The objectives of the study were to investigate the genetic behavior of some biological and economical traits of 14 okra populations collected from Dakahlia Governorate, which underwent two cycles of inbreeding with selection. Selection of individual plants based on earliness, high number of pods, and minimum neck/pod ratio was carried out in all generations. The results showed that the means and ranges of all studied traits for all families became smaller in the S2 generation than those in the S0 generation. Highly significant variations were observed among populations for all the studied traits. The mean performance clearly indicated the agronomic superiority of some families over the others. Family 9 followed by family 12 showed the earliest flowering plants and the highest yield per plant. Phenotypic variances were higher than the corresponding genotypic variances indicating predominance of environmental effects on the expression of these characters. The magnitude of phenotypic and genotypic coefficients of variation varied from one trait to another. High broad-sense heritability coupled with high genetic advance as percent of mean were shown by the different traits, especially, plant height, number of branches per plant, number of pods per plant, pod length, neck/pod ratio and plant yield. This implicates that these traits were under the control of additive genetic effects, and could be effectively improved through selection. Plant yield had positive and highly significant correlation at genotypic and phenotypic level with number of pods per plant, plant height and neck/pod ratio.
The most widely produced okra cultivars in Egypt are known as Balady. Mostly, this cultivar is cultivated based on local open-pollinated seeds, which are maintained by farmers, produced for self-consumption and sold at local market.
Various okra cultivars show a lot of variability in many characters, such as yield, days to first flower appearance, number of pods per plant and plant height (Abo El-Khar 2003; Masoud
Inbreeding with selection was very sufficient in recovering desirable families from okra. In this respect, Hussein (1994), Moualla
Therefore, this study was conducted to select superior families from local okra populations (collected from different locations in Dakahlia Governorate) through inbreeding and selection programs.
This investigation was carried out at El-Baramoon Experimental Farm, Dakahlia Governorate (latitude 31°04′31″N, longitude 31°37′67″E and altitude 19 m above sea level) during the three summer seasons of 2010, 2011 and 2012. The soil texture at the experimental site is clay-loam.
A total of 14 mature pods were obtained from 14 okra populations collected from different locations in Dakahlia Governorate. Seeds were separately extracted from each pod. Each genotype was cultivated on April 5, 2010 in two rows. The row was 5 m long and 70 cm wide with 30 cm in-row spacing. The seedlings were thinned out to only one plant per hill. The culture practices were applied as recommended for okra production.
Plants were self-pollinated to produce S1 seeds. The following traits were recorded in selected individual plants: days to first flower appearance, number of pods per plant, pod length (cm), pod diameter (cm), neck/pod ratio (expressed as neck weight/pod weight X 100) and pod yield per plant (g), moreover, plant height (cm) and number of branches per plants were recorded at 90 days after planting.
Days to first flower appearance, neck/pod ratio and pod yield per plant were used as the basis of selection of plants in all generations. The best plant from each family was selected and kept separately. The selected families were denoted (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13 and 14).
In 2011 season, seeds of selected plants after first cycle (S1) were separately sown on March 28. The cultural practices were conducted similarly as the first season. The 14 selected families were self-pollinated to continue the inbreeding program. The plants were selected from each family to produce S2 families.
In 2012 season, all S2 selected families were sown on March 26 using a randomized complete block design with three replications. Each experimental unit contained three rows, 5 m long and 70 cm wide, while plants were 30 cm apart within each row.
The field procedures were the same as in the first and second seasons. A random sample of ten plants from each plot was used for taking observations on the eight above-mentioned traits as in the 2010 season.
The data were statistically analyzed according to Snedecor and Cochran (1982). Comparisons among means of families were tested using LSD values at 5% and 1% levels.
The components of variance were computed using the observed mean square values as outlined by Johnson
Broad-sense heritability (h2B) was calculated according to Allard (1999) as the ratio of the genotypic variance (
Expected genetic advance after one generation of selection (GA) and GA as percentage of the mean assuming selection of the superior 5% of the genotypes were estimated according to the formulae given by Johnson
where K is the selection differential (2.06 for selecting 5% of the genotypes).
In order to estimate the genotypic and phenotypic correlations between pairs of traits, a covariance analysis was made between all possible pairs of studied traits, and they were calculated from the following equations as outline by Singh and Choudhary (1979):
Cov g1g2 = the genotypic correlation between any pairs of traits.
Cov ph1ph2 = the phenotypic correlation between any pairs of traits.
σ2g1 and σ2g2 are the genotypic variance of the first and second trait, respectively.
σ2ph1 and σ2ph2 are the phenotypic variance of the first and second trait, respectively.
The significance of the rg and rph were tested with “t” test as described by Cochran and Cox (1957).
The results of the selection program for 14 families were recorded in Tables 1 and 2. These tables represent the means and ranges for each selected generation for each of the 14 families. The mean performance of selected families show a remarkable change in all studied traits in S2 generation compared in S0 generation. A decrease in mean values of all studied traits were observed from S0 to S2 generation. At the same time, the results show that the ranges of all studied traits for all selected families became smaller in the S2 generation than those in the S0 generation.
Moreover, the analysis of variance for the studied traits showed that the differences among genotypes were highly significant for all studied traits (Table 3).
The means of the 14 selected families of the second cycle for all studied traits are presented in Table 4. Obtained results clearly indicate that all 14 selected families in the second cycle exhibited highly significant differences for all studied traits. The mean performance showed a clear indication of agronomic superiority of some families over the other. Data revealed that plants in families 12 and 9 were the earliest to flower (65.2 days) and (67.0 days), respectively. On the other hand, plants in family 5 were the latest ones to flower (78.3 days) compared with the other families. Regarding to plant height, the tallest plants (150.9 cm) belong to family 12 whereas, family 11 plants exhibited the shortest growth (97.1cm). For number of branches per plant, family 1 showed the profuse plants (9.14) whereas the family 4 possessed the lowest branched plants (3.32). For number of pods per plant, family 1 possessed the highest values for number of pods (91.8) among the 14 genotypes. On the other hand, family 5 possessed the lowest value (56.1). Results also showed that family 2 possessed the highest value for pod length (3.8 cm), but family 5 possessed the lowest value (2.3 cm). Family 12 had maximum pod diameter (2.1 cm), whereas, family 1 had the lowest mean (1.3 cm). Family 11 had maximum neck/pod ratio (30.04), while family 5 had the lowest mean (18.98). For yield per plant families 9 and 12 exhibited the highest value of 403.8 and 394.1 g/plant, respectively. On the other hand, family 5 exhibited the lowest value (255.5 g/plant) compared with the other families.
The data presented in Table 5 showed that the genotypic and phenotypic estimated variances of all traits being studied appeared large, in comparison with the estimated values of error variance. Furthermore, phenotypic variances were higher than the corresponding genotypic variances.
Table 5 data set also revealed that the magnitude of phenotypic and genotypic coefficients of variation varied from one trait to another. The phenotypic coefficient of variation (PCV) was higher than genotypic coefficient of variation (GCV) for all studied traits. In particular, a higher PCV and GCV estimates were found for number of branches per plant. The moderate PCV and GCV estimates were found for pod diameter (16.43, 13.12), number of pods per plant (16.37, 15.84), pod length (14.73, 13.83), plant height (14.25, 14.02), plant yield (13.95, 13.25) and neck/pod ratio (12.97, 12.89). Minimum values of phenotypic coefficient of variation (PCV) and genotypic coefficient of variation (GCV) were recorded for days to first flower appearance (5.15, 4.71). A narrow range of difference between PCV and GCV was recorded for days to first flower appearance, plant height, number of pods per plant, pod length, neck/pod ratio and plant yield (Table 5). On the contrary, a wide difference between PCV and GCV was observed for pod diameter and number of branches per plant.
The broad-sense heritability estimates were generally high for all the characters under study except for pod diameter which registered moderate value (0.64) (Table 5). High broad-sense heritability coupled with high genetic advance as percent of mean were shown by the different traits, especially, plant height, number of branches per plant, number of pods per plant, pod length, neck/pod ratio and plant yield (Table 5). Pod diameter showed moderately high heritability (0.64) with high genetic advance as percent of mean (21.59). High heritability (0.83) but low genetic advance as percent of mean (8.86) was noted for days to first flower appearance trait.
Genotypic and phenotypic correlations for all possible combinations for traits under study are presented in Table 6. The results clearly show that the magnitudes of the genotypic correlations were almost similar or very close to the corresponding phenotypic correlation.
Plant yield had positive and highly significant correlation at genotypic and phenotypic level with number of pods per plant, plant height and neck/pod ratio (Table 6).
Days to first flower appearance showed negative and significant association with plant height, number of pods per plant, pod diameter and plant yield at genotypic and phenotypic level (Table 6).
Plant height was positively and significantly correlated with number of pods per plant and pod diameter while it had negative and highly significant association with number of branches per plant at genotypic and phenotypic level (Table 6).
Negative and significant correlations were observed between number of branches per plant and pod diameter at genotypic and phenotypic level. Number of pods per plant showed positive and significant correlation with plant height and neck/pod ratio at genotypic and phenotypic level. Pod length was found to be positively and significantly correlated with pod diameter (Table 6).
The narrow ranges of values which were noticed in the S2 generation indicate that all studied traits reached a certain degree of uniformity and less degree of variability due to inbreeding and direct selection. These results are in agreement with those obtained by Hussein (1994), Moualla
The highly significant differences detected among the means of the 14 selected families of the second cycle for all studied traits indicated that there was a wide range of variation among the studied genotypes for all studied traits which provides an opportunity for selecting suitable genotypes with better performance for the traits. This result also implied that these populations of okra genotypes would respond positively to selection. Similar results were obtained by Martinello
For all the studied traits, the genotypic and phenotypic estimated variances appeared large, in comparison with the estimated values of error variance; such result seemed to indicate that the number of replicates used in the evaluation experiment of these genotypes were adequate to give a better estimation for the error variance. Furthermore, phenotypic variances were higher than the corresponding genotypic variances indicating predominance of environmental effects on the expression of these characters. Moreover, the genotypic variance contributed a major proportion of total variance in all the studied traits suggesting that these traits were under the genetic control. The present findings were in conformity with the reports of AdeOluwa and Kehinde (2011). The estimates of phenotypic coefficient of variation (PCV) in general, were higher than the estimates of genotypic coefficient of variation (GCV) for all the characters, which suggested that the apparent variation is not only due to the genotypes but also due to the influence of environment. Number of branches per plant showed high PCV and GCV estimates. While, the characters
The high estimates of heritability for all the characters under study except for pod diameter which registered moderate value suggest the feasibility of selection for these traits. In this respect, Khanorkar and Kathiria (2010) reported that the higher values of narrow-sense heritability for a particular character indicated that it is controlled largely by genes acting in an additive effect. Thus, if heritability is high for a trait, the plant breeder can go for selection of individuals or group of individuals. In crops like okra high narrow-sense heritability estimates may be helpful for the development of improved varieties. These results are in close conformity with the findings of Bendale
However selection should be made very carefully as heritability is measured in broad-sense, which may be influent. High heritability does not mean a high genetic advance for a particular quantitative character. Johnson
High broad-sense heritability coupled with high genetic advance as percent of mean for plant height, number of branches per plant, number of pods per plant, pod length, neck/pod ratio and plant yield, deserve greater attention in future breeding programs for developing better okra, and it is suggested that pedigree phenotypic selection method is a useful breeding program for improving these traits. A similar findings were reported by Jindal
The inheritance of quantitative traits is often influenced by variation in other traits, which may be due to genetic linkage or pleiotropy. So, Estimation of genotypic and phenotypic correlations among traits is necessary in plant breeding. A positive correlation between desirable characters is valuable to the plant breeder because it helps in determining the extent of improvement that could be brought in the characters and also in selecting suitable genotypes. In the present study, the magnitudes of the genotypic correlations were almost similar or very close to the corresponding phenotypic correlation. These results were expected, since the magnitude of the error covariance was relatively small compared with the respective values of genotypic covariance. The results emphasize that selection based on number of pods per plant, plant height or neck/pod ratio will be essential enough in improving plant yield. The results are in line with the findings of Ahmed (2001), Magar and Madrap (2009), Ramya and Senthilkumar (2009), Rashwan (2011) and El-Gendy (2012). Moreover, selection for early flowering resulted in an increased number of pods per plant, ultimately led to an increased yield. Similar results were obtained by Ahmed (2001), Ramya and Senthilkumar (2009), Rashwan (2011) and Simon
In conclusion, our study demonstrated that inbreeding with selection program is very efficient in improving the yield and yield component traits of okra. Selection based on number of pods per plant, plant height and neck/pod ratio is essential enough to effectively improve the yield of okra.
The means of all studied traits for 14 selected families of okra at S0, S1 and S2 generations.
|Families||Days to first flower appearance||Plant height (cm)||No. branches||No pods/plant||Pod length (cm)||Pod diameter (cm)||Neck/pod ratio||Plant yield (g)|
The ranges of all studied traits for 14 selected families of okra at S0, S1 and S2 generations.
|Families||Days to first flower appearance||Plant height (cm)||No. branches||No pods/plant||Pod length (cm)||Pod diameter (cm)||Neck/pod ratio||Plant yield (g)|
Analysis of variance for all studied traits in 14 selected families of okra after two cycles of pedigree selection.
|Source of variation||df||Days to first flower appearance||Plant height (cm)||No. branches||No pods/plant||Pod length (cm)||Pod diameter (cm)||Neck/pod ratio||Plant yield (g)|
Mean performance of 14 selected families of okra after two cycles of pedigree selection for studied traits.
|Families||Days to first flower appearance||Plant height (cm)||No. branches||No. pods/plant||Pod length (cm)||Pod diameter (cm)||Neck/pod ratio||Plant yield (g)|
Genetic estimates of all studied traits in 14 selected families of okra after two cycles of inbreeding with selection.
|Traits||Variance||Coefficient of variation||Heritability ||GA*|
|Days to first flower appearance||11.49||13.77||2.27||4.71||5.15||0.83|
|Plant height (cm)||303.05||312.94||9.89||14.02||14.25||0.97|
|Pod length (cm)||0.193||0.219||0.026||13.83||14.73||0.88|
|Pod diameter (cm)||0.046||0.072||0.026||13.12||16.43||0.64|
|Plant yield (g)||1896.89||2102.60||205.71||13.25||13.95||0.90|
*GA = Genetic advance as percent of mean
Estimates of genotypic (rg) and phenotypic (rph) correlations among all studied traits of S2 okra.
|Traits||Plant height (cm)||No. branches||No. pods/plant||Pod length (cm)||Pod diameter (cm)||Neck/pod ratio||Plant yield (g)|
|Days to first flower appearance||rg||−0.31*||0.41**||−0.45**||−0.35*||−0.70**||−0.30*||−0.73**|
|rph||−0.33 *||0.24ns||−0.47 **||−0.19ns||−0.67 **||−0.30 *||−0.73 **|
|Plant height (cm)||rg||−0.60**||0.39**||−0.05ns||0.46**||−0.12ns||0.36*|
|rph||−0.48 **||0.41 **||−0.09ns||0.44 **||−0.10ns||0.39 **|
|No. branches||rg||0.09 ns||0.30*||−0.62**||0.014ns||−0.08ns|
|rph||0.16 ns||0.14ns||−0.31 *||0.04ns||0.04ns|
|Pod length (cm)||rg||0.61**||0.03ns||0.32*|
|Pod diameter (cm)||rg||−0.09ns||0.23ns|