correlation studies of some yield related traits in linseed, linum ...

Correlation studies in linseed
121
CORRELATION STUDIES OF SOME YIELD RELATED
TRAITS IN LINSEED, LINUM USITATISSIMUM
Tahira Bibi*, Talat Mahmood**, Yasin Mirza***, Tariq Mahmood*
and Ejaz-ul-Hasan*
ABSTRACT
This study was conducted at Crop Sciences Institute, National Agricultural
Research Centre, Islamabad, Pakistan during the year 2010-11. In this study
extent of genetic variability and interrelationship among agronomic traits was
estimated in a set of 19 linseed (Linum usitatissimum) genotypes. Highly
significant differences (p ≤ 0.01) were observed for each trait except number of
seeds per capsule. All traits except days to flower initiation and harvest index
were positively associated with seed yield. Seed yield per plot had significant
positive correlation with plant height and number of capsules per plant. The
analysis for coefficient of variations was the highest in seed yield per hectare
followed by seed yield per plot, harvest index and capsules per plant.
Heritability (81 to 91%) was found for days to flower initiation (0.825), days to
flower completion (0.809), days to maturity (0.832), 1000-grain weight (0.862),
capsules per plant (0.816) and harvest index (0.908). Maximum genetic advance
was observed for seed yield per hectare (384.523) followed by seed yield per
plot (362.843), capsules per plant (25.939), plant height (13.191) and seeds per
capsule (9.156). Seed yield per plot, capsules per plant, plant height and seed
yield per hectare had high and significant heritability estimates coupled with
corresponding high value of genetic advance. The characters like harvest
index, 1000-grain weight, days to flower initiation and days to maturity gave
high heritability values with low genetic advance indicating the control of nonadditive
gene action.
KEY WORDS: Linum usitatissimum; genotypes; agronomic characters;
Pakistan.
INTRODUCTION
Linseed (Linum usitatissimum) is an important oil and fibre crop with multiple
uses. Its seed is used in treatment of some provocative human and animal
diseases in Pakistan. Linseed oil is mainly utilized in the preparation of
paints, printing ink, varnish, resins and several by-products. The oil cake is
*Oilseeds Research Institute, AARI, Faisalabad. ** PMAS Arid Agriculture University,
Rawalpindi. ***Crop Sciences Institute, Oilseed Programme, NARC, Islamabad. email:
tahira167_uaar@hotmail.com
J. Agric. Res., 2013, 51(2)

122
T. Bibi et al.
the most valuable feeding cake for animals. Linseed has also great medicinal
values including anti-hyper cholesterolemic and anti-carcinogenic effects and
is also beneficial for development of brain and retinal tissues of infants (23).
Pakistan is producing 2779 tons of linseed from an area of 4018 hectares
with an average yield of 692 kg per hectare (5). In Pakistan, the potential of
crop is not being exploited satisfactorily due to many constraints like
inappropriate nutrients supply (3). Harvesting is also a major problem with
linseed and particularly if it is late, incompletely desiccated or lodged.
Selection for high seed yielding genotypes is difficult because it is controlled
by multiple traits that are highly influenced by environment. Due to these
factors, success rate of selection procedure is low. Indirect selection through
yield component has been proved more effective (9). This selection criterion
takes into account the information on extent and nature of interrelationship
among agronomic traits and also their relationship with seed yield. However,
selection for yield via highly correlated characters becomes easy.
Genetic traits such as coefficient of variations, heritability and genetic
advance provide precise estimate of genetic variation of quantitative traits
(13, 14, 15). Equipped with such information, a breeder can formulate
efficient scheme of multiple trait selection.
Estimation of genotypic and phenotypic variances, coefficients of genotypic
and phenotypic variation, heritability and genetic advance provide precise
estimates of genetic variability of quantitative characters (21, 25). Studies of
character association may supply more reliable information on the nature and
level of interrelationships of linseed yield with its yield components. Economic
characters such as seed yield are controlled by many genes, and has a
complex type of inheritance (24).
Genotypic coefficient of variation helps measure the range of genetic
variability in a character and provides a way to compare the genetic variation
in quantitative traits (12). Heritability is the only useful genetic parameter for
predicting direct and correlated response (7). Relative amount of heritable
portion of variation can be assessed through estimation of heritability.
However, some time high heritability is not indicative of high genetic advance.
Therefore, various researchers like Johanson et al. (10) have suggested that
high heritability should be combined with genetic advance in the prediction of
phenotypic expression of a trait, instead heritability estimates alone, whereas
low heritability of a trait proves its stability as desirable genetic parameters
(18).
J. Agric. Res., 2013, 51(2)

Correlation studies in linseed
123
The present study was conducted to find association among yield and yield
related traits in linseed that may be helpful in selection of genotypes.
MATERIALS AND METHODS
This study was conducted at Crop Sciences Institute, National Agricultural
Research Centre, Islamabad, Pakistan during the year 2010-11. Nineteen
linseed genotypes viz. LS-1, LS-2, LS-3, LS-4, LS-8, LS-11, LS-12, LS-13,
LS-14, LS-17, LS-19, LS-24, LS-25, LS-26, LS-30, LS-31, LS-32 and LS-33
including one check variety (Chandni). The experiment was planted in RCBD
with three replications. Each experimental unit comprised four rows and plot
size was 5 m x 0.45 m. Plant to plant distance was kept as 10 cm. Data were
recorded for days to flower initiation, days to flower completion, days to
maturity, plant height (cm), number of branches per plant, number of
capsules per plant, number of seeds per capsule, 1000-seed weight (g) and
harvest index (%) from five randomly selected plants in each plot. Seed yield
per plot and seed yield per hectare was calculated from whole plot. Data
recorded were subjected to analysis of variance as described by Steel and
Torrie (27). Subsequently genotypic and phenotypic variances, heritability
(h 2 ), genetic advance (GA), genotypic coefficient of variations (GCV), and
phenotypic coefficient of variations (PCV) were worked out following Al-
Jibouri et al. (4). Variance and covariance components were estimated by
equating observed mean squares to their expectations according to Singh
and Chaudhary (26). The characters with significant mean squares values
were subjected to Duncan multiple range test (DMRT) for their comparison.
RESULTS AND DISCUSSION
Highly significant differences (p ≤ 0.01) were observed for each trait studied
except number of seeds per capsule. It indicated presence of significant
variability for effective selection to identify potential genotypes. The mean
squares of all the characters analyzed are presented in Table 1.
For days to flower initiation, LS-3 and LS-30 were found earlier (108 days
each) whereas LS-17 completed their flowering in about 155 days as
compared to check (160 days) (Table 2). The genotype LS-17 matured in
181 days that significantly differed from check variety. Similarly number of
branches per plant ranged from 5-7 showing genetic differences among the
genotypes indicating potential for improvement. Number of capsules per
plant ranged from 62 to 12 in LS-17 and LS-1, respectively. All genotypes
J. Agric. Res., 2013, 51(2)

124
T. Bibi et al.
Table 1. Mean square values for various yield related traits of linseed genotypes during
Rabi 2010-11.
S.O.V d.f Days to
flower
initiation
Days to
flower
completion
Days to
maturity
Plant
height (cm)
No. of
branches/
plant
No. of
capsules/
plant
Replications 2 2.30 0.5 2.02** 34.67 0.11 264.5**
Genotypes 18 10.59** 9.06** 9.69** 200.76** 1.02** 627.25**
Error 36 0.71 0.66 0.607 25.08 0.10 43.97
LSD (5%) 1.29 1.51 1.20 8.29 0.60 13.02
No. of seeds
capsule
1000-grain
weight (g)
Harvest
index (%)
Seed yield/
plot(g)
Seed yield/ha
(kg)
911.41 2.31** 5.81** 98948.91** 48567.17
1036.37 1.57** 22.04** 119064.22** 183405.58**
622.31 0.080 0.73 7423.81 26324.87
0.38 1.19 1.45 119.72 166.40
** Significant at 1 % level of probability.
showed significant variations for number of seeds per capsule and LS-33 and
LS-11 had maximum value (9 seeds) against check variety (8 seeds) which
was statistically significant (Table 2). 1000-grain weight also depicted large
and dependable genetic variation for its improvement. LS-12 and LS-25 were
found as bold seeded with seed weight of 9 g as against check (7 g). Harvest
index ranged from 12-22 percent indicating large genetic variation. LS-19
was found with maximum harvest index (22%).
Seed yield is an important quantitative trait in linseed. It showed statistically
significant genetic differences. Maximum yield (1439 kg/ha) was recorded in
LS-30 followed by LS-4 (1400 kg) (Table 2). The data indicated that LS-1,
LS-3, LS-11, LS-13, LS-14 and LS-25 showed significant increase in yield i.e
1245, 1339, 1342, 1250, 1352 and 1346 kg per hectare, respectively as
compared to check (1062 kg/ha).
Genetic parameters
The progress of breeding for such economic traits is determined by
magnitude and nature of their genotypic and phenotypic variability. Genotypic
variances for all the characters ranged from 0.3 (branches/plant) to 52360.24
kg (seed yield/ha) (Table 3), while the phenotypic variance estimates ranged
from 0.4 (branches/plant) to 78685.11 kg (seed yield/ha). Similar findings in
linseed were reported by Naqvi et al. (20) and Kadir et al. (11). All the traits
showed higher values of phenotypic variances than their respective genotypic
variances. Falconer and Mackay (8) also observed that genotypic variance is
the variance of genotypic values only while phenotypic variance is the sum of
separate components (V G +V E ).
J. Agric. Res., 2013, 51(2)

Correlation studies in linseed
125
J. Agric. Res., 2013, 51(2)

126
T. Bibi et al.
J. Agric. Res., 2013, 51(2)

Correlation studies in linseed
127
Estimation of coefficient of variation
Genotypic coefficient of variation ranged from 0.939 to 20.251 percent (Table
3) for days to maturity and seed yield per hectare, respectively, whereas
phenotypic coefficient of variations for these traits ranged from 1.029 to
24.825 percent. The highest GCV was observed in seed yield (20.251%)
followed by seed yield per plot (18.664%), harvest index (16.319%) and
capsules per plant (16.068%). The highest PCV values were observed in
seed yield per hectare (24.825%) followed by seed yield per plot (20.441%),
seeds per capsule (18.465%) and capsules per plant (17.792%).
The character with high values of genetic coefficient of variation indicates
high potential for effective selection. These findings are in line with those of
Alam et al. (2) and Kadir et al. (11). On the other hand, characters like seeds
per capsule showed the highest difference between genotypic and phenotypic
coefficient of variation indicating more environmental influences on this
character. The differences between PCV and GCV were less than 1.85 for all
traits except seeds per capsule and seed yield per hectare (Table 3). This
showed less environmental influence in phenotypic variance development.
Similar results were reported by Akbar et al. (1) in linseed.
Estimation of heritability
Heritability values for the traits studied ranged from 0.158 - 0.908 (Table 3).
The highest and significant heritability estimates were found for days to
flower initiation (0.825), days to flower completion (0.809), days to maturity
(0.832), 1000-grain weight (0.862) and capsules per plant (0.816), whereas
heritability estimates for seeds per capsule (0.158) were of low magnitude
and maximum for harvest index (0.908). Khan et al. (14) reported low
heritability for seeds per capsule.
Estimation of genetic advance
Maximum genetic advance was observed for seed yield (384.523) followed
by yield per plot (362.843), capsules per plant (25.939), plant height (13.191)
and seeds per capsule (9.156). Low genetic advance observed for remaining
characters ranged from 0.977 (branches/plant) to 5.234 (harvest index) which
could be attributed to low genotypic coefficient of variation rather than
heritability estimates. Patil et al. (21) also reported similar results in linseed.
The results (Table 3) further showed that plot yield, capsules per plant, plant
height and seed yield per hectare had high and significant heritability
J. Agric. Res., 2013, 51(2)

128
T. Bibi et al.
estimates of 0.834, 0.816, 0.700 and 0.665 coupled with corresponding high
values of genetic advance (362.843, 25.939, 13.191 and 384.523). It could
be concluded that for these traits, the additive gene action was a norm (22).
The characters like harvest index, 1000-grain weight, days to flower initiation
and days to maturity gave high heritability values with low genetic advance
which indicated the control of non-additive gene action. It implies that high
value of heritability is not always an indication of high genetic advance. Akbar
et al. (1) have supported the present findings. One other character i.e. seeds
per capsule revealed low values for heritability and it may not respond
favorably to phenotypic selection.
Correlation coefficients
Correlation studies permit the feasibility of indirect selection, when direct
selection becomes inapplicable due to various causes. The data regarding
genotypic and phenotypic correlation estimates (Tables 4 and 5) depicted
that magnitude of phenotypic correlation coefficient for days to flower
initiation with other traits was higher than that at genotypic level confirming
the reports of Das et al. (6) in soybean.
Days to flower completion showed positive and significant genotypic and
phenotypic correlation with days to maturity. Genotypic correlation coefficient
of plant height with capsules per plant (0.526) and seed yield per plot (0.558)
was positive and significant (Table 4). However, these results do not agree
to findings of Khan et al. (13) in linseed. Phenotypic correlation coefficient of
plant height with capsules per plant (0.583), seed yield per plot (0.614) and
seed yield per hectare (0.497) was positive and significant (Table 5). Alam et
al. (2) also reported positive genotypic and phenotypic association of plant
height with capsules per plant and seed yield.
Genotypic and phenotypic correlation coefficient of capsules per plant with
seed yield per plot (0.705, 0.736) and seed yield per hectare (0.679, 0.691)
was positive and significant. Khan et al. (14), Mahto and Mahto (17), Das et
al. (6) and Kumar and Yadav (16) also reported similar findings.
Genotypic correlation of seed yield per plot showed positive and significant
with seed yield per hectare (0.735). At phenotypic level correlation coefficient
between seed yield per plot and seed yield per hectare was also significant
and positive (0.862). Mirza et al. (19) also reported similar findings.
Genotypic correlation coefficients were lower than corresponding phenotypic
correlation coefficients in maximum character pairs. Genotypic correlation
coefficient of days to flower completion with days to maturity (0.579); plant
J. Agric. Res., 2013, 51(2)

Correlation studies in linseed
129
J. Agric. Res., 2013, 51(2)

130
T. Bibi et al.
height with capsules per plant (0.526) and seed yield per plot (0.558); seed
yield per plot with seed yield per hectare (0.735); capsules per plant with
seed yield per plot (0.701) and seed yield per hectare (0.679) showed
positive and significant association which indicated that selection for these
traits can improve the yield.
The present study concluded that hybridization programme involving
genotypes LS-3, LS-11, LS-14, LS-25 and LS-30 with higher number of
capsules per plant and higher seed yield could generate transgressive
segregates possessing higher seed yield potential provided that selection
would be done on the basis of medium height plants bearing higher number
of capsules per plant and higher seed yield per plot in early segregating
generations.
REFERENCES
1. Akbar, M., T. Mahmood, M. Anwar, M. Ali, M. Shafiq and J. Salim. 2003.
Linseed improvement through genetic variability, correlation and path
coefficient analysis. Int. J. Agric. Biol. 5: 303-305.
2. Alam, M. S., A. R. M. S. Rahman and A. B. M. A. Khair. 1985. Genetic
variability and character association in groundnut. Bangladesh J. Agric.
Res. 10: 9-16.
3. Ali, S., M. A. Cheema, M. A. Wahid, A. Sattar and M. F. Saleem. 2011.
Comparative production potential of linola and linseed under different
nitrogen levels. Crop Environ. 2: 33-36
4. Al-Jibouri., H. A., P. A. Miller and H. F. Robinson. 1958. Genotypic and
environmental variances and covariances in upland cotton cross of
interspecific origin. Agron. J. 50: 633-636.
5. Anon. 2011. Agri. Statistics of Pak. 2010-11. Finance Division,
Economic Advisor’s Wing. Islamabad, Pakistan.
6. Das, M. L., A. Rehman and A. J. Miah. 1989. Correlation, path
coefficient and regression studies in soybean. Bangladesh J. Agric. Res.
14: 27-29.
7. Dudley, I. W. and R.E. Moll. 1969. Interpretation and use of estimates of
h 2 and genetic advance in plant breeding. Crop Sci. 9: 257-262.
8. Falconer, D. S. and T. F. C. Mackay. 1996. Introduction to Quantitative
Genetics. Prod. Longman Malaysia. Pp. 122.
9. Ford, J. H. 1964. The influence of time of flowering on seed
development of flax. Crop Sci. 4: 52-54.
10. Johanson, H. W., H. F. Robinson and R. E. Comstock. 1955. Estimation
of genetic and environmental variability in soybeans. Agron. J. 47: 314-
318.
J. Agric. Res., 2013, 51(2)

Correlation studies in linseed
131
11. Kadir, M., B. K. Biswas, P. Podder, M. S. Alam and S. K. Adhikary.
1996. Genetic parameters, character association and path analysis in
sesame. Bangladesh J. Agric. Res. 21: 37-43.
12. Katiyar, R. P., S. N. Singh and Y. S. Chauhan. 1974. Genetic variability,
heritability and genetic advance of yield and its components in mustard.
Indian J. Agric. Sci., 44: 291-293.
13. Khan, N. I., F. Din, M. N. Khan and M. T. H. Shahid. 1998. Correlation
and path analysis in linseed (Linum usitatissimum L.). J. Agric. Res.
36(2):83-87.
14. Khan, N., M. Akbar, N. Iqbal and S. Rasul. 2000 (b). Bio-agronomic
evaluation of linseed genotypes. Pak. J. Biol. Sci. 3: 1172-1173.
15. Khan, N., N. Iqbal, S. Rasul and S. Fazal. 2000 (a). Factor analysis of
seed yield in linseed. Pak. J. Biol. Sci. 3: 328-329.
16. Kumar, P. R. and T. P. Yadav. 1978. Selection criteria for the seed yield
in Brassica species. 5 th Int. Rapeseed Conf. June, 1978.
17. Mahto, R. N. and J. L. Mahto. 1998. Variability, correlation and path
coefficient analysis in linseed. Birsa J. Res. 10: 25–29.
18. Malik, A. J., M. M. Rajput, B. A. Ansari and S. M. Pragani. 1988. Genetic
gain expected from selection for different quantitative characters in
wheat. Pak. J. Agric. Engg. Vet. Sci. 4: 80-84.
19. Mirza, S. H., D. Nessa and S. Islam. 1996. Genetic studies of
interrelationship between seed yield and its components in linseed.
Bangladesh. J. Bot. 25: 197-201.
20. Naqvi, P. A., M. Rai and A. K. Vasishtha. 1987. Association of different
components of seed and oil in linseed. Indian J. Agric. Sci. 57: 231-236.
21. Patil, V. D., V. G. Makne and P. R. Chopde. 1981. Genetic variability
and character association in intervarietal crosses of linseed. Indian J.
Agric. Sci. 51: 631-633.
22. Payasi, S. K., U. S. Bose and A. K. Singh. 2000. Pooled analysis of
genetic parameters in linseed (Linum usitatissimum L.). Adva. in Pl. Sci.
13: 559-562.
23. Payne, T. J. 2000. Promoting better health with flax seed in bread.
Cereal Foods World. 45: 102-104.
24. Poehlman, J. M. 1987. Breeding Field Crops. AVI Publishing Company,
Inc., Westport, Connecticut, New York.
25. Rao, S. K. and S. P. Singh. 1985. Breeding behaviour and genetic
variation for yield and its components in varietal crosses of linseed.
Indian J. Agric. Res. 9: 111-118.
26. Singh, R. K. and B. D. Chaudhary. 1985. Biometrical Methods in
Quantitative Genetic Analysis. Kalyani Pub., New Delhi, India. Pp. 57.
J. Agric. Res., 2013, 51(2)

132
T. Bibi et al.
27. Steel, R. G. D and J. H. Torrie.1980. Principles and Procedures of
Statistics. A Biometrical Approach, McGraw Hill Book Co., New York,
USA.
28. Yadav, T. P. and T. L. Dalal. 1972. Genetic variability and correlation
studies in linseed (Linum usitatissimum L.). PAU. J. Res. 11: 35-39.
J. Agric. Res., 2013, 51(2)