suggested by
Smith (1952).
D.E. = F1 −MP/0.5 × P2 – P1, where F1 =mean value of the hybrid
population; MP =mid-parent; P2 =mean of the highest parent;
P1 =mean of the lowest parent.
Complete dominance was realized when D.E. = +1; while partial
dominance is indicated when D.E. is between −1 and +1; D.E. = zero
indicates absence of dominance. Over dominance was considered
when D.E. exceeds ±1. The ‘+’ and ‘−’ signs indicate the direction of
dominance of either parent.
2.4.4. Estimation of combining ability and gene action
Analysis of variance table for combining ability with expectation of mean square was set up as follows:
Source d.f. M.S. Expectation of mean square
General combining ability (p − 1) Mg
2
e + (P + 2)
1/ (p − 1)
g
2
i
Specific combining ability p(p − 1)/2 Ms
2
e +
2/p (p − 1)
i
j
s
2
ij
Error m M
e
2
e
Combining ability variances and effects were worked out
according to Griffing’s (1956) Model 1 and Method 2. Method 2
is applicable to the present study as parents and one set of nonreciprocal F
1
s were included. Model 1 assumes that variety and
block effects are constant but environmental effect is variable and
the experimentalmaterial is the population aboutwhich inferences
are to be made.
The additive and non-additive genetic variances were estimated
from the combining ability components as follows:
ˆ
2
a
(additive) = 2ˆ
2
g
where ˆ
2
g = 1/ (p − 1)
i
ˆ
2
g
i
=
Mg −M
e
p+2
ˆ
2
na = (non-additive) = ˆ
2
s
where ˆ
2
s = 2/p (p − 1)
i
j
sˆ
2
ij
= Ms
− M
e
and M
e = ˆ
2
e
Statistical analyses were done using SPSS Professional Statistics
version 7.5 (SPSS Inc., Chicago, IL).
3. Results
3.1. Genetic diversity of the genotypes through multivariate
analysis
The intra-and inter-cluster distance represent the index of
genetic diversity among clusters. The distance among the genotypes revealed that Cluster I exhibited the maximum intra-cluster
value indicating that genotypes belonging in this cluster are diverse
in nature (Table 1). On the other hand, Cluster VI had the minimum intra-cluster value. At inter-cluster level, the minimum value
was observed between Cluster I and II indicating close relationship among the genotypes included inthese clusters. Themaximum
inter-cluster value was observed between cluster III and V followed
by between Cluster III and VI which indicated that the genotypes
included in these clusters had the maximum divergence.