Volume 31(8).

Published July, 01, 2016

www.jnsciences.org
E-ISSN 2286-5314

Genetic analysis of barley (Hordeum vulgare L.) grain yield

components

M. A. BARGOUGUI

School of Higher Education in Agriculture of Mateur

*Corresponding author: [email protected]

Abstract - The purpose of this study is to develop a selection approach to improve barley (Hordeum
vulgare L.) grain yield, based on the various yield components, respectively, the spike-number per plant,
grain-number per spike and the thousand-kernel weight. A complete diallel cross, including five barley
genotypes, was analyzed for estimating variance of genetic effects, combining ability and heritability.
The result revealed a highly significant general combining ability of the spike-number, sustained by
high narrow sense heritability equal to 0.49. We noted that the genotypes Manel and INAT102 have the
best values of general combining ability, and their progeny presented the most important specific
combination abilities. Thus, selection based on the spikes-number appears to be quite adequate. The
pedigree method is recommended, in which the genotypes Manel and INAT102 make good head that
can be crossed together or with other parents to create new interesting genetic combinations for grain
yield in barley.

Keywords: Diallel cross, barley (Hordeum vulgare L) grain yield components, genetic effects,
heritability, selection approach.

1. Introduction
Among the cereal crops, barley is a species with the greatest adaptability to a wide range of
environments. Barley is cultivated from arctic latitudes to tropical areas, grown at the highest altitudes
and adapted to specific sets of agro-ecological areas (Alemayehu and Parlevliet, 1997). It is adapted to
a broad range of agro-ecological environments and tolerant to soil salinity, drought and frost to a
considerable level. The crop grows successfully in the arid climates. In terms of the area and production
worldwide, barley is the fourth most important cereal after wheat, rice and maize (Abebe, 2010). In a
2007 ranking of cereal crops in the world, barley was fourth both in terms of quantity produced (136
million tons) and in area of cultivation (566,000 km²).
In Tunisia, barley (Hordeum vulgare L.) is one of the principal cereal crops. Elsewhere the fertile land,
generally limited to the north, barley is produced in quantity in the southern desert and plateau, where
desert farming is precarious. As reported by Abdellaoui et al (2010), among cereals, barley (Hordeum
vulgare L.) is one of the principal cereal crops in Tunisia since it is cultivated in different zones from
northern to southern Tunisia and occupies about the third of Tunisian cereals’ cultivated area. The
national production was about 772,500 (tons) in 2014. It is used as a livestock feed, for malt and for
preparing foods. The roasted grains are a coffee substitute.
Grain yield is made up of three main different yield components, specifically, the number of spikes, the
number of grains per spike and the thousand-kernel weight. Regarding interactions between grain yield
and yield components in barley, Barczak and Majcherczak (2009) found the highest interaction between
grain yield and spike number per m2, and between grain yield and grain number per spike. Ebrahim,
(2015) also revealed that grain yield is positively correlated to kernel number per spike. In addition to
that, positive and significant correlations of grain yield with spike number per m2 and thousand-kernel
weight were reported by Ataei (2006).
The choice of an efficient breeding program depends to a large extent on the knowledge of gene action
involved in the expression of the character. Diallel analyses were used to estimate gene action of yield
components in barley and provide an opportunity to obtain a rapid and overall picture of genetic control
of a set of parents in the early generation. Besides genetic effects, breeders would also like to know how
much of the variation in a crop is genetic and to what extent this variation is heritable, because the
efficiency of selection mainly depends on additive genetic variance (Eshghi and Akhundova, 2009).

Bargougui (2016) / Journal of new sciences, Agriculture and Biotechnology, 31(8), 1794-1799 1794
Volume 31(8). Published July, 01, 2016
www.jnsciences.org
E-ISSN 2286-5314
These authors reported additive genetic effects of tillers number and an over-dominance gene action for
number of grains per spike.
Genetic study of forage yield and other traits in barley (Hordeum vulgare L.), based on twenty one
genotypes of six – parent half diallel analysis, realized by Pal and Kumar (2009) revealed significant
role of additive genetic component for the inheritance of days to 50% heading, plant height and spikelets
per ear. A non-additive component was found to be important for the genetic control of all the traits
except for days to 50% heading and tillers per plant.
Concerning the approaches of grain yield improvement in barley, Lalić et al (2006) reported that genetic
gain estimates showed that pedigree method was efficient for shorter stem, higher number of spikes per
plant, and single grain weight and grain yield per plant. While single seed descent method was superior
for grain yield per plant. This approach is also based on the heritability of the various characteristics and
on the general combining abilities of the parental genotypes. Thus, the choice of the hybridization
between the head of lines is made according to the best specific combining abilities.

2. Materials and methods

To quantify the yield, we measured its three most interesting components, respectively the spike-number
per plant, the grain-number per spike, and the thousand-kernel weight.
A complete diallel cross including five barley genotypes, Rihane, Manel, Martin, INAT-102 and Souihli
(Bargougui, 2016) had been realized. The five parental lines and their 20 hybrids progenies were grown
the same year and under the same crop conditions, according to a completely randomized design, in
which the experimental unit is represented by a row of 125 cm length for each genotype.
Twenty plants had been scored for the spike-number, and forty spikes per genotype had been scored for
the grain-number and the thousand-kernel weight.

3. Results and discussion

3.1. Analysis of the variance
The data concerning the three yield components obtained from all parents and their F1 offspring is
subjected to an analysis of the variance taking into consideration genotypes and replicates. Statistical
data and analysis of the variance will be presented respectively for the spike-number, the grain-number
and the thousand-kernel weight.

3.1.1. The spike-number

The analysis of the variance for the spike-number (Table 1) showed that only the genetic effect
contributed to a significant variation, the effect of the replicates was not significant

Table 1: analysis of variance for the spike-number

Source D.F. SS MS Variance ratio Pr>F
Total 499 27430.232
Genotypes 24 6190.632 257.955 5.87 < 0.0001
Replicates 19 1186.232 62.433 1.42 < 0.1121
Error 456 200053.068 43.976

Duncan t test revealed an important variability of the spike-number and if we consider the genotype
INAT102 as having the minimal value for the best cross choice, we note that the majority of its own
hybrids are significantly superior. Indeed the most interesting hybrid was resulting from INAT102
crossed with Manel.
The decomposition of the genetic effect (Table: 2) showed the preponderance of the general combating
ability, indicating that this characteristic depends probably more on the additive genetic action than on
the interactions. This is of a great interest for the genetic improvement of this component.

Bargougui (2016) / Journal of new sciences, Agriculture and Biotechnology, 31(8), 1794-1799 1795
Volume 31(8). Published July, 01, 2016
www.jnsciences.org
E-ISSN 2286-5314
Table 2: Analysis of variance for the number of the spike-number according to Griffing (1956)
Sources F.D SS MS F Pr>F
General combining ability 4 125.34425 31.336 13.74 < 0.0001
Specific combining ability 10 209.5404 8.4165 3.69 < 0.0001
Reciprocal maternal effect 4 287.9324 19.598 8.56 < 0.0001
Reciprocal specific effect 6 21.61425 3.6 1.58 < 0.1000
Error 475 1083.57925 2.28

The classification of parents according to their general combining abilities (Table 3) revealed a very
important variation, which supports already the analysis of the variance. This classification puts at the
head the genotype INAT-102 followed by Manel and Rihane. Thus, to improve this characteristic the
genotype INAT-102 seems to be the best head of lines in any future cross.

Table 3: General combining abilities classification for the spike-number

Parent General combining ability Classification for α = 5%
INAT102 2.371 A
Manel 0.411 a b
Rihane 0.271 b
Martin - 0.524 b c
Souihli - 2.0529 c
Var (gi – gj) = 1.0640929

3.1.2. The grain-number

For the grain-number, both genotypes and replicates showed high significant variation (Table 4). This
indicates that this characteristic is influenced by the environment conditions generating an individual
variation and an essential genetic effect.

Table 4: analysis of variance for the grain-number

Source D.F. SS MS Variance ratio Pr>F
Total 999 221410.799
Genotypes 24 32182.244 1340.926 8.39 < 0.0001
Replicates 39 39556.959 1014.281 6.34 < 0.0001
Error 936 149671.616 159.9056

Duncan t test showed that the genotype Martin had the best value, which differs significantly from the
other parents except when compared to the INAT102. An increased value is noted for the majority of
their hybrids. The decomposition of the genetic effect (Table 5) showed that there is a prevalence of the
specific combining ability and reciprocal effect. This indicates that genetic improvement of grain yield
based on this characteristic is less probable.

Table 5: Analysis of variance for the grain-number according to Griffing (1956).

Sources F.D SS MS F Pr>F

General combining ability 4 38.03675 9.5092 2.38 < 0.05
Specific combining ability 10 572.54968 57.255 14.32 < 0.0001
Reciprocal maternal effect 4 160.565 40.1413 10.04 < 0.0001
Reciprocal specific affect 6 33.4046875 5.5675 1.39 < 0.1000
Error 936 3741.7904 3.9976

Similar results are obtained by Madić et al (2014), particularly for grain number per spike. The
classification of parents according to their general combining abilities for the grain-number (Table 6)
sets the genotypes Rihane and Souihli in the best ranks. INAT-102 showed a negative value, which
differs from the preceding classification according to the spike-number.

Bargougui (2016) / Journal of new sciences, Agriculture and Biotechnology, 31(8), 1794-1799 1796
Volume 31(8). Published July, 01, 2016
www.jnsciences.org
E-ISSN 2286-5314
Table 6: General combining abilities classification for the grain-number
Parent General combining ability Classification for α = 5%
Rihane 0.9715 A
Souihli 0.9265 a
INAT102 - 0.1035 a b
Manel - 0.461 a b
Martin - 1.3335 b
Var (gi – gj) = 0.799528

3.1.3. Thousand-kernel weight

For the thousand-kernel weight, the analysis of variance (Table 7) indicates that the replicate effect
remains relatively important. This indicates that this characteristic undergoes the environment effects.
However, genotypic variation is more important.

Table 7: Analysis of variance for the thousand-kernel weight

Source D.F. SS MS Variance ratio Pr>F
Total 999 151327.9859
Genotypes 24 63600.7397 2650.0308 31.49 < 0.0001
Replicates 39 8958.3905 229.7023 2.73 < 0.0001
Error 936 78768.8557 84.1548

The genotypes classification according to the thousand-kernel weight showed that Martin is the
genotype that has the most important thousand-kernel weight followed by Manel and souihli. For this
characteristic INAT-102 has a relatively moderate value. However, its hybrids remain among the best
hybrids. Overall, the hybrids expressed a more important performance, compared to the parental
genotypes. This avoids any restriction to choose the best hybrid combinations. According to the analysis
of the variance (Table 8), we notice the prevalence of the effects of genetic interaction and the reciprocal
effects compared to the general combining ability.

Table 8: Analysis of variance for the thousand-kernel weight according to Griffing (1956)
Sources F.D SS MS F Pr>F
General combining ability 4 167.232136 41.81 19.87 < 0.0001
Specific combining ability 10 1199.250798 119.93 57 < 0.0001
Reciprocal maternal effect 4 116.9545711 29.24 13.9 < 0.0001
Reciprocal specific affect 6 106.5635312 17.76 8.44 < 0.0001
Error 936 1969.22139 2.10

The classification of the genotypes according to their general combining ability for thousand-kernel
weight (Table 9) puts the genotype Martin at the first rank followed by Manel variety.

Table 9: General combining abilities classification for the thousand-kernel weight

Parent General combining ability Classification for α = 5%
Martin 2.998465 A
Manel 1.005315 b
Rihane - 0.60901 c
Souihli - 1.155285 c
INAT102 - 2.23951 d
Var (gi – gj) = 0.420774

As a conclusion, we may retain that the spike-number constitutes the component that imply the additive
genetic effect, and can be used to genetic improvement of the grain-yield of the Barley.
Taking into account these results, INAT102 and Manel genotypes can be considered as being suitable
parental genotypes and their cross as the better hybrid combinations to select best resistant varieties with
better grain yield.

Bargougui (2016) / Journal of new sciences, Agriculture and Biotechnology, 31(8), 1794-1799 1797
Volume 31(8). Published July, 01, 2016
www.jnsciences.org
E-ISSN 2286-5314
3.2. Heritability
The narrow sense heritability values of the three yield-components were estimated by the regression of
the hybrid values on their respective parents (Falconer, 1960). As it is shown in the table 10, only the
spike-number has significant heritability values.

Table 10: Narrow sense heritability Values for the three yield-components
Characteristic Heritability " h² " Probability > | T |
Spike-number 0.49 0.0456
Grain-number per spike 0.24 0.6506
Thousand-kernel weight 0.25 0.4582

This result sustains another time the prevalence of the general combining abilities, noted previously for
these characteristics. Particularly, the spike-number is a heritable yield-component and it can be useful
for the improvement of barley grain-yield. In fact, Yadav et al (2015) reported corresponding
conclusions.

3.3. Choice of head of line crosses

Let us recall first that the genetic analysis allowed us to consider the genotypes Manel and INAT102 as
the best parents for the creation of new genetic combinations interesting for the barley grain-yield. The
choice of the parents has been done according to their general combining abilities. The choice of their
hybrid crossings will be based on the specific combining ability given by the table 11.

Table 11: The specific combining abilities for the spike-number

Rihane Manel Martin INAT-102 Souihli
Rihane - 2.086
Manel - 0.136 - 2.616
Martin 0.394 2.119 - 2.796
INAT-102 2.0789 3.959 2.699 - 3.636
Souihli - 0.036 1.509 1.174 0.364 - 4.936

Crosses including Manel and INAT102 as parental genotypes exhibit an interesting specific combining
abilities which likely can be considered as being suitable heads of lines in a genetic improvement
program to increase the spike-number.
Selection based on the spike-number appears to be possible to create new high yielding barley varieties.
A pedigree method, in which the Manel and INAT102 genotypes constitute the best parents who can be
cross between them or with other parents, is recommended.

4. Conclusion
Barley (Hordium vulgare L.) is self pollinated species and improved genotypes are often developed by
inbreeding within population formed by artificial hybridization. In such case, improvement schemes
involve crossing between parental lines followed by a selection among their segregating generations. It
is necessary to establish the number of generations required to get a sufficient and adequate level of
uniformity to obtain lines possessing desirable attributes. Cultivars displaying high performances that
can be released for use are derived as early as the F2 and as late as the F11 generation.
For highly heritable traits, pedigree selection is appropriate to obtain homogeneous improved lines
promptly. F2 genotypes are subjected to individual plant selection for desirability in one or more
important traits. Then, selected plants are harvested individually and followed by progeny testing.
Such a study revealed high heritability of the spike-number, allowing efficient selection for yielding
barley varieties. The combining abilities analyses showed a mainly additive genetic effect. So selection
should be earlier in order to carry the characters within individual homozygous lines.
A pedigree method in which the Manel and INAT102 genotypes constitute the best parents that can be
inter-crossed or crossed with other parents is recommended. Hybridization can include more than two
parents and more complex crosses can be used to combine different interesting genes within potential
new improved varieties. According to Bargougui (2016) result, we recommend simultaneous selection
for the spikes-number and powdery mildew “Erysiphe graminis f. sp.” resistance to improve grain yield

Bargougui (2016) / Journal of new sciences, Agriculture and Biotechnology, 31(8), 1794-1799 1798
Volume 31(8). Published July, 01, 2016
www.jnsciences.org
E-ISSN 2286-5314
of barley. In deed Simmonds (1988) indicated that the main problem lies in combining the resistance
with other improved characteristics to create high-yielding varieties.

5. References
Abdellaoui R. Kadri K. Ben Naceur M. and Bettaib Ben Kaab L. (2010) Genetic diversity in some Tunisian
barley landraces based on RAPD markers. Pak. J. Bot., 42(6): 3775-3782.
Abebe T.D. (2010). Genetic Diversity and Population Differentiation Analysis of Ethiopian Barley (Hordeum
vulgare L.) Landraces using Morphological Traits and SSR Markers. Wellega, Ethiopia.
Alemayehu F, Parlevliet J.E. (1997). Variation between and within Ethiopian barley landraces Euphytica 94:183-
189.
Ataei M (2006). Path analysis of barley (Hordeum vulgare L.) yield. Ankara Univ. Fac. Agric. J. Agric. Sci.
12:227-232.
Barczak B. and Majcherczak E. (2009). Effect of varied fertilization with sulfur on selected spring barley yield
structure components. Journal of Central European Agriculture, 9(4): 777-784.
Bargougui M.A. (2016). Diallel analysis of the resistance to powdery mildew (Erysiphe graminis f.sp. hordei) in
barley (Hordeum vulgareL.). Journal of New Sciences. Vol : 27(2) p : 1466-1473.
Ebrahim S., E. Shiferaw and F. Hailu. (2015). Evaluation of genetic diversity in barley (Hordeum vulgare L.)
from Wollo high land areas using agro-morphological traits and hordein. Afican Journal of Biotechnology
Vol. 14(22), pp. 1886-1896.
Eshghi R. and E. Akhundova. (2009) Genetic analysis of grain yield and some agronomic traits in hulless barley.
African Journal of Agricultural Research Vol. 4 (12), pp. 1464-1474.
Falconer, D.S. (1960) Introduction to Quantitative Genetics. Oliver & Boyd, Edinburgh, London, 380.
Griffing B. (1956). Concept of general and specific combining ability in relation to diallel crossing systems.
Austral. J. Biol. Sci 9, 463-493.
Lalić A, Novoselovi D., Kovacelic J., Drezner G., Babic D., Abicic I. Dvojkovic K. (2010). Genetic gain and
selection criteria effects on yield and yield components in barley (Hordeum vulgare L.). Periodicum
biologorum, Vol 112, No 3. pp 311–316.
Madić Milomirka R., Djurović Dragan S., Knezević Desimir S., Paunović Aleksandar S. and Tanasković
Snežana T. (2014). Combining abilities for spike traits in a diallel cross of barley. Journal of Central
European Agriculture, 15(1), p.108-116.
Pal, D. and S. Kumar (2009) Genetic analysis offorage yield and other traits in barley (Hordeum vulgare L.).
Barley Genetics Newsletter 39: 13-19.
Simmonds N. W. (1988) La résistance aux maladies. Principes d’amélioration génétique des végétaux. Les presses
de l’université Laval. Québec, p. 245-246.
Yadav Sunil K., Ashok K. Singh, Praveen Pandey and Smita Singh. (2015). Genetic Variability and Direct
Selection Criterion for Seed Yield in Segregating Generations of Barley (Hordeum vulgare L.). American
Journal of Plant Sciences, 2015, 6, 1543-1549.

Bargougui (2016) / Journal of new sciences, Agriculture and Biotechnology, 31(8), 1794-1799 1799