Abstract
Cotton breeders use various approaches to develop improved
cultivars. Mutation breeding through irradiation of seed is one
method used either in isolation or in combination with other breeding
tools to improve the genetic potential of cotton for agronomic and
fibre quality characteristics. Multi environment trials assist
breeders in selecting stable cultivars. Objectives of this study
were to assess the possibility of obtaining stable mutants within the
first three generations following irradiation and to compare the
performance of mutants to conventionally bred commercial cultivars.
Six mutants selected for their favourable agronomic traits were
evaluated in a Randomized Complete Block trial Design during three
seasons at Cotton Development Trust Research Station. Two
commercial cultivars were used as controls in the study. Analysis of
variance was performed using Genstat. Stability was assessed by
comparing the linear regression coefficients of the mean performance
of genotypes over the seasonal (environmental) mean. Some mutant
lines performed better than controls in a number of traits. The
stability of mutants was comparable to controls in most
characteristics studied. Results of this study demonstrate that
mutation through irradiation has potential in effecting genetic
improvements in cotton. Early selection of lines for advances testing
is possible in cotton mutation breeding as shown by stability of some
mutants in the study.
Introduction
Cotton in Zambia is grown over
an area of more than 150,000 ha. However, the average yield per unit
area (600 kg ha-1) is much lower than that obtained in
some of the leading cotton producing
areas in the world. Improvement
in seed cotton yield is a top priority of the cotton-breeding
programme in Zambia. Evolution of cotton varieties with improved
yield potential can be realized if the there is sufficient genetic
variability for the trait. Genetic variability through cross breeding
depends on the genetic diversity of the parents involved. Mutagenesis
of cotton cultivars or in combination with hybridization has been
shown to be a suitable technique for creating genetic variability in
cotton (Al-Didi, 1965 and Kuliev, 1983). Direct development of
varieties has been considered as one of the most useful roles of
induced mutation in plant breeding (Allard, 1960). Induced mutation
has resulted in some cotton varieties of earlier maturity. Rault et
al 1971 reported cotton cultivars developed through mutation breeding
requiring 60-75 less days to reach maturity.
A number of cultivars obtained
from irradiation of cottonseed have been shown to perform well in a
diverse range of environments (Muhammad et al, 2002). Growing
genotypes in a wide range of environments enables breeders to detect
lines that show specific or general adaptation. Environments can be
defined as locations or seasons (years) or a combination of locations
and seasons. Bucio (1966) used sixteen seasons on one location to
compare the stability of nicotiana rustica progenies with
their parents. Most studies on stability of cotton genotypes have
focused on strains obtained through conventional hybridization
techniques. However, detection of stable mutants in early generations
could lead to more rapid development of improved cultivars by
facilitating early generation selection of promising lines. The
objectives of the present study were;
To assess the possibility of obtaining a stable mutant of cotton in the early generations following irradiation i.e. the M2 or M3 generation.
To compare the performance of mutants with conventional commercial cultivars
Materials
and Methods
Nomenclature
of Genotypes
In the Zambian cotton
mutation-breeding programme, the genotypes are allotted a prefix ‘M’
to denote the process of derivation of the line-as mutation. The
prefix is then followed by the original initials for the genotype and
a numeric indication of the dosage of gamma radiation used to derive
the genotype e.g. MF-20 means this mutant was derived from cultivar
F135 irradiated at 20 Kr.
Field
and Laboratory MethodologyIn
early 2002, six cotton genotypes (FKF, CF, CKF, Chureza, F 135 and
G319-16) were subjected to gamma irradiation. Dry Seeds, 300g per
treatment of each genotype were irradiated at different doses i.e.
15Kr, 20Kr, 30Kr and 40kr. After irradiation, the seed were planted
in non-randomized blocks during the 2002/03 season as M1.
Selection was done on the genotype by treatment populations based on
earliness, hairiness and productivity. Selected lines were grown in a
randomized complete block trial design during the seasons of
2003/04(M2), 2004/05(M3) and 2005/06(M4).
Four replications were used with inter row spacing of 90 cm and intra
row spacing of 30 cm. Each plot consisted of three rows 9 m long and
the centre row was harvested as net plot. Data collected included,
boll weight, plant height, days to flowering, days to boll opening,
lint percent and yield. Boll weight was determined by measuring the
seed cotton weight of twenty hand-harvested bolls and dividing the
weight by 20 to estimate the weight of one boll. Plant height was
obtained by measuring the distance from the base of the stem to the
tip of five arbitrarily chosen plants in each plot. Days to flowering
and boll opening were estimated by counting days to the time when 50%
of the plants in the net plots had flowered or open bolls
respectively. Data were pooled to assess the stability of mutants in
comparison to commercial cultivars as checks. Fibre quality data was
not included in the analysis because it was available in only one
season. In the analysis, years were considered as environments.
Genstat statistical software was used for analysis of variance.
Microsoft excel was used to generate the slopes for stability
determination. The principle developed by Eberhard and Russel (1966)
was used to assess the stability of genotypes. Stability was
determined by the degree of deviation of the slope from 1.00 in a
given character.
ResultsAnalysis
of Variance
Differences were detected among
entries in average boll weight, lint percent and plant height. Table
1 shows the analysis of variance for the studied traits over the
three seasons.
There was significant variation
due to Genotype and Year main effects for average boll weight trait
in the study. Genotype, Year and Genotype by Year accounted for
24.6%, 67.0% and 8.5% of the total variation in sums of squares
respectively. Blanche et al (2006) found that boll weight was
influenced more by Genotype than Environment or Genotype by
Environment interaction effects. Averaged across years, MCZA-20Kr had
significantly (p=0.05) less average boll weight than all the
genotypes used in the study including its parent Chureza (Table 2).
This result indicates a potential negative effect of irradiation on
some cotton genotypes.
Only
the Genotype component of variation was significant for lint percent.
Genotype, Year and Genotype by Year accounted for 63.8%, 3.8% and
32.4% of the total variation in sum of squares (Table1). Kerby et al
(2000) found that variation in sums of squares for lint percent was
influenced more by Genotype and Genotype by Environment than by
Environment. The overall mean for lint percent of MG-15Kr was
significantly higher (p=0.05) than one of the check cultivars,
Chureza (Table 3).
All
sources of variation were significant for the plant height character.
Genotype, Year and Genotype by Year accounted for 4.2%, 87.3% and
8.5% of the total variation in sums of squares for plant height
(Table 1).
Stability
Analysis
The linear regression
coefficient of the mean performance of the genotypes over the
environmental (seasonal) mean was used as a measure of response of
genotype to varying environments. Based on our stability criteria of
degree of deviation of slope of regression line from 1.00, MCKF-40Kr
and MG-15Kr were more stable than both commercial checks for boll
weight trait. MG-25Kr was the most stable genotype for this trait
with a slope of 1.00 (Table2).
All mutants had slopes
significantly different from the 1.00 standard for slope of
regression line. They were therefore more unstable than commercial
cultivar checks for this character (Table 3).
Plant height was the only
character with significant genotype by year interaction in the study.
The stability of mutants in this trait was comparable to that of
commercial checks (Table 4). Similar stability trends were observed
for yield with slopes of mutants comparable to commercial cultivar
checks (Table 7).
Trait | Source | D.f | SS | F-Value | Prob | % Total variation |
| Days to Flowering | Genotype | 7 | 47.88 | 0.5 | 0.833 | 2.8 |
| Year | 2 | 1425.86 | 51.63 | 0.001** | 84.6 |
| Genotype x Year | 14 | 211.22 | 1.09 | 0.389 | 12.6. |
| | | | | | |
Days to Boll Opening | Genotype | 7 | 94.83 | 0.92 | 0.501 | 0.98 |
| Year | 2 | 9336.33 | 316.29 | 0.001** | 96.9 |
| Genotype x Year | 14 | 208.33 | 1.01 | 0.462 | 2.2 |
| | | | | | |
Average Boll Weight | Genotype | 7 | 4.19 | 4.83 | 0.001** | 24.6 |
| Year | 2 | 11.41 | 45.98 | 0.001** | 67.0 |
| Genotype x Year | 14 | 1.44 | 0.83 | 0.633 | 8.4 |
| | | | | | |
Lint Percent | Genotype | 7 | 38.9 | 4.67 | 0.001** | 63.8 |
| Year | 2 | 2.33 | 0.98 | 0.384 | 3.8 |
| Genotype x Year | 14 | 19.75 | 1.18 | 0.320 | 32.4 |
| | | | | | |
Plant Height | Genotype | 7 | 2683.9 | 3.12 | 0.009* | 4.2 |
| Year | 2 | 55666.4 | 226.62 | 0.001** | 87.3 |
| Genotype x Year | 14 | 5407 | 3.14 | 0.002* | 8.5 |
| | | | | | |
Yield | Genotype | 7 | 780340 | 1.13 | 0.359 | 3.3 |
| Year | 2 | 21167576 | 107.59 | 0.001** | 90.4 |
| Genotype x Year | 14 | 1460650 | 1.06 | 0.416 | 6.2 |
Table 1- Degrees of freedom, sums of squares, and
significance levels by trait
Traits
marked with asterisks * and ** are significantly different (p=0.05
and p=0.01 respectively)
| Genotype | 2003/04 | 2004/05 | 2005/06 | Mean | Slope for Regression |
| MFKF- 20Kr | 4.6 | 5.5 | 5.7 | 5.3 | 1.22 |
MCF-30Kr | 5.0 | 5.1 | 5.5 | 5.2 | 0.44 |
MCF-20Kr | 4.5 | 5.6 | 5.6 | 5.2 | 1.33 |
MCKF-40Kr | 4.5 | 5.3 | 5.5 | 5.1 | 1.13 |
MCZA-20Kr | 3.7 | 4.8 | 4.9 | 4.5 | 1.32 |
MG-15Kr | 4.6 | 5.4 | 5.5 | 5.1 | 1.00 |
Chureza © | 4.8 | 5.3 | 5.6 | 5.2 | 0.85 |
| F 135 © | 4.8 | 5.2 | 5.5 | 5.2 | 0.70 |
| Mean | 4.6 | 5.3 | 5.5 | | |
| CV % | 3.5 | 8.5 | 2.5 | 6.9 | |
LSD (5%) | ns | ns | 0.2 | 0.3 | |
Table 2-mean boll weight in each year, across years and
stability statistics for each genotype
Table
3-mean Lint Percent (%) in each year, across years and slope of
regression for each genotype
| Genotype | 2003/04 | 2004/05 | 2005/06 | Mean | Slope for Regression |
| MFKF- 20Kr | 40.7 | 42.5 | 42.1 | 41.8 | 3.8 |
MCF-30Kr | 40.2 | 41.9 | 41.9 | 41.3 | 4.3 |
MCF-20Kr | 42.4 | 41.8 | 41.8 | 42.1 | -1.2 |
MCKF-40Kr | 43.5 | 42.5 | 42.6 | 42.9 | -2.3 |
MCZA-20Kr | 40.1 | 41.4 | 40.8 | 40.8 | 2.2 |
MG-15Kr | 43.5 | 42.1 | 43.8 | 43.2 | -0.1 |
Chureza © | 41.7 | 41.7 | 42.1 | 41.8 | 0.8 |
| F 135 © | 42.2 | 42.7 | 42.3 | 42.4 | 0.5 |
| Mean | 41.8 | 42.1 | 42.2 | | |
| CV % | 2.8 | 2.5 | 2.1 | 2.6 | |
LSD (5%) | 2.0 | ns | 1.5 | 1.0 | |
Table
4-mean Plant Height (cm) in each year, across years and slope of
regression for each Genotype
| Genotype | 2003/04 | 2004/05 | 2005/06 | Mean | Slope for Regression |
| MFKF- 20Kr | 72.3 | 121 | 156 | 116.4 | 1.23 |
MCF-30Kr | 99.3 | 120 | 153.3 | 124.2 | 0.77 |
MCF-20Kr | 80.7 | 121.7 | 132 | 111.4 | 0.78 |
MCKF-40Kr | 76 | 130.7 | 148 | 118.2 | 1.09 |
MCZA-20Kr | 69.3 | 132 | 124 | 108.4 | 0.87 |
MG-15Kr | 73.3 | 121.7 | 158 | 117.7 | 1.25 |
Chureza © | 95 | 116.7 | 167.3 | 126.3 | 1.02 |
| F 135 © | 93 | 122 | 161.3 | 125.4 | 0.99 |
| Mean | 82.4 | 123.2 | 150.0 | | |
| CV % | 10.8 | 12.4 | 0.5 | 9.3 | |
LSD (5%) | 15.6 | 26.7 | 1.26 | 10.5 | |
| Genotype | 2003/04 | 2004/05 | 2005/06 | Mean | Slope for Regression |
| MFKF- 20Kr | 129.3 | 108 | 131.0 | 122.8 | 0.92 |
MCF-30Kr | 131.3 | 110.3 | 135.0 | 125.6 | 0.96 |
MCF-20Kr | 134.6 | 105. | 132.0 | 123.9 | 1.17 |
MCKF-40Kr | 134.3 | 107 | 134.3 | 125.2 | 1.13 |
MCZA-20Kr | 132.3 | 107.3 | 133.3 | 124.3 | 1.05 |
MG-15Kr | 129.7 | 106.3 | 134.3 | 123.4 | 1.07 |
Chureza © | 130 | 107 | 130.3 | 122.4 | 0.96 |
| F 135 © | 127 | 110.3 | 129.7 | 122.3 | 0.75 |
| Mean | 131.1 | 107.6 | 132.5 | | |
| CV % | 3.5 | 3.7 | 1.8 | 3.1 | |
LSD (5%) | ns | ns | ns | ns | |
Table 5-mean Days to Boll Opening in each
year, across years and slope of regression for each genotype
| Genotype | 2003/04 | 2004/05 | 2005/06 | Mean | Slope for Regression |
| MFKF- 20Kr | 78.7 | 77.3 | 68.7 | 74.9 | 0.94 |
MCF-30Kr | 81.3 | 73.3 | 70.3 | 75 | 0.99 |
MCF-20Kr | 83.7 | 70.3 | 70.7 | 74.9 | 1.14 |
MCKF-40Kr | 82.0 | 76.7 | 69.7 | 76.1 | 1.13 |
MCZA-20Kr | 78.3 | 76.7 | 68.3 | 74.4 | 0.94 |
MG-15Kr | 79.7 | 75.3 | 68.7 | 74.6 | 1.02 |
Chureza © | 80.0 | 79.3 | 69.7 | 76.3 | 0.98 |
| F 135 © | 78.0 | 74.3 | 68.7 | 73.7 | 0.86 |
| Mean | 80.2 | 75.4 | 69.3 | | |
| CV % | 3.4 | 7.0 | 1.5 | 5.0 | |
| LSD (5%) | ns | ns | ns | ns | |
Table 6-mean Days to flowering in each year, across
years and slope of regression for each genotypeTable
5-mean Seed cotton Yield (Kg/ha) in each year, across years and slope
of regression for each genotype
| Genotype | 2003/04 | 2004/05 | 2005/06 | Mean | Slope for Regression |
| MFKF- 20Kr | 1333 | 1430 | 2639 | 1801 | 1.09 |
MCF-30Kr | 1377 | 1219 | 2323 | 1639 | 0.87 |
MCF-20Kr | 953 | 1358 | 2244 | 1518 | 0.98 |
MCKF-40Kr | 1150 | 1488 | 2304 | 1649 | 0.88 |
MCZA-20Kr | 799 | 1226 | 2374 | 1466 | 1.21 |
MG-15Kr | 1036 | 1690 | 2546 | 1758 | 1.08 |
Chureza © | 1400 | 1194 | 2434 | 1676 | 0.96 |
| F 135 © | 1387 | 1115 | 2342 | 1614 | 0.92 |
| Mean | 1179.4 | 1340.0 | 2400.8 | | |
| CV % | 35.9 | 17.9 | 8.6 | 19.1 | |
LSD (5%) | ns | ns | ns | ns | |
Discussion
Stability data in tables 2
through 7 suggest that early-generation mutants of cotton can be as
stable as the established conventional cultivars in most
characteristics. Lines MFKF-20Kr and MG-15Kr were the most stable
mutants with instability in only one character, lint percent. The
lines were even more stable than one of the check varieties, F 135.
Overall, Chureza was the most stable entry in the study with
stability in all the characteristics studied. Some mutants performed
better than the commercial check Chureza in lint percent (MG-15Kr and
MCKF-40Kr). Mohammed et al (2002) found similar results in their
study of cotton mutants obtained through irradiation. The
significantly shorter plants (p=0.05) obtained among some m0pmutant
lines could indicate the potential of this breeding method in
breeding for shorter but high yielding genotypes. There is need for
further research on the effectiveness of this breeding tool or in
combination with other methods such as interspecific hybridization in
improving the fibre properties of cotton and the quality of products
obtained from cottonseed such as seed proteins and gossypol.
Conclusion
Results of the analysis of
variance demonstrate that mutation breeding through irradiation has
potential in effecting genetic improvements of cotton in essential
agronomic traits without excessive deleterious effects. One potential
advantage of this method is that the genetic gains can be made in a
shorter period than using conventional methods of hybridization
followed by pedigree selection.
References
Allard, R, W, 1960 Principles of Plant Breeding, John Wiley and Sons Inc.New York, pp 485
Al-Didi, A.A., 1965. Development of new Egyptian cotton strains by seed Irradiation. The use of induced mutation in plant breeding. Radiat.Bot. Suppl., 5: 579-583
Eberhart, S.A and W.A Russell, 1966. Stability parameters for comparing varieties. Crop Sci.6: 36-40
Bucio Alanis, L. 1966 Environmental and genotypes by environmental components of variability. Inbred lines. Heredity, 21, 387-97
Blanche B.S, Gerald O.M, Jimmy Z.Z, David C. and James H. (2006). Stability Comparisons Between Conventional and Near Isogenic Transgenic Cotton Cultivars. Cotton Science 10: pp 17-28
Kerby,T, J.Burges, M.Bates, D. Albers and K. Lege, 2000. Partitioning Variety and Environment contribution to variation in Yield, Plant growth and Fiber quality.p. 528-532. In proc. Beltwide Cotton Conference, New Orleans, L.A, 7-10 Jan.. Natl Cotton Counc. Am. Memphis TN.
Kuliev, R.A., 1983. Combining hybrid and mutational variation in Cotton Breeding. Sel’skokhozyaistvennaya Biology, No. 9: 4-40
Linn, R and Binn, T. 1980, Understanding Genotype by Environment Interaction in Wheat, Theoretical and Applied Genetics 145-150.
Muhammad Mureed Kandhro, Sawan Laghari, Mahboob Ali Sial and Ghulam Shah Nizamani. 2002, Performance of Early Maturing Strains of Cotton (Gossypium hirsutum L Developed through Induced Mutation and Hybridization. Asian Journal of Plant Sciences, Volume 1 Number 5: 581-582,
Rault R.N, Jain H.K, Parnwar R.S. 1971, International Symposium on use of Isotopes Radiation in Agricultural Animal Husbandry Research, New Delhi pp 223
