Jayashree Madhavan
Multivariate analysis of young and mature clones in para rubber (Hevea brasiliensis Muell. Arg.) - Vellayani Department of Plant Breeding and Genetics, College of Agriculture 2001
A study on young and mature clones of rubber (Hevea brasiliensisy was
undertaken with the objectives of estimating the variability, correlations, direct
and indirect effects of various traits influencing yield, estimating the genetic
divergence and the factors contributing to divergence, identifying those variables
that remain stable in both stages of the crop, and to examine the extent to which
mature yield could be predicted using immature attributes. A number of
morphological, structural, physiological and biochemical parameters were observed
at the two stages of growth.
Significant clonal differences at both stages were observed for almost all
the traits. There was high genetic variability at the mature stage for most traits,
especially girth increment, laticifer area index, initial flow rate, final volume of
latex, latex biochemical parameters (except for total solids content) and yield.
However number of stomata per unit area, density of latex vessels, diameter of
latex vessels, total solids content and chlorophyll a:b ratio had very low variability
as indicated by their genotypic coefficients of variation. It was shown that the
traits yield, girth, laticifer area index, initial flow rate, final volume of latex, plugging
index, latex thiols, inorganic phosphorous, sucrose, magnesium, chlorophyll band
total chlorophyll were most likely to respond to selection since there was a
preponderance of additive gene action in the inheritance of these traits, as indicated
2
by their moderate to high estimates of heritability coupled with high genetic advance.
High heritability coupled with low genetic advance observed for stomatal density,
bark thickness, leaf midrib lamina and palisade layer thickness indicate that these
are governed by non additive gene action.
Among the immature plants, high genetic variability was observed for
immature yield, time taken to sprout, number of flushes shed by the end of the
first year (W3), diameter increment, number of new flushes produced and those
retained on the entire plant in the second year (W5 and 7), number of latex vessel
rows, laticifer area index, latex thiols, inorganic phosphorous, sucrose and
magnesium and juvenile yield. Leaf size, specific leaf weight, density and diameter
of latex vessels and chlorophyll a: b ratio showed extremely low estimates of
genotypic coefficient of variation. Higher heritability estimates were recorded for
all the anatomical and biochemical traits except density of latex vessels, than for
the morphological traits, indicating the greater influence of environment on the
latter. High heritability combined with high genetic advance was recorded for most
of the anatomical and biochemical parameters and yield (except density and
diameter of latex vessels and .chlorophyll a:b ratio which had very low genetic
advance estimates) indicating the preponderance of additive gene action in these
traits. Moderate to high heritability followed by low genetic advance were seen
for the traits height, first year scion diameter bark thickness, diameter of latex
vessels, which implies the inheritance of these traits is governed mainly by non
additive gene action, and hence will not respond to selection.
At the mature phase, strong genotypic correlations of average annual yield
were observed with final volume of latex and initial flow rate, girth, girth increment,
number of latex vessel rows, laticifer area index, bark thickness and inorganic
3
phosphorous content, while at the immature stage, laticifer area index, scion
diameter in the second year, number of latex vessel rows, bark thickness, inorganic
phosphorous, thiol content, girth increment, number of new flushes produced and
those retained on the main stem in the second year, latex magnesium, chlorophyll /)
(
a, chlorophyll b, total chlorophyll, leaf size, and number of flushes produced on
the main stem in the two years (W8), were positively associated with testap yield
at both the phenotypic and genotypic levels. Biochemical parameters at both stages
contributed relatively less than the other characters to variation in yield, except
for inorganic phosphorous in the immature stage which showed moderate
association with yield.
Path analysis at the mature phase revealed that initial flow rate and bark
thickness could be used effectively as selection parameters for yield, as their direct
effects were the highest. Inspite of the moderate to high correlations of girth, girth
increment, number of latex vessel rows, final volume of latex, density and diameter
of latex vessels, selection for these traits per se will not effectively improve yield,
as their direct effects are 10w . .A negligible residue was obtained in the present
study, implying that almost all the variation in mature yield in the present study
could be accounted for by these variables. At the immature stage, number of latex
vessel rows was found to exert the highest positive direct effect on yield, while
bark thickness had a very high negative direct effect on yield. A residue of 0.33
indicates that other variables contributing to variability in yield at this stage have
not been included. These could be the physiological parameters initial flow rate
and final volume of latex, which had strong correlations with mature yield.
4
Genetic divergence was assessed using the D2 statistic and Tocher's
method of clustering was employed to group the clones in the two stages. Seven
and five clusters respectively were formed for the mature and immature groups of
clones. A great deal of similarity was found in the clustering pattern of the clones
at the two stages, inspite of the difference in age and the variables used to compute
the genetic distance. Most of the clones fell into one major group (Cluster I) with
18 and 19 clones respectively, of which 16 clones were in common. The clustering
patterns of the remaining clones were also similar, with many of them being
independent or forming two clone clusters. This indicates that though most of the
clones were genetically close as they fell into one cluster, the remaining clones
included in different clusters having divergence can be exploited in hybridization
programmes.
Factor analysis carried out in the two stages identified 10 factors at the
mature stage which were principally associated with yield, stomatal density, latex
biochemical components, initial flow rate, bark structural traits and chlorophyll
content. The nine factors identified at the immature stage were mainly associated
with vigour, yield, chlorophyll content and leaf structural traits.
The perfomance of the 25 clones at the two stages of growth was evaluated
on the basis of indices formulated using discriminant function analysis. There was
no correlation between the performance of the clones at the two stages.
Correlations between immature attributes and corresponding mature
attributes reveal that latex biochemical traits thiols, inorganic phosphorous, sucrose
and magnesium, bark structural traits number of latex vessel rows, laticifer area
5
index, stomatal density, density and diameter of latex vessels were relatively stable
over the years, while morphological traits and yield appear to be the most affected
with increase in age. Correlations between mature yield in the BI 2 panel and
immature attributes of two year old plants revealed that the association with
immature bark thickness and number of latex vessel rows were still retained at
this age. Step wise regression of mature yield on immature attributes showed that
number of latex vessel rows could explain only 21 per cent of the variability in
mature yield. As no good fit was obtained, yield at this stage cannot be predicted
using the first two years'attributes included in this study.
630.28 / JAY/MU
Multivariate analysis of young and mature clones in para rubber (Hevea brasiliensis Muell. Arg.) - Vellayani Department of Plant Breeding and Genetics, College of Agriculture 2001
A study on young and mature clones of rubber (Hevea brasiliensisy was
undertaken with the objectives of estimating the variability, correlations, direct
and indirect effects of various traits influencing yield, estimating the genetic
divergence and the factors contributing to divergence, identifying those variables
that remain stable in both stages of the crop, and to examine the extent to which
mature yield could be predicted using immature attributes. A number of
morphological, structural, physiological and biochemical parameters were observed
at the two stages of growth.
Significant clonal differences at both stages were observed for almost all
the traits. There was high genetic variability at the mature stage for most traits,
especially girth increment, laticifer area index, initial flow rate, final volume of
latex, latex biochemical parameters (except for total solids content) and yield.
However number of stomata per unit area, density of latex vessels, diameter of
latex vessels, total solids content and chlorophyll a:b ratio had very low variability
as indicated by their genotypic coefficients of variation. It was shown that the
traits yield, girth, laticifer area index, initial flow rate, final volume of latex, plugging
index, latex thiols, inorganic phosphorous, sucrose, magnesium, chlorophyll band
total chlorophyll were most likely to respond to selection since there was a
preponderance of additive gene action in the inheritance of these traits, as indicated
2
by their moderate to high estimates of heritability coupled with high genetic advance.
High heritability coupled with low genetic advance observed for stomatal density,
bark thickness, leaf midrib lamina and palisade layer thickness indicate that these
are governed by non additive gene action.
Among the immature plants, high genetic variability was observed for
immature yield, time taken to sprout, number of flushes shed by the end of the
first year (W3), diameter increment, number of new flushes produced and those
retained on the entire plant in the second year (W5 and 7), number of latex vessel
rows, laticifer area index, latex thiols, inorganic phosphorous, sucrose and
magnesium and juvenile yield. Leaf size, specific leaf weight, density and diameter
of latex vessels and chlorophyll a: b ratio showed extremely low estimates of
genotypic coefficient of variation. Higher heritability estimates were recorded for
all the anatomical and biochemical traits except density of latex vessels, than for
the morphological traits, indicating the greater influence of environment on the
latter. High heritability combined with high genetic advance was recorded for most
of the anatomical and biochemical parameters and yield (except density and
diameter of latex vessels and .chlorophyll a:b ratio which had very low genetic
advance estimates) indicating the preponderance of additive gene action in these
traits. Moderate to high heritability followed by low genetic advance were seen
for the traits height, first year scion diameter bark thickness, diameter of latex
vessels, which implies the inheritance of these traits is governed mainly by non
additive gene action, and hence will not respond to selection.
At the mature phase, strong genotypic correlations of average annual yield
were observed with final volume of latex and initial flow rate, girth, girth increment,
number of latex vessel rows, laticifer area index, bark thickness and inorganic
3
phosphorous content, while at the immature stage, laticifer area index, scion
diameter in the second year, number of latex vessel rows, bark thickness, inorganic
phosphorous, thiol content, girth increment, number of new flushes produced and
those retained on the main stem in the second year, latex magnesium, chlorophyll /)
(
a, chlorophyll b, total chlorophyll, leaf size, and number of flushes produced on
the main stem in the two years (W8), were positively associated with testap yield
at both the phenotypic and genotypic levels. Biochemical parameters at both stages
contributed relatively less than the other characters to variation in yield, except
for inorganic phosphorous in the immature stage which showed moderate
association with yield.
Path analysis at the mature phase revealed that initial flow rate and bark
thickness could be used effectively as selection parameters for yield, as their direct
effects were the highest. Inspite of the moderate to high correlations of girth, girth
increment, number of latex vessel rows, final volume of latex, density and diameter
of latex vessels, selection for these traits per se will not effectively improve yield,
as their direct effects are 10w . .A negligible residue was obtained in the present
study, implying that almost all the variation in mature yield in the present study
could be accounted for by these variables. At the immature stage, number of latex
vessel rows was found to exert the highest positive direct effect on yield, while
bark thickness had a very high negative direct effect on yield. A residue of 0.33
indicates that other variables contributing to variability in yield at this stage have
not been included. These could be the physiological parameters initial flow rate
and final volume of latex, which had strong correlations with mature yield.
4
Genetic divergence was assessed using the D2 statistic and Tocher's
method of clustering was employed to group the clones in the two stages. Seven
and five clusters respectively were formed for the mature and immature groups of
clones. A great deal of similarity was found in the clustering pattern of the clones
at the two stages, inspite of the difference in age and the variables used to compute
the genetic distance. Most of the clones fell into one major group (Cluster I) with
18 and 19 clones respectively, of which 16 clones were in common. The clustering
patterns of the remaining clones were also similar, with many of them being
independent or forming two clone clusters. This indicates that though most of the
clones were genetically close as they fell into one cluster, the remaining clones
included in different clusters having divergence can be exploited in hybridization
programmes.
Factor analysis carried out in the two stages identified 10 factors at the
mature stage which were principally associated with yield, stomatal density, latex
biochemical components, initial flow rate, bark structural traits and chlorophyll
content. The nine factors identified at the immature stage were mainly associated
with vigour, yield, chlorophyll content and leaf structural traits.
The perfomance of the 25 clones at the two stages of growth was evaluated
on the basis of indices formulated using discriminant function analysis. There was
no correlation between the performance of the clones at the two stages.
Correlations between immature attributes and corresponding mature
attributes reveal that latex biochemical traits thiols, inorganic phosphorous, sucrose
and magnesium, bark structural traits number of latex vessel rows, laticifer area
5
index, stomatal density, density and diameter of latex vessels were relatively stable
over the years, while morphological traits and yield appear to be the most affected
with increase in age. Correlations between mature yield in the BI 2 panel and
immature attributes of two year old plants revealed that the association with
immature bark thickness and number of latex vessel rows were still retained at
this age. Step wise regression of mature yield on immature attributes showed that
number of latex vessel rows could explain only 21 per cent of the variability in
mature yield. As no good fit was obtained, yield at this stage cannot be predicted
using the first two years'attributes included in this study.
630.28 / JAY/MU