Plant Gene and Trait, 2015, Vol.6, No.7, 1-11
4
no. of secondary branches/plant, days to first
flowering and days to first fruiting, fruit length (cm)
and fruit stalk length (cm), fruit circumference(cm)
and fresh weight/fruit (g), no. of fruits in
inflorescence/plant and fruits/plant also showed strong
positive association among them. Significant positive
association also observed among different pairs of
traits both genotypic and phenotypic level indicated
that simultaneous selection of these traits would be
effective for improvement of yield as same as
concluded by Shinde et al. (2012).
The correlation coefficient between yield and a
particular trait is the ultimate result of direct effect of
that attribute and indirect effect over other attributes.
Direct selection based on correlation coefficient may
be misleading, at that case path coefficient analysis
assess the actual impact of causal variable on yield
through direct and indirect effects as presented in
Table 3. Highest positive direct effect on yield was
observed by no. of fruits in inflorescence/plant and
fresh weight/fruit (g) in both genotypic and
phenotypic level. So direct selection based on these
two traits would be effective to increase yield of
brinjal as same as reported by Shende et al. (2014) for
length of fruit, number of fruits per cluster, plant
height, days to last picking, average weight of fruit
and number of fruits per plant.
No. of secondary branches/plant had high positive
direct effect on yield at genotypic level and positive
direct effect on yield at phenotypic level. No of fruits
in solitary/plant showed significant positive direct
effect on yield both genotypic and phenotypic level.
These are the moderate contributing characters
towards yield and selection based on these characters
could also be effective for developing high yielding
brinjal varieties. Similar conclusion were reported by
Kumar et al. (2013b) for Plant height, number of
branches per plant, fruit length, fruit pedicel length,
number of fruits per plant, average fruit weight and
little leaf incidence because of positive direct effect
on yield. Fruits/plant would also be effective to
increase yield because of its strong positive
correlation with yield/plant and high indirect effect
on yield/plant via the no.
of fruits in
inflorescence/plant. Similar observation in brinjal
was reported by Ullah et al. (2014).
At genotypic level residual effect was recorded
0.36375 and at phenotypic level residual effect was
recorded 0.31839 (Figure 1). This indicated that
character considered in the present study responsible
for 64% variation in genotypic level and 69%
variation in phenotypic level. Residual effects both
genotypic and phenotypic level indicated that some
other possible characters need to include in the
analysis to get entire variation in fruit yield as same as
concluded by Kumar and Arumugam (2013).
The character yield/plant (g), fresh weight/fruit (g), no.
of fruits in inflorescence/plant, no. of fruits in
solitary/plant and no. of secondary branches/plant
were considered to estimate selection index (Table 4)
based on genetic variability, significant genotypic
correlation with yield and high direct positive effect
towards yield/plant as same suggested by
Chattopadhyay et al. (2011).
It was observed that selection for individual character
index I
3
(no. of fruits in inflorescence/plant) showed
highest expected genetic gain (2.19) with highest
relative efficiencies (166.79%) over straight selection
for yield. It indicated that this trait is enough to
determine genotypic value for yield and can be
utilized as important aspect in the improvement of
these brinjal genotypes. Relative efficiencies of fresh
weight/fruit (g) (93.20%) and no. of fruits in
solitary/plant (97.55%) are close to the efficiency of
yield/plant (100%). Higher relative efficiencies over
straight selection for grain yield in rice was observed
by Habib et al. (2007) for individual characters like
filled grains/panicle, plant height and days to maturity.
When two characters were included in the selection
index, maximum relative efficiency (231.89%) with
highest expected genetic gain (3.04) over selection for
yield was obtained for the index I
13
(yield/plant (g) +
no. of fruits in inflorescence/plant). Purohit and
Majumder (2009) observed maximum relative
efficiency in rice for two characters namely grain
yield and number of effective tillers. But in a
conventional approach, breeders consider at least three
characters function. Considering three characters,
maximum relative efficiency (258.14%) was obtained
for index I
134
in combination of yield/plant (g), no. of
fruits in inflorescence/plant, no. of fruits in
solitary/plant with highest expected genetic gain (3.39).
