Genetic options for improving fodder yield and quality in forage sorghum
Aruna Ca., M. Swarnalathaa, P. Praveen Kumara, Visala Devendera, M. Sugunaa, M Blummelb, JV
Patila
a-
Directorate of Sorghum Research, Rajendranagar, Hyderabad, India
b-
International Livestock Research Institute, ICRISAT, Patancheru, Hyderabad, India
Corresponding author
Dr C. Aruna
Principal Scientist
DSR, Rajendranagar
Hyderabad- 500 030
India .
Email- aruna@sorghum.res.in
Telephone- +914024018651
Fax- +914024016378
Genetic options for improving fodder yield and quality in forage sorghum
Aruna C., M. Swarnalatha, P. Praveen Kumar, Visala Devender, M. Suguna, M Blummel, JV Patil
Abstract
Forage sorghum improvement for fodder quantity and quality is important, especially in semi-arid tropics
where sorghum is a major source of forage for the livestock. The aim of this work was to understand the
character associations among fodder yield and quality traits, and to estimate combining ability of the
parents. The experiment was carried out for two years using ten parents crossed in half diallele design.
Highly significant differences among the genotypes for fodder yield, quality and cell wall constituents
were observed. The correlation of the important quality traits, crude protein and digestibility (IVOMD)
with fodder yield were non-significant, indicating the possibility to improve yield and quality
simultaneously in forage sorghum. Partial least square (PLS) analysis showed that for improving fodder
yield and digestibility together, plant height, ADF, ADL, NDF, leaf number, stem diameter, days to
flower and leaf length were important. GCA and SCA variances and heterosis showed that for almost all
characters both additive and non-additive gene effects were important with predominance of non-additive
effects. Parental lines SEVS4, HC308 and UPMC503 were good combiners for fodder yield and quality.
The brown midrib lines, EC582508 and EC582510 were good combiners for low lignin and high
IVOMD. Strategies for improving forage sorghum to suit animal and biofuel industries were discussed.
Key words- Cellulose, hemicelluloses, digestibility, partial least square regression, diallel analysis, gene
effects
1. Introduction
Sorghum is a versatile species that can be used as a source for food, feed, fodder and fuel in the semi arid
tropical regions of the world. Sorghum with its high biomass production potential can be used for biofuel
production and for fodder purpose. The demand for fodder increases because of the recent efforts towards
increasing milk and meat production, which puts an immense pressure for increasing quantities of green
and dry fodder. Under the semi-arid situations, sorghum is the major supplier of fodder, and its role
becomes important during the lean period of winter and summer months. The management practices that
might improve fodder yield and quality, such as higher application of nitrogen, may not be suitable in
semi-arid regions where the environment is becoming highly unpredictable and drought prone (Hall et al.
2004). The best options for increasing the yield and quality of forage sorghum appears to be genetic
improvement of both these characteristics in currently available cultivars. Considering the growing
demand for more and better quality fodder for livestock, forage sorghum improvement programs need to
become multidimensional, targeting improvement for both yield and quality traits. There is limited
information available on feed quality for improved forage sorghums which is important for
commercialization of forage cultivars. There is a need to develop forage cultivars with improved feed
quality so that reliable and better quality fodder is available year round.
Besides its utility as a fodder crop, sorghum has the potential to provide abundant and sustainable
resources of lignocellulosic biomass for the production of ethanol as biofuel (Carpita and McCann 2008).
The shorter life cycle of bioenergy grasses and their different cell wall composition specifically lower
lignin content, makes processing of biomass from grasses much less energy intensive (Vermerris et al.
2011). For using sorghum as a fodder and biofuel crop, it is important to improve the biomass quality in
terms of digestibility and saccharification of the stalk. Biomass quality is influenced by cell wall
composition and structure, and can be determined by content and composition of lignin, cellulose and
hemicelluloses. Lignin content of cell walls determine, among other factors, sugar release and thus the
efficiency of the ethanol extraction process (Vermerris et al. 2007; Lorenz et al. 2009). The main goal of
forage sorghum breeders is to develop cultivars with high fodder yield as well as cellulose, hemicelluloses
content and high digestibility, which will be stable across years and locations.
In comparison to other crop plants, maize has received particular attention regarding cell wall component
research. Before biofuel gained attractiveness, maize cell wall components were studied because of
digestibility in ruminants (Xie et al. 2009). Genetic improvement of biomass crops can significantly
reduce the overall cost of biomass to ethanol conversion. Improvements in fodder quality in sorghum are
dependent on genetic variability within the species, the heritability of the traits, selection intensity and the
ability of the plant breeders to understand the genetic architecture controlling these traits. Fodder yield
and quality are quantitative traits which are polygenetically controlled.
The nature and magnitude of gene action involved in the expression of quantitative traits is important for
successful development of crop varieties and cultivars through proper choice of parents for hybridization
(Griffings 1956; Falconer 1989). Diallel analysis can be used to provide important information on general
and specific combining abilities (GCA and SCA), determine genetic variances and estimate heritability.
Combining ability describes the breeding value of parental lines to produce hybrids. General combining
ability (GCA) refers to the average performance of a parent in hybrid combinations and specific
combining ability (SCA) is the performance of a parent relatively better or worse than expected on the
basis of the average performance of the other parents involved (Griffings 1956). Combining ability
analysis helps in the identification of parents with high gca and parental combinations with high sca.
Based on combining ability analysis of different characters, higher sca values refer to dominance gene
effects and higher gca effects indicate a greater role of additive gene effects controlling these characters.
The estimation of additive and non-additive gene action through this technique helps in determining the
possibility of commercial exploitation of heterosis.
The objectives of this study were to understand the genotypic variation for fodder yield and quality in a
set of forage sorghum genotypes, finding out the associations among the yield and quality, and to
determine combining ability of the forage sorghum genotypes and to understand the genetic basis of the
important fodder yield and quality traits. The selected promising parents could then be used in forage
sorghum breeding programs aiming towards improvement in fodder yield and quality.
2. Materials and methods
The study was conducted on the research farm of the Directorate of Sorghum Research (DSR),
Hyderabad, India for two years during the rainy seasons of 2009 and 2010. During both the years, the
crop was sown during the second week of June and harvested during third or fourth week of August
depending on flowering of the genotype. The environmental characterization during the crop period of the
years under study is given in Table 1. The rainfall during July month was observed to be high in second
year of study.
2.1 Field material for the study
The lines used in the study comprised of ten sorghum cultivars including seven forage sorghum lines, one
sweet sorghum genotype and two brown midrib genotypes; and 45 hybrids derived by crossing the ten
parents in a half diallel fashion. Description of the ten forage sorghum genotypes is presented in Table 2.
The parents and the F1s were evaluated in a randomized complete block design with three replications.
Each line was grown in a single row of the length 4m with an inter row spacing of 45 cm and 15 cm with
in row. Parents and F1s were randomly assigned and labeled accordingly.
2.2 Observations recorded
In each year, observations were recorded on the following fodder yield and quality parameters. Days to
flower were recorded on plot basis, while other parameters were recorded on10 plants/plot.
2.2.1 Field observations
Days to 50% flowering (DTF) were recorded as the number of days required from planting for 50% of the
plants in a plot to reach mid-anthesis. Plant height (PH) was recorded as the height of the plant from
ground level to the tip of the main tiller at flowering. The number of leaves per plant (NLP) were counted
at flowering, and the length (LL) and breadth (LB) of fourth leaf from ground level was recorded. Stem
diameter (SD) was measured as the diameter at the fourth internode from ground level. The plants were
harvested at 50% flowering to estimate fodder yield and to recover samples for quality analysis (see
below). The fodder yield (FY) was recorded immediately after harvest to avoid moisture loss. For this,
plants were harvested manually by cutting the stem at the base and the entire above ground plant material
was weighed.
For estimation of fodder quality, the chopped sample of green plant was dried in the hot air drier at 60700C for 72 hrs and powdered.
2.2.2 Observations on fodder quality
Brix, a measure of the mass ratio of soluble solids to water, is a widely used approximation for sugar
content (Audilakshmi et al. 2010). Brix values were recorded with hand refractometer at flowering from
the juice of stalk of each of the ten plants which were subsequently averaged. All forage samples were
analyzed by Near Infrared Spectroscopy (NIRS) calibrated for this experiment against conventional
chemical and in vitro analyses. The NIRS instrument used was a FOSS Forage Analyzer 5000 with
software package Win SI. Crude protein (CP) percentage was estimated by determining total nitrogen (N)
in the sample by Auto Analyzer and neutral detergent (NDF), and acid detergent fiber (ADF) and acid
detergent lignin (ADL) were analyzed according to Goering and Soest (1970). In vitro organic matter
digestibility (IVOMD) was determined and calculated according to Menke and Steingass (1988) using an
in vitro gas production test with manual syringes as modified by Blümmel and Ørskov (1993).
The cellulose and hemicelluloses fractions were calculated from NDF and ADF values. NDF estimates
the sum of cellulose, hemicelluloses and lignin, ADF the sum of cellulose and lignin, while ADL
measures lignin (ADL) only. Therefore, NDF minus ADF yields the hemicelluloses (HCL) content, and
ADF minus ADL corresponds to cellulose (CLL) content.
2.3 Data analysis
Data collected over the two environments was used for analysis of variance, and simple correlations using
the software Genstat. Partial least square (PLS) analysis was carried out for each environment with the
help of the statistical package SAS version 9.3 (SAS Institute, Carey, NC, USA). The fodder yield (FY)
and digestibility (IVOMD) were taken as dependent/ response variables, and plant height (PH), days to
50% flowering (DTF), number of leaves (NLP), leaf length (LL), leaf breadth (LB), stem diameter (SD),
brix, crude protein% (CP), ADF, NDF and ADL as independent/ predictor variables. For each
environment, three PLS models were tried: (i) fodder yield as dependent variable, (ii) digestibility as
dependent variable and (iii) fodder yield and digestibility as dependent variables. Coefficients of
determination (R2) and root mean prediction residual error sum of squares (PRESS) were used to select
the best models. The variable importance projection (VIP) of independent variables represented the value
of the variable in fitting the PLS model for both predictors and responses. The VIP of each factor was
defined as the square root of the weighted average times the number of predictors. If a predictor had a
relatively small coefficient (in absolute value) and a small value of VIP, then it was a prime candidate for
deletion (Wold 1994). Variables with VIP values less than 0.8 and outliers were removed from the
variable list.
The analysis of variance for GCA and SCA effects was carried out according to Griffing’s (1956) method
1, model 2 involving parents with one set of F1s but reciprocals were not included. Windostat (Indostat,
2004) software was used for analysis. GCA and SCA effects for different traits were calculated across
environments. The model: Yijk = µ + gi +gj +sij + eijk, where Yijk is the observed measurement for the ijth
cross grown in the kth replication or environment; µ is the population mean; gi and gj are the GCA
effects; sij the SCA effect; and eijk the error term associated with the ijth cross evaluated in the kth
replication or environment. The restrictions imposed on the combining ability effects are: ∑gi=0, and
∑sij=0 for each j (Griffing 1956). Estimates of σ²g (general combining ability), σ²s (specific combining
ability) and their variances were computed for the random-effects model to estimate σ²A, σ²D and h2
(Zhang and Kang 2005).
Heterosis (MP: mid-parent) and heterobeltiosis (BP: better parent) values were, respectively calculated by
using the formula [MP= (value of F1 – mean of parents/mean of parents) x 100]; [BP= (value of F1 –
value of better parent/ value of better parent) x 100]. The critical differences for testing the significance of
heterosis were calculated as follows: critical difference (MP) = √3Me/2r x t, critical difference (BP) =
√2Me/r x 4t, where Me is the error mean squares, r is the number of replications, and t is the table value
of t at 5 or 1% level of significance.
The genetic divergence among the parents used was computed by means of Mahalanobis’ D2 technique
(Mahalanobis 1930). The genotypes were grouped into clusters following Tocher’s method as described
by Rao (1952) using Windostat software (Indostat 2004).
3. Results
3.1 Mean values and heretabilities
The mean values and range for all the fodder yield and quality traits along with the level of significance
for 55 sorghum genotypes studied were reported in Table 3. Highly significant (P <0.001) differences
were observed for all fodder yield and fodder nutritional quality characteristics. Heritabilities for all
fodder quality and yield traits were moderate to high, which is sufficiently high to realize progress from
selection for improved fodder quality. The cultivar effects were highly significant (P <0.001) for all the
traits indicating the presence of variability for all the traits studied. The year effects were significant for
majority of the traits except for DTF, NLP, SD and brix. Cultivars x year interaction were significant for
all traits (Table 4).
3.2 Relationships among fodder yield and quality traits
No significant relationships were observed between fodder quality traits and fodder yield (data not
shown). Fodder quality and fodder yield seem to be compatible traits. Fodder crude protein was
nonsignificantly associated negatively with fodder yield (r= -0.08, p=0.56). Similarly NDF (r= -0.16,
p=0.25), ADF (r= -0.15, p=0.27) and ADL (r= -0.001, p=0.99) were also nonsignificantly associated
negatively with fodder yield, while IVOMD was positively but nonsignificantly associated with fodder
yield (r= 0.07, p=0.63). IVOMD recorded significant negative correlation with ADL (r= -0.86,
p=<0.001), while it is negative but non-significant with CP (r= -0.02, p=0.91). Brix showed significant
positive association with IVOMD (r=0.35, p=<0.001). CP showed non-significant associations with all
the fodder quality traits except with brix which is significant and negative (r= -0.52, p=<0.001).
Since simple correlation studies showed that the independent variables were correlated in the present
study, PLS analysis was carried out to understand the importance of each independent variable. PLS
regression is typically used when the independent variables are correlated, or the number of independent
variables exceeds the number of observations. It differs from standard multiple linear regression because
it accounts for the variation of both response and predictor variables (Jurs 1990). The PLS regression
matrix is presented in Table 5. For each environment, three regression models were tried: (i) fodder yield
as dependent variable, (ii) IVOMD as dependent variable and (iii) fodder yield and IVOMD as dependent
variables. The R2 for X (the independent variables) were similar across models, ranging from 0.58 to
0.68. For Y, the dependent variable, it ranged from 0.68 to 0.74. VIP is presented in Table 6. Different
traits were important for improvement of fodder yield and digestibility individually or both together. For
the improvement of fodder yield, PH, NLP, SD, DTF and LL were the important traits for explaining the
variability in fodder yield. For explaining variability in digestibility, ADF, ADL, NDF and plant height
were shown to be important. When both fodder yield and digestibility were considered, the important
traits being PH, ADF, ADL, NDF, NLP, SD, DTF and LL, while LW, Brix and CP were found to have
less contribution towards variability in either fodder yield or digestibility.
3.3 ANOVA, GCA and SCA variances
The mean square values after analysis of variance for genotypic differences and combining ability for all
the traits is presented in Tables 7&8. There were significant genotypic differences for all the characters
studied (p<0.001). Significant differences (p< 0.001) due to environment were also observed for all the
traits. The partitioning of genotype mean squares into GCA and SCA mean squares showed GCA and
SCA mean squares to be significantly different (p<0.001) for all the traits. Estimates of highly significant
gca and sca variances for these characters indicated the importance of both additive and non additive
genes in the expression of the characters. For HCL and NDF additive gene action was predominant. Both
additive and non-additive gene actions were equally important for the traits, LB, IVOMD, CLL and ADF.
While the non-additive gene actions were predominant for the traits brix, CP, LL, SD, PH, NLP, ADL,
DTF and fodder yield. Significant differences for genotype, GCA and SCA mean squares implied that
there was scope for improving these traits.
Genotype by environment interactions were highly significant (P <0.001) for all traits. Partitioning of
genotype by environment interactions into GCA by environment (GCA x E) and SCA by environment
(SCA x E), showed that: (a) GCA x E was significant (P <0.05) for all the traits except PH and NLP, and
(b) SCA x E was significant (P <0.05) for all the traits except CP. Significant GCA x E interactions for
the above traits is an indication of variation of general combining ability of lines under different
environments. Significant SCA x E interaction for the traits mean that specific hybrids differed in the way
they expressed these traits under different environments.
3.4 GCA and SCA effects
3.4.1 Fodder yield traits
Estimates of gca effects of the ten genotypes for fodder yield traits showed that SEVS4, HC308 and
UPMC503 were the best combiners for FY (Table 9). SEVS 4 and HC 308 recorded high per se
performance for fodder yield (Table 11). Apart from fodder yield, SEVS4 and HC308 were good general
combiners for other yield components including PH, SD and leaf parameters such as NLP, LL and LB.
For DTF, EC582508, Nizamabad forage, Keller, EC582510 and SSG59-3 were good general combiners
with significantly negative gca effects for flowering. The parents, PC 23, Nizamabad forage, EC 582508
and Keller were early with DTF less than 65 days. The sweet sorghum line, Keller was a good combiner
for LL, LB and SD, besides earliness.
The estimates of sca for fodder yield traits for promising cross combinations are presented in Table 12.
The crosses, PC23 x EC582508, UPMC503 x EC582508, UPMC512 x SEVS4, EC582510 x Niz forage,
SSG59-3 x UPMC512, HC308 x EC582510, EC582510 x SEVS4 and PC23 x Keller exhibited high sca
effects for fodder yield. All these crosses were good combiners for the yield components such as PH and
leaf parameters also.
3.4.2 Fodder quality traits
Estimates of gca effects for fodder quality traits indicate that the genotypes, Keller, HC308 and
UPMC503 were the best genotypes for brix which indicates the sweetness of the stem (Table 10). High
brix values were observed in the parents, Keller and EC 582510 (Table 11). For CP, only one genotype,
Nizamabad forage was a good general combiner and the same parent recorded high CP value individually.
For IVOMD, the brown midrib genotypes, EC582508 and EC582510 and the sweet sorghum genotype,
Keller were good combiners. All the three parents had high per se performance for IVOMD.
The crosses, PC23 x EC582508, UPMC503 x UPMC512, PC23 x Nizamabad forage, UPMC512 x
HC308, UPMC512 x Nizamabad forage, PC23 x Keller, PC23 x SEVS4 and HC308 x EC582508
recorded significant sca effects for brix. For CP, Keller x EC582510, UPMC 503 x EC 582510 and
Nizamabad forage x EC 582508 showed high sca effects. While for IVOMD the crosses, SSG59-3 x
Keller, UPMC512 x PC23, UPMC503 x HC308, PC23 x HC308, SSG59-3 x EC582510, UPMC512 x
EC582508 and HC308 x EC582510 recorded significant sca effects.
3.4.3 Cell wall constituents
For the fibre components, NDF and ADF, the genotypes Keller, SEVS4, EC582508 recorded negative
significant gca effects. While for HC308 only gca of ADF was significant and for Nizamabad forage gca
for NDF was significant. Positive significant GCA effects for ADF and NDF were observed for PC23
and UPMC512. Significantly positive gca effects for CLL and HCL were also observed for PC23 and
UPMC512, while the genotypes Keller, SEVS4, EC582508 and HC308 recorded negative gca effects for
CLL and HCL. Low CLL and HCL values were observed in the genotypes, Keller and SEVS 4 (Table
11). For ADL, the brown midrib genotypes, EC582508 and EC582510 had significantly negative gca
effects besides HC308. Both the brown midrib lines, EC 582508 and EC 582510 had low lignin compared
to other genotypes.
The crosses, SSG59-3 x Keller, PC23 x HC308, PC23 x Nizamabad forage and EC582510 x Nizamabad
forage recorded negative sca effects for NDF and ADF. While positive sca effects for ADF and NDF
were reported in the crosses, UPMC 512 x Keller, HC 308 x EC 582510, Keller x EC 582510, UPMC 503
x SEVS 4, UPMC 512 x EC 582510 and PC 23 x Keller. The crosses, UPMC 512 x Keller, PC 23 x
Keller, HC 308 x EC 582510, Keller x EC 582510 exhibited positive and significant sca effects for both
cellulose and hemicelluloses. The crosses, SSG59-3 x Keller, SSG 59-3 x EC 582510, UPMC 512 x EC
582508, PC 23 x SEVS 4 and HC308 x EC 582510 had significant sca effects in desirable direction for
lignin.
3.5 Heterosis for fodder yield and quality traits
High heterosis was observed for FY followed by brix and CP (Fig. 1). Mean heterosis for FY was
observed to be 22.2 and 12.6% over the mid and better parents respectively. The best hybrid with high
heterosis was PC23 x EC582508 with 142% and 80% heterosis over mid and better parents, followed by
UPMC503 x EC582508. Both the hybrids showed significant heterosis over better parent, while thirteen
more hybrids expressed significant heterosis over midparent value (data not shown). For brix, maximum
heterosis was observed in PC23 x EC582508 which was 54% and 38.4% over mid and better parents.
Other hybrids with significant heterosis were PC23 x Nizamabad forage and UPMC512 x PC23. For CP,
the best hybrid was UPMC503 x EC582510 with heterosis of 34.7% and 31.1% over mid and better
parents respectively, followed by SSG59-3 x EC582510 (32.7% and 24.2%) and SSG59-3 x UPMC503
(25.9% and 15.2%). For IVOMD only one hybrid, PC23 x Nizamabad forage recorded positive
significant heterosis. Heterosis for less ADL was observed in PC23 x Nizamabad forage (-7.53% and 2.38% over mid and better parents), HC308 x Nizamabad forage (-5.19 and -1.78%) and UPMC503 x
HC308 (-6.96% and -5.5%).
3.6 Genetic divergence studies
The ten sorghum genotypes included in the study were grouped into three clusters based on D2 values
(Table 8). Cluster I was having five genotypes, comprising of the two brown midrib genotypes
(EC582508 and EC582510), and the forage lines HC308, SSG59-3 and SEVS4. Cluster II consists of four
genotypes including the sweet sorghum line, Keller and the forage lines Nizamabad forage, UPMC503
and PC23. The cluster III consists of a single genotype, UPMC512. The cluster distance ranged from 32.6
to 34.4 with in clusters and from 39.8 to 52.1 between clusters (Table 9). The maximum diversity of 52.1
was evident between Cluster II and Cluster III.
4. Discussion
The forage sorghum breeding program should aim for improvement of the important fodder quality traits
such as high digestibility of fodder and high protein content besides fodder yield. Forage sorghum has
good potential as a biofuel and biogas crop also (Mahmood and Honermeier 2012). In recent times,
ethanol demand has been increasing drastically due to Indian government’s policy to blend it with
automotive fuels, for achieving fuel sufficiency. To produce ethanol from plant biomass, cellulose/
hemicelluloses in the cell wall will need to be saccharified, which is the hydrolyzation of cell wall
polysaccharides into fermentable sugars (glucose and xylose). During sachharification process, lignin acts
as a physical barrier and retards the action of cellulases, impeding swelling of cellulose fibres and
nonspecifically binding cellulose proteins (Vermerris et al. 2007). Reducing lignin had a highly beneficial
effect for converting cellulose to glucose resulting in high ethanol yields (Dien et al. 2009).
The brown midrib mutants (bmr) in forage maize and sorghum with reduced lignin and greater
digestibility (Barriere et al. 2004; Sattler et al. 2010; Rao et al. 2012) could lead to the development of
forage and sweet sorghums as novel biomass crops (Sarath et al. 2008). This is because the accessibility
of cellulose and hemicelluloses by rumen microorganisms is improved by lowering the amount of lignin
present in cell walls. Biorefineries present a comparable system with rumen digestive system where
improved cellulose breakdown with enzyme mixtures rather than rumen bacteria to sugars. Goals shared
between forage and biomass feedstock improvement include high biomass yield, high cellulose,
hemicellulose content and low lignin content.
Among the grasses, maize has received the most attention in genetic and genomic studies related to cell
wall lignifications and degradability (Barriere et al. 2008, Lorenz et al. 2009). Biomass/fodder yield of
energy crops positively correlates with plant height, stem density, and stem thickness and leaf parameters
such as leaf number, leaf length etc as demonstrated in Miscanthus, swithgrass and sorghum (Atienza et
al. 2003; Murray et al. 2008; Zhao et al. 2009; Iyanar et al. 2010; Tariq et al. 2012). Height is independent
of stem structural composition, i.e. cellulose, hemicelluloses and lignin content (Murray et al. 2008),
which means that a variety with tall stem can be bred to contain more cellulose and less lignin.
In the present study, all traits were significant by the F test, indicating that there is sufficient genetic
variability in the parents and hybrids, which is of fundamental importance for obtaining genetic gains in
hybrid combinations. The highly significant differences among genotypes for the fodder yield and quality
traits indicate that genetic improvement of fodder quality of forage sorghums should be possible. Highly
significant genotypic as well as genotype x year differences were observed for all the traits studied. The
estimates of heritability of pertinent fodder quality traits were around 0.5 suggesting opportunities for
further improvement of fodder quality by genetic enhancement. The magnitudes of genotypic differences
found in this study should prove meaningful for animal performance, as small changes of in vitro dry
matter digestibility of 3-4% units have been observed to result in improvements of 17-24% in daily gains
and production per hectare (Vogel and Sleper, 1994). However, the very large genotype-environment
interactions (GEI) observed in this study could seriously reduce the realized gains for biomass quality of
sorghum. Although few studies have been specifically designed to examine genotype by year interactions
for stover quality, there are reports of highly significant genotype by year interactions for in vitro
digestibility (Aruna et al. 2012) or NDF (Jung et al., 1998) while other studies failed to find major GxE
interactions for stover digestibility (Badve et al., 1994). The presence of significant genotype x
environment (G x E) interactions for yield and quality traits suggests that evaluation in more than one
environment may be required for accurate selection for biomass yield and quality as was reported in
maize (Lorenz et al. 2009).
Associations amongst the fodder yield and quality traits, their interaction with the environment will help
in guiding future research strategies. In general nonsignificant associations were recorded between fodder
yield and all important fodder quality traits, such as IVOMD, CP, ADF, NDF and ADL indicating that
these traits have independent inheritance. This paves way for simultaneous genetic improvement of both
fodder yield and quality. Significant positive association of fodder yield was observed with PH, NLP, SD,
DTF, LL and LW, which was supported by partial least square analysis showing that all these traits,
except for LW contributed for variability in fodder yield. Strong positive genotypic and phenotypic
correlations between fodder yield and stem diameter, fresh weight per plant, leaf length, plant height and
number of leaves was reported earlier (Iyanar 2010; Tariq et al. 2012). IVOMD was observed to have
positive association with PH, NLP and SD among the yield traits which is significant at 5% level of
significance. Partial least square analysis also indicated that for improving both fodder yield and
digestibility the important traits to be addressed are PH, ADF, ADL, NDF, NLP, SD, DTF and LL. The
leaf component is important for both yield and quality not only because of the generally high nutritive
value in leaves compared to stems, but leaves are more acceptable to animals as they are easier to chew
and more digestible (Reddy et al. 2003). Brix was observed to have significant positive association with
IVOMD indicating the suitability of sweet sorghum varieties for forage purpose.
The GCA and SCA variances were significant for all traits. Since genotypic variance was significant,
varietal improvement could contribute to raising the nutritional quality of sorghum fodder above current
levels. Presence of highly significant GCA and SCA variances for most of the characters indicated the
importance of both additive and non-additive genes in the expression of the traits. In the present study the
ratio of GCA/SCA variance was less than unity for all the characters except leaf breadth, which indicated
the pre-ponderance of nonadditive genetic variance. Predominance of non-additive gene action for fodder
yield and yield components in forage sorghum was reported earlier (Aruna et al. 2012; Prakash et al.
2010). For the traits where, both additive and non-additive gene effects were important, dominance
variance (σ²D) was found to be larger than the additive variance (σ²A), showing the importance of nonadditive gene effects in the control of these traits indicating good prospects for the exploitation of nonadditive genetic variation for fodder yield and quality traits in forage sorghum through hybrid breeding.
Although there was a pre-ponderance of non-additive gene action for most of the characters, the presence
of a considerable amount of additive gene action could not be totally neglected. Since the non-additive
gene effects were predominant, direct selection for fodder yield and quality may not be effective.
Population breeding, where there is a chance to accumulate genes from different genotypes, can be one of
the approaches to improve the level of yield. Predominance of nonadditive gene effects for brix was
reported earlier (Audilakshmi et al. 2010).
The GCA effect is considered as the intrinsic genetic value of the parent for a trait, which is due to
additive gene effects and it is fixable (Simmonds 1979). To get outstanding recombinants in segregating
generations, the parents of the hybrids must be good general combiners for the characters to which
improvement is sought (Gravois and McNew 1993, Manonmani and Fazullah Khan 2003). Hence, the
potentiality of parents to produce better offspring with a reservoir of superior genes was evaluated based
on their GCA effects.
The presence of heritable variation for both fodder yield and quality traits and their independent nature of
association prompts that simultaneous improvement of fodder yield and quality is possible, but it depends
on the parents selected in the breeding programme. In the current study, cluster analysis indicated that
cluster II had a high inter-cluster distance from the other two clusters. Cultivars from different clusters, if
chosen for hybridization programme, may give a broad spectrum of variability in segregating generations.
In the present study, some genotypes in different clusters with good general combining ability for fodder
yield and quality were identified. The genotypes HC308 and SEVS4 from cluster I were the best
combiners for most of the fodder yield parameters such as plant height, leaf paramaters, stem girth etc.
and for some of the fodder quality traits such as high brix, low lignin in HC308 and low CLL and HCL in
SEVS4. Both these lines were good combiners for low cellulose which is better for animal feed. The
brown midrib genotypes, EC582508 and EC582510 were good combiners for earliness, IVOMD and low
lignin content. These lines can be used as a source for improving the fodder quality in terms of
digestibility. The genotype, Keller from cluster II was a good combiner for earliness, leaf parameters,
stem girth and fodder quality traits such as high brix, IVOMD, low cellulose, low hemicelluloses and low
lignin. Nizamabad forage from the same group was a good combiner for CP and earliness. These can be
crossed with HC308 and SEVS4 of cluster I for improvement of forage sorghum for animal feed. PC23 of
cluster II and UPMC512 of cluster III had high positive combining ability for cellulose and
hemicelluloses and hence can be utilized for improvement of genotypes for biofuel production. Breeding
programmes can be designed to utilize these lines for improving biomass/fodder yield and quality, and
multiple crosses involving these parents would result in identification of superior segregants with
favourable genes for most of the traits associated with fodder yield and quality. Biparental mating in early
segregating generations of the crosses involving these parents for simultaneous exploitation of both
additive and non-additive gene action can be recommended to develop sorghum genotypes with improved
fodder yield and quality. It is suggested that intermating of the randomly selected progenies in early
segregating generations (especially in F2 and F3) obtained by crossing these parents will release the
hidden genetic variability through breakage of undesirable linkages involved in different characters. It
may produce an elite population for selection of lines with high fodder yield and quality in advanced
generations.
The most promising specific combiners for fodder yield and other characters were SSG59-3 x UPMC512,
UPMC512 x SEVS4, PC23 x Keller, PC23 x EC582508, HC308 x EC582510, EC582510 x Nizamabad
forage, EC582510 x SEVS4 and Nizamabad forage x SEVS4 (Table 16). These hybrids involved parents
with high/low, average/low and low/low GCA effects for yield indicating additive x dominance and
dominance x dominance types of gene interactions for expression of the traits. The superiority of these
crosses may be due to complementary and duplicate gene actions (Griffings 1956; Falconer 1989). When
additive x additive gene interactions are present in the control of a trait, simple selfing in the early
generations will give desired genotype. If additive x dominance and dominance x dominance gene
interactions are present, hybrid breeding will be useful, as in hybrids heterozygous condition is fixed.
Epistatic interactions were found to play a major role in the genetic basis of fibre related traits (Shiringani
and Friedt 2011). According to Griffing (1956), the hybrids with high specific combining ability effects,
and involving at least one parent with high or average GCA effects for a particular trait is a good strategy
for plant breeding.
It is interesting to note that majority of hybrids shown by this study as being the best in SCA effects for
some traits, were also among the best performing for the same traits as shown by their heterosis (Table
16); hence, utilizing heterosis in improving such traits might be rewarding. Hybrids low in lignin and
fibre components appears to be attainable without sacrificing high yield levels. It was concluded that
exploiting heterosis in forage sorghum to improve quality traits might be promising.
The main conclusions from this study are that both additive and non-additive gene effects are important,
with a predominance of non-additive gene effects in governing fodder yield and quality in sorghum.
Multiple crosses involving the best combiners for different traits would result in obtaining superior
segregants with favourable genes for most of the traits associated with fodder yield and quality. The study
also indicates the brown midrib genotypes can be used to develop cultivars with low lignin and high
digestibility which are suitable for both animal and biofuel industry. The potential for significant genetic
improvement of C4 grasses as biofuel crops is good. Better understanding of the genes involved in cell
wall biosynthesis, and the precise role of these gene families will ultimately help in targeted modification
of cell wall architecture in such a way that processing characteristics are optimized without compromising
plant performance in the field. This confirms that there is a great opportunity to improve both fodder yield
and quality in breeding programs aiming at genetic enhancement of forage sorghums. Since many traits
contribute for fodder yield and quality, population breeding or marker assisted selection would be fruitful
in forage sorghum improvement. However, population breeding is time consuming. Identifying markers
for the component traits associated with yield and quality and pyramiding them into elite cultivars would
help in developing forage sorghum cultivars with improved quality. The improvement in the quantity and
the nutritional quality of the fodder of forage sorghum cultivars could make a significant impact on
livestock productivity in the sorghum growing areas.
References
1. Aruna C, Shrotria PK, Pahuja SK, Umakanth AV, Venkatesh Bhat B, Vishala Devender A, Patil
JV (2012) Fodder yield and quality in forage sorghum: scope for improvement through diverse
male
sterile
cytoplasms.
Crop
&
Pasture
Science
63,
1114–1123.
http://dx.doi.org/10.1071/CP12215
2. Atienza SG, Satovic Z, Petersen KK, Dolstra O, Martin A (2003) Identification of QTLs
influencing agronomic traits in Miscanthus sinenesis Anderss. I. Total height, flag-leaf height and
stem diameter. Theoretical and applied genetics 107, 123-129. Doi: 10.1007/s00122-003-1220-5
3. Audilakshmi S, Mall AK, Swarnalatha M, Seetharama N (2010) Inheritance of sugar
concentration in stalk brix), sucrose content, stalk and juice yield in sorghum. Biomaas and
bioenergy 34, 813-820. http://dx.doi.org/10.1016/j.biombioe.2010.01.025
4. Barriere Y, Ralph J, Mechin V, Guillaumie S, Grabber JH, Argillier O, Chabber B, Lapierre C
(2004) Genetic and molecular basis of grass cell wall biosynthesis and degradability. II. Lessons
from brown midrib mutants. C R Biol 327, 847-860. http://dx.doi.org/10.1016/j.crvi.2004.05.010
5. Barriere Y, Thomas J, Denoue D (2008) QTL mapping for lignin content, lignin monomeric
composition, p-hydroxycinnamate content, and cell wall digestibility in the maize recombinant
inbred
line
progeny
F838
x
F286.
Plant
Sci
175,
585-595.
http://dx.doi.org/10.1016/j.plantsci.2008.06.009
6. Blummel M, Ørskov ER (1993) Comparison of in vitro gas production and nylon bag
degradability of roughages in predicting feed intake in cattle. Animal Feed Science and
Technology 40, 109-119.
7. Badve VC, Nisal PR, Joshi AL, Rangnekar DV (1994) Genotype and environment effects on
sorghum stover production and quality. (Eds. Joshi AL, Doyle PT, Oosting SJ), Variation in the
quantity and quality of fibrous crop residues. Proceedings of the National Seminar held at the
BAIF Development Research Foundation, Pune, Maharashtra, India, February 8-9, 1994. IndoDutch Project on Bioconversion of Crop Residues, pp. 9-19.
8. Carpita MC, McCann MC (2008) Maize and sorghum: genetic resources for bioenergy grasses.
Trends Plant Sci 13, 415-420. http://dx.doi.org/10.1016/j.tplants.2008.06.002
9. Dien BS, Sarath G, Pedersen JF, Sattler SE, Chen H, Funnell-Harris D, Nichols NN, Cotta MA
(2009) Improved sugar conversion and ethanol yield for forage sorghum (Sorghum bicolour L.
Moench) lines with reduced lignin contents. Bioenerg. Res. 2, 153-164. Doi: 10.1007/s12155009-9041-2
10. Falconer D (1989) "Introduction to quantitative genetics. Long man, Scientific and Technical
Group Ltd, 3rd edn. Wiley, New York."
11. Goering HK, Van Soest PJ (1970) Forage Fiber analyses: Apparatus, Reagents, Procedures and
some applications, Agricultural Handbook No. 379, USDA-ARS, Washington, DC.
12. Gravois KA, McNew WR (1993) Genetic relationships and selection for rice yield and yield
components. Crop Science 33, 249-52.
13. Griffings B (1956) "Concept of general and specific combining ability in relation to diallel
crossing systems." Aust J Biol Sci 9, 463-493.
14. Hall A, Blummel M, Thorpe W, Bidinger FR, Hash CT (2004) Sorghum and Pearl millet as foodfeed crops in India. Animal Nutrition and Feed Technology 4, 1-15.
15. Indostat Servies (2004) Windostat, Indostat Services, Hyderabad, India.
16. Iyanar K, Vijayakumar G, Fazllullah Khan AK (2010) Correlation and path analysis in multicut
fodder sorghum. Electronic journal of Plant Breeding 1, 1006-1009.
17. Jung HG, Mertens DR, Buxton DR (1998) Forage quality variation among maize inbreds: in vitro
fiber digestion kinetics and prediction with NIRS. Crop Sci. 38, 205-210.
18. Jurs PC (1990) Chemometrics and multivariate analysis in analytical chemistry. In Reviews in
Computational Chemistry 1 (Eds KB Lipkowitz and DB Boyd), pp. 169-212. NewYork: VCH
Publishers.
19. Lorenz AJ, Coors JG, de Leon N, Wolfrum EJ, Hames BR, Sluiter AD, Weimer PJ (2009)
Characterization, genetic variation, and combining ability of maize traits relevant to the
production of cellulosic ethanol. Crop Sci. 49: 85-98. doi:10.2135/cropsci2008.06.0306
20. Mahalanobis PC (1930) On the test and measure of group divergence. Proceedings of Asiatic
Society of Bengal 26, 541-548.
21. Manonmani S, Fazlullah Khan SK (2003) Analysis of genetic diversity for selection of parents in
rice. Oryza 40, 54-56.
22. Mahmood A, Honermeier B (2012) Chemical composition and methane yield of sorghum
cultivars
with
contrasting
row
http://dx.doi.org/10.1016/j.fcr.2011.12.010
spacing.
Field
Crops
Res.
128,
27-33.
23. Menke KH, Steingass H (1988) Estimation of the energy feed value obtained from chemical
analysis and in vitro gas production using rumen fluid. Animal Research and Development 28, 755.
24. Murray SC, Sharma A, Rooney WL, Klein PE, Mullet JE, Mitchel SE, Kresovich S (2008)
Genetic improvement of sorghum as a biofuel feedstock II: Quantitative loci for stem and leaf
structural carbohydrates. Crop Sci. 48, 2180-2193. doi:10.2135/cropsci2008.01.0068
25. Prakash R, Ganesamurthy K, Nirmalakumari A, Nagarajan P (2010) Combinng ability for fodder
yield and its components in Sorghum (Sorghum bicolor L.). Electronic Journal of plant breeding
1, 124-128.
26. Rao CR (1952) Advanced statistical methods in biometrical research. John Wiley and sons Inc.,
New York.
27. Rao PS, Blummel M, Reddy BVS (2012) Enhancement of in vitro digestibility of sorghum
(Sorghum bicolor (L) Moench) in brown midrib (bmr) mutant derivatives of bmr1 and bmr7.
European J Plant Sci and Biotechnology 6, 76-80.
28. Reddy BVS, Reddy PS, Bidinger F, Blummel M (2003) Crop management factors influencing
yield
and
quality
of
crop
residues.
Field
Crops
Research
84,
57-77.
http://dx.doi.org/10.1016/S0378-4290(03)00141-2
29. Sarath G, Mitchell RB, Sattler SE, Funnell D, Pedersen JF, Graybosch RA, Vogel KP (2008)
Opportunities and roadblocks in utilizing forages and small grains for liquid fuels. J Ind
Microbiol Biotechnol 35, 343-354. Doi: 10.1007/s10295-007-0296-3
30. Sattler SE, Funnell-Harris DL, Pedersen JF (2010) Brown midrib muatations and their importance
to the utilization of maize, sorghum and pearl millet lignocellulosic tissues. Plant Science 178,
229-238. http://dx.doi.org/10.1016/j.plantsci.2010.01.001
31. Shiringani AL, Friedt W (2011) QTL for fibre related traits in grain x sweet sorghum as a tool for
the enhanement of sorghum as a biomass crop. Theor Appl Genet 123, 999-1011. Doi:
10.1007/s00122-011-1642-4
32. Simmonds NW (1979) Principles of crop improvement. Long Man Group Ltd., London, pp. 110116.
33. Tariq AS, Akram Z, Shabbir G, Gulfraz M, Saifullah Khan K, Iqbal MS, Mahmood T (2012)
Character association and inheritance studies of different sorghum genotypes for fodder yield and
quality under irrigated and rainfed conditions. African J Biotechnology 11, 9189-9195.
34. Vermerris W, Saballos A, Ejeta G, Mosier NS, Ladisch MR, Carpita NC (2007) Molecular
breeding to enhance ethanol production from corn and sorghum stover. Crop Sci 47, 142-153.
doi:10.2135/cropsci2007.04.0013IPBS
35. Vermerris W (2011) Survey of genomic approaches to improve bioenergy traits in maize,
sorghum and sugarcane. J Integr. Plant Biol. 53, 105-119.
36. Vogel KP, Sleper DA (1994) Alteration of plants via genetics and breeding. In: Forage quality,
evaluation and utilization ( Eds. George J, Fahey C). American Society of Agronomy, Madison,
WI, pp. 891-921
37. Wold S (1994) PLS for multivariate linear modelling. In QSAR: Chemometric methods in
molecular design: Methods and Principles in Medicinal Chemistry (Ed. H. Van de Waterbeemd),
pp. 195-281. Weinheim, Germany: Verlag-Chemie.
38. Xie HL, Ji HQ, Liu ZH, Tian GW, Wang CL, Hu YM, Tang JH (2009) Genetic basis of
nutritional content of stover in maize under low nitrogen conditions. Euphytica 165, 485-493.
Doi: 10.1007/s10681-008-9764-8
39. Zhang Y, Kang M (2005) "DIALLEL-SAS05: a comprehensive program for Griffing's and
Gardner-Eberhart analyses." Agronomy Journal 197, 1097-1106.
40. Zhao YL, Dolat A, Steinberger Y, Wang X, Osman A, Xie GH (2009) Biomass yield and changes
in chemical composition of sweet sorghum cultivars grown for biofuel. Field Crops Res 111, 5564. http://dx.doi.org/10.1016/j.fcr.2008.10.006
Table 1. Meteorological data during the crop period for two years (2009 & 2010)
Month and
Year
2009
June
July
August
2010
June
July
August
Temperature
Minimum
Maximum
Relative
humidity %
I
II
Rainfall
(mm)
36.3
32.0
31.2
24.8
23.4
23.3
72
80
81
41
57
64
82.0
154.0
203.7
35.2
29.4
30.3
24.7
22.5
22.6
82
89
91
60
77
75
113.7
278.9
203.1
Table 2. Description of the parents used in the study
Lines
SSG59-3
Origin
HAU, Hisar, India
Characteristics
Popular multi-cut forage sorghum variety released under
All India Co-ordinated Sorghum Improvement Program
(AICSIP)
UPMC503
GBPUAT, Pantnagar, Male parent of the popular forage sorghum hybrid, CSH
India
UPMC512
20MF
GBPUAT, Pantnagar, Improved forage sorghum line from Pantnagar
India
PC23
HC308
IARI, New Delhi,
Forage sorghum variety from Indian Agricultural Research
India
Institute (IARI), New Delhi, India
HAU, Hisar, India
Popular forage sorghum variety released under All India
Co-ordinated Sorghum Improvement Program (AICSIP)
Keller
USA
Sweet sorghum variety
EC582510
N598 from Nebraska Brown midrib line
University, USA
EC582508
Atlas bmr-12 from Brown midrib line
Nebraska University,
USA
Nizamabad
Nizamabad, India
forage
SEVS4
Forage sorghum variety from Nizamabad area of Andhra
Pradesh, India
AICSIP, India
Dual purpose sorghum variety under All India Coordinated Sorghum Improvement Project (AICSIP)
Table 3. Overall means, ranges in individual cultivar means, and entry mean heretabilities for
sorghum fodder quality traits and fodder yield. Data are from a 55- entry sorghum trial grown in
two environments (**p <0.001)
Variable
FY (g/pl)
CP (%)
NDF
ADF
ADL
IVOMD
Brix
DF
PH
LL
LB
LN
SD
Mean
1860
9.62
64.42
40.35
4.87
50.76
12.07
66
294
76
5.9
11
1.34
Range in the cultivars
610-2743**
7.43-11.7**
58.03-73.1**
36.01-46.18**
3.59-5.68**
45.73-55.38**
6.63-16.77**
59-76.3**
165-355**
61-86**
4.6-7.3**
9-13**
0.94-1.60**
lsd
513.6
1.6
3.5
2.7
0.47
2.5
1.9
3.1
21.3
4.5
0.6
1.2
0.16
Heritability
0.54
0.36
0.48
0.43
0.47
0.39
0.59
0.83
0.82
0.70
0.60
0.50
0.47
DTF- Days to flower; FY- Fodder yield; PH- Plant height; NLP- Number of leaves per plant; LL- Leaf length;
LB- Leaf breadth; SD- Stem diameter; CP- Crude Protein; IVOMD- In vitro organic matter digestibility;
NDF- Neutral Detergent Fibre; ADF- Acid Detergent Fibre; ADL- Acid Digestible Lignin
Table 4: G x E effects of forage dry matter yield and forage fodder quality traits
FY
CP
NDF
ADF
Cultivar
1470721** 4.12**
Year
3008728** 791.4* 10996*
Cultivar x
378957*
2.84*
43.5**
ADL
IVOMD
Brix
17.3**
20.0**
19.96** 0.76**
20.1**
2071*
53.4**
3183*
11.1
11.2**
0.33**
8.03**
8.78**
Year
DTF
PH
146.4**
9190**
Year
683.7
64233*
Cultivar x Year
52.9**
724.2**
Cultivar
NLP
LL
LB
SD
6.55** 174.5**
2.34**
0.11**
29.8
5736.4*
151.4*
0.24
2.01*
60.3**
0.88**
0.04**
DTF- Days to flower; FY- Fodder yield; PH- Plant height; NLP- Number of leaves per plant; LL- Leaf length;
LB- Leaf breadth; SD- Stem diameter; CP- Crude Protein; IVOMD- In vitro organic matter digestibility;
NDF- Neutral Detergent Fibre; ADF- Acid Detergent Fibre; ADL- Acid Digestible Lignin
Table 5. PLS regression matrix
No.
of R2
for
latent
model (X)
variables
2
58.1
R2
for Root
dependent
PRESS
variable
73.0
0.57
2
59.8
73.8
0.47
Fodder
3
yield and
IVOMD
67.6
67.8
0.60
Dependent
variable
Fodder
yield
IVOMD
Equation prediction variable and coefficients
Intercept
DF (7.32); PH (0.97); NL (28.7); LL (2.25); LW
(17.7): SD (93.3); Brix (6.88); CP (-9.23); NDF
(1.49); ADF (1.91); ADL (11.2); IVOMD (-4.71)
DF (-2.51); PH (-0.01); NL (-0.06); LL (-0.02); LW
(0.095): SD (0.55); Brix (0.071); CP (0.003); NDF (0.06); ADF (-0.11); ADL (-0.76)
FY: DF (9.37); PH (1.27); NL (28.7); LL (1.17); LW
(16.8): SD (1.37); Brix (6.61); CP (-37.7); NDF
(0.49); ADF (2.20); ADL (14.1)
IVOMD: DF (-0.01); PH (-0.01); NL (-0.099); LL (0.004); LW (0.12): SD (1.3); Brix (0.06); CP (0.15);
NDF (-0.09); ADF (-0.16); ADL (-0.95)
-1149.5
53.3
-1159.2
66.5
Table 6. Variable importance for projecting scores of the predictors
Predictor
PH
FY
1.45
IVOMD
0.86
FY
+ 1.33
IVOMD
NLP
1.41
0.32
1.04
SD
1.16
0.71
0.95
DTF
1.15
0.46
0.85
LL
1.00
0.45
0.84
LW
0.79
0.57
0.64
ADF
0.61
1.57
1.26
Brix
0.56
0.65
0.60
NDF
0.54
1.41
1.14
ADL
0.43
1.54
1.26
CP
0.28
0.16
0.59
DTF- Days to flower; FY- Fodder yield; PH- Plant height; NLP- Number of leaves per plant; LL- Leaf length; LB- Leaf breadth; SD- Stem diameter;
CP- Crude Protein; IVOMD- In vitro organic matter digestibility; NDF- Neutral Detergent Fibre; ADF- Acid Detergent Fibre; ADL- Acid Digestible
Lignin
Table 7. Analysis of variance for combining ability for fodder yield traits in forage sorghum involving 10 x 10 half diallel analysis
DF
Environments 1
Genotypes
54
Gen * Env
54
Error
216
GCA
9
SCA
45
GCA*Env
9
SCA*Env
45
Error
216
∂2G
∂2 S
∂2a
∂2 D
GCA/SCA
Ratio
h² (b)
DTF
683.7**
146.4**
52.9 **
7.17
175.4 **
23.5 **
51.2 **
10.9 **
2.39
7.21
10.55
14.42
10.55
FY
30087.3 **
14707.2 **
3789.6 **
2037.3
16702.6 **
2542.4 **
1905.4 **
1134.7 **
679.1
667.65
931.6
1335.3
931.6
PH
64279.3 **
9189.1 **
724.0 **
350.5
8322.0 **
2011.2 **
196.2
250.4 **
116.8
341.9
947.2
683.7
947.2
NLP
29.52 **
6.62 **
2.01 **
1.17
6.39 **
1.37 **
0.7
0.66 **
0.39
0.25
0.49
0.5
0.49
LL
5743.6 **
174.4 **
60.12**
15.7
131.0 **
43.55 **
27.40 **
18.57 **
5.23
5.24
19.16
10.48
19.16
LB
15175.5 **
233.5 **
88.02 **
27.96
312.5 **
30.90 **
30.45 **
29.12 **
9.32
12.63
10.79
25.27
10.79
SD
24.4 **
10.95 **
3.90**
2.03
8.26 **
2.73 **
2.63 **
1.03 *
0.68
0.32
1.02
0.63
1.02
0.684
0.717
0.361
0.51
0.274
1.171
0.309
0.83
0.54
0.82
0.50
0.70
0.60
0.47
*p<0.05, **p<0.01
DTF- Days to flower; FY- Fodder yield; PH- Plant height; NLP- Number of leaves per plant; LL- Leaf length; LB- Leaf breadth; SD- Stem diameter
1
Table 8. Analysis of variance and combining abilities for fodder quality traits in forage sorghum involving
2
10 x 10 half diallel analysis
Brix
11.7*
20.3**
8.88**
2.82
13.9**
5.33**
8.68**
1.82**
0.94
0.54
2.19
1.08
2.19
3
4
5
6
CP
792.0**
4.12**
2.84**
1.57
2.53**
1.14**
2.16**
0.70
0.52
0.08
0.31
0.17
0.31
IVOMD
3183.6**
17.32**
8.02**
4.22
20.04**
2.92**
3.35*
2.54**
1.4
0.78
0.76
1.55
0.76
NDF
10996.3**
43.51**
24.05**
9.44
56.2**
6.16**
14.98**
6.62**
3.15
2.21
1.51
4.42
1.51
ADF
2071.2**
19.97**
11.2**
5.25
21.75**
3.64**
7.59**
2.97**
1.75
0.83
0.94
1.67
0.94
CLL
1459.8**
14.18**
8.26**
3.81
15.98**
2.48**
5.40**
2.22**
1.27
0.61
0.6
1.23
0.6
HCL
3518.8**
8.02**
5.17**
2.09
10.28**
1.15**
3.054**
1.46**
0.7
0.4
0.23
0.8
0.23
Environments
Genotypes
Gen * Env
Error
GCA
SCA
GCA*Env
SCA*Env
Error
∂2G
∂2 S
∂2a
∂2 D
GCA/SCA
0.246
0.27
1.026
1.468
0.882
1.017
1.756
Ratio
h² (b)
0.59
0.36
0.39
0.48
0.43
0.41
0.41
*p<0.05, **p<0.01
CP- Crude Protein; IVOMD- In vitro organic matter digestibility; NDF- Neutral Detergent Fibre; ADF- Acid
Detergent Fibre; CLL- Cellulose; HCL- Hemicellulose; ADL- Acid Digestible Lignin
ADL
5338.6 **
75.7 **
32.8 **
16.66
66.9 **
16.9 **
23.4 **
8.44 *
5.55
2.56
5.67
5.11
5.67
0.451
0.47
7
8
9
10
11
Table 9. General combining ability effects for fodder yield traits in forage sorghum parents from 10 x 10
12
half diallel analysis
SSG 59-3
UPMC 503
UPMC 512
PC 23
DTF
FY
PH
NLP
LL
LB
SD
-1.12**
2.04 **
-0.57
1.25 **
-16.97 **
10.29 *
-19.96 **
-36.46 **
5.47 **
16.20 **
-15.76 **
18.76 **
-0.46 **
0.051
-0.45 **
0.11
-0.53
2.69 **
-0.39
-1.01 *
-6.94 **
-1.84 **
1.31 *
-3.73 **
-1.05 **
0.12
-0.38 *
-0.61 **
30
HC 308
Keller
EC582510
Nizamabad
forage
EC582508
SEVS 4
SE (gi)
SE (gi-gj)
13
14
15
16
5.42 **
-2.56 **
-1.82 **
30.66 **
-4.26
-8.55
15.48 **
-20.26 **
-19.31 **
0.99 **
-0.30 *
-0.21
2.66 **
1.45 **
-3.74 **
2.19 **
2.41 **
0.9
0.47 **
0.52 **
0.21
-2.19 **
-0.93
-0.6
-0.30 *
-0.22
-1.38 *
-0.29
-2.90 **
2.46 **
0.30
0.45
-8.89
55.06 **
5.05
7.52
-24.52 **
24.55 **
2.09
3.12
-0.26 *
0.82 **
0.12
0.18
-3.35 **
2.44 **
0.44
0.66
1.12
5.95 **
0.59
0.88
0.08
0.92 **
0.16
0.24
*p<0.05, **p<0.01
DTF- Days to flower; FY- Fodder yield; PH- Plant height; NLP- Number of leaves per plant; LL- Leaf length;
LB- Leaf breadth; SD- Stem diameter.
17
18
19
20
21
22
Table 10. General combining ability effects for fodder quality in forage sorghum parents from 10 x 10
23
half diallel analysis
SSG 59-3
UPMC 503
UPMC 512
PC 23
HC 308
Keller
EC582510
Nizamabad
forage
EC582508
SEVS 4
SE (gi)
SE (gi-gj)
24
Brix
CP
IVOMD
NDF
ADF
CLL
HCL
ADL
-0.46 *
0.24
-0.44
-0.12
0.19
0.05
-0.3
1.42 **
0.44 *
-0.58 **
-0.3
-0.18
-0.09
-0.1
-0.1
0.1
-1.10 **
0.06
-0.99 **
1.70 **
1.15 **
1.02 **
0.55 **
1.28 **
0.25
-0.11
-1.74 **
3.54 **
2.06 **
1.76 **
1.47 **
3.00 **
0.93 **
-0.35
0.43
-0.38
-0.62 *
-0.50 *
0.23
-1.17 *
1.39 **
-0.04
1.07 **
-1.54 **
-1.11 **
-1.03 **
-0.42 **
-0.83
-0.23
-0.14
0.51 *
-0.11
-0.18
-0.05
0.07
-1.29 **
-0.66 **
0.59**
-0.13
-0.68 *
-0.08
-0.17
-0.60 **
0.83
-0.50 **
0.17
1.21 **
-0.79 *
-0.71 **
-0.44 *
-0.07
-2.76 **
-0.05
0.15
0.37
-1.43 **
-0.60 *
-0.54 *
-0.83 **
-0.59
0.19
0.48
0.23
0.34
0.26
0.22
0.16
0.46
0.28
0.71
0.34
0.51
0.38
0.32
0.24
0.68
*p<0.05, **p<0.01
31
25
26
27
CP- Crude Protein; IVOMD- In vitro organic matter digestibility; NDF- Neutral Detergent Fibre; ADF- Acid
Detergent Fibre; CLL- Cellulose; HCL- Hemicellulose; ADL- Acid Digestible Lignin
28
29
30
31
32
33
Table 11. Mean performance of the parents for important fodder yield and quality traits over two years
SSG 59-3
UPMC 503
UPMC 512
PC 23
HC 308
KELLER
EC582510
Niz forage
EC 582508
SEVS 4
Mean
C.V.
C.D. 5%
C.D. 1%
DTF
67.17
68.17
65.33
60.00
76.33
63.00
65.17
61.67
62.50
72.83
66.22
3.90
4.33
5.93
Plant
ht (cm)
278.00
289.37
165.00
258.77
303.00
224.37
204.90
280.63
211.63
305.67
252.13
4.59
19.34
26.50
FY
(g/pl)
261.67
326.33
240.00
122.00
443.33
270.33
207.33
335.33
250.00
450.67
290.70
23.09
125.3
171.6
Brix
10.72
12.23
9.73
8.23
12.37
16.77
14.23
10.23
10.33
12.43
11.73
18.29
3.74
5.13
CP% ivomd%
10.79
49.68
7.90
50.47
11.60
50.02
11.42
45.73
10.06
50.65
10.05
53.72
9.10
54.24
12.42
49.52
11.92
55.38
10.27
52.06
10.55
51.15
9.66
4.25
1.75
3.50
2.40
4.80
ADL Cellulose Hemicellu
4.96
35.94
23.41
4.92
35.50
23.78
4.94
36.32
23.92
5.68
40.49
26.93
4.77
35.62
24.04
4.47
31.54
22.03
3.69
33.23
24.55
5.11
36.08
23.45
3.59
33.99
25.79
4.35
32.85
22.33
4.65
35.16
24.02
5.85
3.63
3.82
0.50
2.31
1.78
0.69
3.17
2.44
34
35
36
37
38
39
32
40
Table 12. Specific combining ability of selected hybrids for fodder yield traits involving 10 x 10 half diallel
41
analysis
Pedigree
P1 x P3
P1 x P9
P1 x P10
P2 x P3
P2 x P9
P3 x P4
P3 x P5
P3 x P7
P3 x P9
P3 x P10
P4 x P6
P4 x P7
P4 x P8
P4 x P9
P5 x P7
P5 x P9
P7 x P8
P7 x P10
P9 x P10
SE (sij-sik)
SE (sij – skl)
DTF
FY
PH
NLP
LL
LB
SD
7.56 **
-2.27 *
-4.14 **
-0.94
3.72 **
5.68 **
3.02 **
-4.25 **
-3.99 **
-2.52 *
4.00 **
2.93 **
3.97 **
3.68 **
-0.73
0.68
0.38
2.22 *
-3.52 **
1.48
1.41
41.26 *
6.19
-7.42
-15.34
60.9 **
20.4
18.3
-10.15
-13.3
53.2 **
34.0 *
21.7
25.7
84.7 **
40.2 *
5.23
42.5 *
35.8 *
-1.67
24.9
23.8
20.1 **
25.7**
3.56
23.6 **
27.57 **
34.54 **
34.82 **
33.6 **
8.27
14.06 *
36.9 **
26.9 **
10.72
59.3 **
20.21 **
25.08 **
12.36
27.1 **
-0.51
10.35
9.87
0.83 *
0.51
-0.21
-0.55
0.69
1.09 **
0.34
0.14
-0.4
-0.22
1.07 **
0.42
0.81 *
1.17 **
1.13 **
0.52
0.96 *
0.81 *
0.39
0.6
0.57
3.32 *
1.48
-0.61
1.6
5.13 **
1.67
3.27 *
-2.79
1.71
-0.85
5.86 **
6.82 **
2.69
3.53 *
3.75 *
5.99 **
2.2
10.04 **
-2.99 *
2.19
2.09
0.52
0.21
-4.35 *
1.89
-0.02
2.98
0.99
-1.35
3.92 *
1.96
-1.25
-0.37
0.64
2.5
2.91
1.15
5.04 *
7.67 **
-0.95
2.92
2.79
1.15 *
0.09
-0.32
-0.48
1.73 **
0.88
1.21 *
-0.17
-0.65
1.01
1.08 *
1.05
0.62
2.78 **
0.65
0.14
0.94
1.39 **
0.08
0.79
0.75
42
43
44
45
46
*p<0.05, **p<0.01
DTF- Days to flower; FY- Fodder yield; PH- Plant height; NLP- Number of leaves per plant; LL- Leaf length;
LB- Leaf breadth; SD- Stem diameter
P1- SSG59-3; P2- UPMC503; P3- UPMC512; P4- PC23; P5- HC308; P6- Keller; P7- EC582510; P8-
47
Nizamabad forage; P9- EC582508; P10- SEVS4
48
49
Table 13. Specific combining ability of selected hybrids for fodder quality traits involving 10 x 10 half
50
diallel analysis
Pedigree
P1 x P6
P1 x P7
P2 x P3
Brix
0.58
-0.94
2.33 **
CP
IVOMD
NDF
ADF
CLL
HCL
ADL
-0.35 1.69 *
-3.18 **
-1.85 * -1.58 *
-1.35 **
-2.72 *
0.87 1.43*
0.77
1.28
0.97
-0.51
-2.98*
-0.52
0.46
-1.03
-1.24
-1
0.21
-2.38
33
P2 x P5
P2 x P9
P2 x P10
P3 x P4
P3 x P5
P3 x P6
P3 x P7
P3 x P8
P3 x P9
P4 x P5
P4 x P6
P4 x P8
P4 x P9
P4 x P10
P5 x P7
P5 x P9
P6 x P7
P7 x P8
SE (sij-sik)
SE (sij – skl)
0.43
0.042 1.40 *
-1.67
-0.99
-0.73
-0.68
-2.56
-0.46
-0.22
0.89
-1.2
-0.41
-0.46
-0.78
0.39
-0.49
0.09
-1.22
2.10*
1.37
1.09
0.73
2.81*
1.22
-0.05 1.66 *
-0.39
-0.94
-0.81
0.54
-1.21
1.87 **
-0.04
0.57
0.047
-0.72
-0.65
0.77
-0.72
-5.73 **
0.59
-2.12 3.81 **
2.58 ** 2.13 **
1.22 *
4.52 **
0.077
-0.05
-1.83
1.72 1.74 *
1.52 *
-0.013
2.13
1.79 **
-0.91
-0.76
1.019
0.91
0.85
0.11
0.51
-0.99 -0.003
1.61*
1.24
1.46
1.15 -0.22
-3.2*
-0.49
-0.41
1.41*
-2.08*
-1.4
-1.15 -0.69
-2.49
1.73 **
0.12
-0.31 2.97 **
1.47 1.29 *
1.50 **
1.8
2.14 **
-0.05
1.67
-1.84 -1.70 *
-1.44 *
-0.14
-2.63
2.48 **
-1.13
-1.14
-0.95
0.02
0.008
-0.97
0.12
1.66 **
-0.39
0.73
-0.64
-0.64
-0.34
-0.001 -2.95 *
-1.53*
0.61
-2.68 ** 3.41 **
2.39 ** 1.82 **
1.02 *
5.73 **
1.35 *
-0.69
-0.18
-1.07
-1.2
-1.15
0.13
-0.46
-1.23 1.25*
-0.52 2.89 **
1.83 *
1.65 *
1.06 *
1.82
-1.54 *
0.06
-0.14 -2.26 *
-1
-1.15 -1.26 *
1.42
0.93
0.69
1.13 1.70
1.27
1.08 0.80
2.26
0.88
0.66
1.08 1.62
1.21
1.03 0.76
2.15
51
52
53
54
55
*p<0.05, **p<0.01
CP- Crude Protein; IVOMD- In vitro organic matter digestibility; NDF- Neutral Detergent Fibre; ADF- Acid
Detergent Fibre; CLL- Cellulose; HCL- Hemicellulose; ADL- Acid Digestible Lignin
P1- SSG59-3; P2- UPMC503; P3- UPMC512; P4- PC23; P5- HC308; P6- Keller; P7- EC582510; P8-
56
Nizamabad forage; P9- EC582508; P10- SEVS4
57
Table 14 . Grouping of lines into different clusters
Cluster
Cluster 1
Cluster 2
Cluster 3
Genotypes
EC582510, EC582508, HC308, SSG59-3, SEVS4
Keller, Nizamabad forage, UPMC503, PC23
UPMC512
58
59
60
Table 15 . Average intra (bold) and inter cluster distances for 10 genotypes
Cluster
I
II
I
II
32.58
47.7
III
47.7
34.37
39.78
52.05
34
III
39.78
52.05
0
61
62
63
64
65
66
67
68
Table 16. Crosses with significant specific combining ability and heterosis for fodder yield and quality
69
traits
Trait
FY
Earliness
Brix
CP
IVOMD
Crosses with significant SCA
Crosses with significant heterosis
Over MP
Over BP
P4 x P9 (84.7), P2 x P9 (60.9), P3 x
P4 x P9 (142.3), P4 x P7 (97.6), P4 x P9 (80.3),
P10 (53.2), P7 x P8 (42.5), P1 x P3
P4 x P6 (82.8), P2 x P9 (72.4),
P2 x P9 (52.2)
(41.3), P5 x P7 (40.2), P7 x P10
P7 x P10 (63.1), P7 x P8 ( 61.4),
(35.8), P4 x P6 (34)
P6 x P7 (59.1), P3 x P10 (58.9),
P4 x P10 (58.3), P1 x P7 (54.1),
P5 x P7 (52.7), P1 x P3 (51.8),
P2 x P6 (47.0), P6 x P10 (45.9),
P2 x P10 (32.9)
P2 x P4 (-5.26), P3 x P7 (-4.25), P1 x P1 x P10 (-9.05), P1 x P7 (P3 x P7 (-8.18),
P10 (-4.14), P3 x P9 (-3.99), P9 x P10 8.56), P3 x P7 (-8.3), P3 x P9 (P1 x P7 (-7.16)
(-3.52), P6 x P10 (-3.54), P3 x P6 (7.69), P9 x P10 (-7.64), P6 x
3.18), P6 x P7 (-2.59), P2 x P5 (P10 (-7.48), P1 x P9 (-7.2), P6 x
2.43),P1 x P9 (-2.27)
P7 (-7.15), P1 x P8 (-6.86), P3 x
P6 (-6.23), P1 x P6 (-6.02)
P4 x P9 (2.48), P2 x P3 (2.33), P4 x
P4 x P9 (38.4),
P4 x P9 (54.0),P4 x P8 (49.5),
P8 (2.14), P3 x P5 (1.87), P3 x P8
P4 x P8 (34.8),
P3 x P4 (38.5), P4 x P10 (34.8), P3 x P4 (27.8)
(1.79), P4 x P6 (1.73), P4 x P10
P2 x P3 (25.0), P3 x P5 (24.6),
(1.66), P5 x P9 (1.35)
P4 x P5 (23.8), P4 x P6 (23.5),
P5 x P9 (21.9)
P2 x P7 (1.40), P8 x P9 (1.31), P6 x
P2 x P7 (34.7), P1 x P7 (32.7),
P1 x P7 (24.2),
P7 (1.25)
P1 x P2 (25.1)
P2 x P7 (31.1),
P6 x P7 (25.6)
P1 x P6 (1.69), P3 x P4 (1.66), P3 x
P3 x P9 (6.31), P4 x P8 (6.16)
P3 x P9 (10.8)
P9 (1.61), P1 x P7 (1.43), P4 x P5
35
70
(1.41), P2 x P5 (1.4)
Lignin
P5 x P7 (-5.73) , P3 x P9 (-3.2), P1 x
P3 x P9 (-8.2), P4 x P5 (-8.0)
Nil
P7 (-2.98), P4 x P10 (-2.95), P1 x P6
(-2.72)
Cellulose
P3 x P6 (2.13), P5 x P7 (1.82), P6 x
P6 x P7 (11.3), P3 x P6 (10.8),
Nil
P7 (1.65), P3 x P7 (1.52), P4 x P6
P3 x P7 (9.2),
(1.29)
Hemicellulose P4 x P6 (1.50), P3 x P6 (1.22), P6 x
P3 x P6 (10.7), P4 x P6 (8.79)
Nil
P7 (1.06), P5 x P7 (1.02)
P1- SSG59-3; P2- UPMC503; P3- UPMC512; P4- PC23; P5- HC308; P6- Keller; P7- EC582510; P8-
71
Nizamabad forage; P9- EC582508; P10- SEVS4
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
36
88
89
(a)
Mid Parent
(b)
Better Parent
150
Mid Parent
Better Parent
25
20
100
15
50
10
5
0
DF
90
PH
NLP
LL
LW
0
FY
-50
-5
DF
PH
NLP
LL
LW
FY
91
Mid Parent
Better Parent
Mid Parent
60
Better Parent
8
50
6
40
4
30
2
20
10
0
0
-2
-10
-20
92
-4
93
94
Fig1. Heterosis for fodder yield and quality traits- a) Maximum heterosis observed among the crosses; b)
95
mean heterosis across all the crosses
96
97
98
99
DF- Days to flower; FY- Fodder yield; PH- Plant height; NLP- Number of leaves per plant; LL- Leaf length;
LB- Leaf breadth; CP- Crude Protein; IVOMD- In vitro organic matter digestibility; CLL- Cellulose; HCLHemicellulose; ADL- Acid Digestible Lignin
100
37
101
38