Genetic relationships among okra ( Abelmoschus esculentus ( L . ) Moench ) cultivars in Nigeria

This study was conducted on okra (Abelmoschus esculentus (L.) Moench) cultivars at the Teaching and Research Farm, University of Maiduguri, Nigeria. The objective was to evaluate the level of genetic divergence and heritability of eight characters in 2015 and 2016 dry seasons using irrigation. The results showed highly significant (p<0.01) differences in the ten okra cultivars for days to anthesis, plant height, fresh capsule length, fresh mass per capsule and fresh capsule diameter across the two years. A high genotypic coefficient of variation, heritability, and genetic advance were detected in all the characters except for days to anthesis and fresh capsule diameter. This implied that different genetic constitution and preponderance of additive effects governed these characters, thus presenting a significant opportunity for selection. The Mahanalobis D analysis allotted the ten cultivars into four clusters. The highest was cluster I comprising four cultivars, followed by cluster II containing three cultivars, cluster III consisting two cultivars, and cluster IV with mono genotypic. The three most superior okra cultivars (Salkade, Y’ar gagure and Kwadag) for yield and related characters could be exploited directly or introgressed with other promising ones in future breeding programs.


INTRODUCTION
Okra (Abelmoschus esculentus (L.) Moench) usually referred to as Lady's finger belongs to the Malvaceae family (Kishor et al., 2016).Okra is proposed to originate from the Tropical Africa from where it extensively spread to Asia, America, Southern Europe and other countries (Muhammad et al., 2013).In 2008, the five top most okra producing countries were Iraq, Nigeria, Togo, Sudan and India (FAOSTAT, 2010).However, Nigeria ranks third in okra amid fruit vegetables based on production and consumption, succeeding pepper and tomato (Ibeawuchi, 2007).The okra local cultivars diverged in growth habits including leaf arrangement and size, fruits branching, height and maturity period.During the vegetative stage, okra growth patterns are similar, but those that were highly vigorous produced improved leaf area and accumulated dry matter (Akanbi et al., 2010).
The collection of desirable plant germplasm relies on the proven accession features and genetic divergence, which are essential in genetic resources utilization (Olaoye et al., 2009;AdeOluwa & Kehinde, 2011).Genetic diversity denotes the variability in different crop species, and its links with accessions identification, which is important in gene bank curators (Bello et al., 2012ab;Bello et al., 2011;Osekitar & Akinyele, 2008).Morphological characterization of plants has been recommended as the first step to be adopted prior to farreaching molecular research and biochemical analyses (Akash et al., 2013).Many researchers reported a substantial morphological degree of variance in the West African okra varieties (Adeniji, et al., 2007;Akanbi et al., 2010;AdeOluwa & Kehinde, 2011).The improvement in plant breeding scheme leans on high genetic differences in the population and the magnitude of inheritance of favorable attributes (Olawuyi et al., 2015, Bello et al., 2014ab).The variability obtained in a population are apportioned into non-heritable and heritable parameters utilizing genetically related components including heritability, the genotypic coefficient of variation and genetic advance, which are the core for selection (Muluken et al., 2016;Seth et al., 2016).The expected response to selection and methods of selection are assessed on high heritability values of the characters.Cluster analysis is one of the powerful tools in determining genetic divergence among varieties of crops.The objective of this study was to evaluate the level of genetic divergence and heritability of ten okra cultivars in the stress-free irrigation conditions of Nigeria Sudan savannah, with the view to devising a breeding strategy for okra selection for further improvement.

MATERIALS AND METHODS
Field experiments were conducted on okra (Abelmoschus esculentus (L.) Moench.)cultivars at the Teaching and Research Farm, University of Maiduguri, Nigeria to evaluate the level of genetic divergence and heritability of eight characters in 2015 and 2016 dry seasons using irrigation.Ten okra cultivars were used; of which six cultivars (Salkade, Yar'gagure, Kwadam, Lai-lai, Yar'duwi and Y'ar kwami.)were obtained from Gagure, Gulani Local Government Area of Yobe State and four (Kwalpuku, Composite Kwadag and Mola Kwadag) were acquired from Borno State Agricultural programme, Maiduguri, Nigeria, and coded as described in Table 1.Kwadam This cultivar has a short and white stem, medium leaves and spiny capsules.
The field experiment was based on Randomized Complete Block Design with three replications.The plot was 216 m 2 , divided into 33 plots of 2 × 2 m with 1 m spacing between replications, and 0.5 m between treatments.Weeding was carried out manually at 3, 6, and 9 weeks after sowing (WAS).A compound fertilizer, N.P.K. 15:15:15 was applied at the rate of 60 kg N/ha in twice, first at three weeks after planting and then at flowering.Two milliliters of Ultracide 40EC insecticide in 15 liters was applied fortnightly to control insect pests.Light watering was applied using a watering can at every morning and afternoon.This was continued for a week for rapid and well establishment of the germinated seedlings.
For the evaluation of the eight studied characters; from each plot, ten (10) plants were randomly chosen.The eight studied characters include fresh capsule length, fresh capsule yield per plant (g), the number of primary branches per plant, days to anthesis, the number of capsules per plant, fresh capsule diameter (cm), fresh mass per capsule (g) and plant height (m).Individual year analysis of variance (ANOVA) was calculated, and then a combined ANOVA across the two years with the use of SAS PROC GLM model (Version 9.2, Volume 1), to determine the mean squares for every character (SAS Institute, 2011).A mixed model of the SAS PROC GLM model was utilized for the ANOVA.
Cultivars were considered as fixed effects, while replication as a random effect.The degree of variation was estimated employing % coefficient of variation p < 0.05.Also, the variations in the character means were computed with the use of Least Significant Difference (LSD).However, genotypic coefficient of variation (GCV) and phenotypic coefficient of variation (PCV) were computed based on the formula proposed by Burton (1952) with the use of PROC GLM model of SAS (SAS Institute, 2011, Version 9.2, Volume 1).The broad sense heritability was determined as recommended by Johnson et al. (1955).The estimate of the expected genetic advance for each character was ascertained using the procedure of Allard (1960).The approach of Comstock & Robinson (1948) was followed for estimating the genetic advance in a percentage of the mean.To evaluate genetic divergences in the ten okra cultivars, the Mahalanobis' D 2 model (Mahalanobis 1936) and its auxiliary analysis were applied following the method of Rao (1952).

RESULTS AND DISCUSSION
The combined analysis of variance revealed exceedingly significant (p < 0.01) differences in the studied okra cultivars for days to anthesis, plant height, fresh capsule length, fresh mass per capsule and fresh capsule diameter across the two years (Table 2).It shows that the genetic parameters of the parental materials were quite dissimilar.This result corroborates with that of several earlier researchers (Akinyele & Osekita, 2006;Nwangburuka et al., 2011;Nwangburuka et al., 2012;AdeOluwa & Kehinde, 2013;Muluken et al., 2016).
Again, the cultivars exhibited significant (P<0.05)differences for fresh capsule yield per plant and the number of capsules per plant.On the other hand, nonsignificant differences in the cultivars for the number of primary branches per plant revealed that the genetic parameters of the okra cultivars were very intact.The ANOVA also showed variations in the studied cultivars for almost all the characters.This variation could be used via selection to improve the okra studied characters.This result is supported by several previous researchers (Düzyaman, 2005;Salesh et al., 2010;Nwangburuka et al., 2012;Hazem et al., 2013;Amoatey et al., 2015).The first-order cultivar × year interaction was significant for all the okra characters.It signified that environment condition affected the wide variation of these characters, and is considered as a key parameter for the yield of crops.
The okra yield potential is essential in producing many capsules per plant which could be attained by means of timely harvesting the fresh fruits to enable the development of more branches.This effort will undeviatingly increase the crop yield.This finding is also in conformity with the previous researchers (Akinyele & Osekita 2006;Mehta et al., 2006;Alade et al., 2008).The mean performance of the studied okra cultivars showed a significant difference in days to anthesis (Table 3).The maximum number of days to anthesis (50.69 days) was recorded for 'P4', while the minimum number of days (42 days) was obtained in 'P10'.The average number of days to anthesis was 46.82 days and five cultivars had higher days to anthesis than the average.It denoted that the assessed cultivars differed morphologically from one other especially on flower bearing habits, similar to the findings of Muluken et al. (2016).Besides, all the studied okra cultivars varied decidedly for plant height with 'P5' being the tallest while P3 cultivar was the shortest.The average plant height in 'P5' and 'P4' was 1.25m which was higher than the overall mean height.The number of primary branches per plant significant differed in the cultivars with 'P1' possessing the greatest number, whereas 'P10' exhibited the least.Subsequently, the greatest number of primary branches was obtained from 'P1', followed by 'P7', 'P5', and 'P4' which were more than the average.It is obvious that the number of primary branches varied significantly at the early growth of okra, as previously reported by Jagan et al. (2013).The average number of capsules per plant recorded was 28.03 with about half of the cultivars accomplishing a greater average.The 'P4' was outstanding with the greatest number of capsules per plant, followed by 'P1'', ''P5', and 'P9'.The highest number of fresh capsule length was attained in 'P1', while 'P6' had the lowest.This variation might be the differences in the number of bearing internodes and plant height in the cultivars.Half of the okra cultivars had an extended fresh capsule length than the average length.Fresh capsule length was reported to vary significantly from one accession to the other since it invariably articulates the distinctiveness (Nwangburuka et al., 2012;Hazem et al., 2013;Amoatey et al., 2015).The 'P2', 'P4', and 'P5' had fresh capsule diameter more than the average, while 40 % of the cultivars produced more than the average fresh capsule mass.The okra capsule length is at variance from one cultivar to the other, perhaps due to differences in days to anthesis and other morphological characters.However, 'P1', 'P4', and 'P5' also possessed capsule yield greater than the average.Variability parameters estimated viz.GCV, PCV, heritability and genetic advance studied characters are depicted in Table 4.The large magnitude of PCV and GCV with a small difference between the two heredity parameters indicated a smaller amount of environmental influence on the phenotypic expression.Muluken et al. (2016) earlier buttressed this extrapolation.The GCV and PCV ranged between 1.1-33.3% and 2.4 %-48.5 %, respectively, for fresh capsule yield and fresh capsule diameter.Several researchers reported the consistent differences of okra cultivars due to cultivars and environmental interactions (Thirupathi et al., 2012;AdeOluwa & Kehinde, 2013;Ehab et al., 2013;Adekoya et al., 2014).This statement also showed the level of productivity in crops, as statistical groupings of the cultivars were believed to be distinguished.Several researchers corroborate these discoveries for West African okra germplasm (Nwangburuka et al., 2012;AdeOluwa & Kehinde, 2013;Adekoya et al., 2014).
The number of primary branches and fresh capsule diameter, on the other hand, were having the lowest estimate of below 10 % for PCV and GCV.This expressed a slight range of difference and hindered possibility for selection of these characters.Furthermore, the least GCV and PCV estimate of characters, implied higher impacts environmental conditions on these characters therefore; selection based on phenotypic basis will not be useful for the genetic progress of the crop (Chaurasia et al., 2011;Bharathiveeramani et al., 2012;Das et al., 2012;Sankara & Pinaki, 2012;Thirupathi et al., 2012;Ehab et al., 2013;Kishor et al., 2016].Contrariwise, days to anthesis possessed medium GCV and PCV values (Chaurasia et al., 2011).It indicated that genetic effects influenced these characters.Therefore, these characters are responsive to selection for onward improvement.
The number of capsules per plant, fruit length, plant height, the number of primary branches, fresh capsule length, fresh capsule yield per plant and fresh mass per capsule exhibited high values more than 20 % for both PCV and GCV with a considerably low degree of variation between the two.This result substantiates with the findings of Ehab et al. (2013).Many researchers nevertheless, noticed that high magnitude of GCV and PCV inferred a low environmental manifestation effects on the characters, which probably increase greater improvement prospects through selection scheme (Salesh et al., 2010;Bharathiveeramani et al., 2012;Nwangburuka et al., 2012;Swati et al., 2014;Kishor et al., 2016).Thus, selections of favorable characters by utilizing high phenotypic and genotypic estimates could be exploited in enhancing the characters during the breeding program.
Broad sense heritability estimates ranged from 25.84 % for the number of capsules per plant to 93.84 % for fresh capsule yield per plant (Table 4).As described by Robinson et al. (1955), heritability is categorized as least with a range of 0-30 %, fair (31-60 %) and best (> 60 %).In the present research, a broad sense heritability of greater than 60 % was obtained for capsule yield, plant height, days to anthesis, capsule length, capsule diameter, and capsule mass.These agronomic characters seemed to respond effectively to the pressure of selection.Whenever heritability is up to 80 % or more of a character, selection would be easy for such character.Thus, selection for all these characters might result in an increase in capsule yield of okra.A great heritability also showed a great genetic base.A close association between the phenotype and cultivar could be the cause of small environmental interplay conditions (Jagan et al., 2013;Muluken et al., 2016).Fairly broad sense heritability estimate was observed for the number of primary branches.The least heritability value, however, was obtained for the number of capsules per plant.This alluded that these cultivars may not be improved via direct selection.Whenever a character is of a range between medium and high heritability, a selection due to specific performance can allow rapid progress.Fairly heritability implied improvement via selection.The least heritability also indicated ineffectual direct selection for the advancement of the characters owing to environmental masking effects (Nwangburuka et al., 2012;Bello & Olawuyi, 2015;Muluken et al., 2016).
As described by Johnson et al. (1955), genetic advance as percent mean were categorized as high (≥20 %), moderate (10 ≤ 20 %) and low (0 ≤ 10 %).Based on this ranking, the number of capsules per plant, the number of primary branches, capsule length, plant height, capsule yield per plant and capsule mass possessed the genetic advance of greater or equal to 20 % (Table 4).This revealed the predominance of additive genetic effects for these characters.Capsule diameter and days to anthesis showed low and moderate genetic advance, respectively.Appropriately, this result depicted that expected progress from the selection of the cultivars is between 16.1 % (days to anthesis) and 51.8 % (capsule yield).This substantiates with the findings of Olawuyi et al. (2015) and Hazem et al. (2013).
The relative amount of heritable variability is not enough to determine the GCV only, except with the aid of heritability and genetic advance.The high estimates of heritability coupled with genetic advance offered sufficient information on each character and indicated a genotypic response to selection (Pradip et al., 2010;Sibsankar et al., 2012).High heritability and genetic advance were observed for all the characters studied except days to anthesis and fresh capsule diameter (Table 4).This showed that differences in the genetic background would enable great opportunity for selection.Furthermore, this demonstrated the preponderance of additive gene effects for these characters, instead of the environmental influences.Thus, selection can be made based on the phenotypic expressions of okra characters for the improvement of yield (Muluken et al., 2016).As moderate heritability and high genetic advance were noted for the number of primary branches of okra, low heritability and high genetic advance estimates were detected for the number of capsules per plant.This could also be on high environmental influences controlling the expression of the characters.This, therefore, possibly hinders the opportunity for selection for crop improvement due to the prevalence of non-additive (dominant and/or epistatic) or non-fixable effects.Jagan et al. (2013) earlier noticed these findings.In a condition of low heritability and genetic advance for the characters, unique approaches such as hybridization and recurrent selection should be adopted (Bozokalfa et al., 2013;Jagan et al., 2013).The parameters of four different Clusters, respective cultivars, and their numbers are presented in Table 5.
Cluster I possessed the highest with 4 cultivars followed by Cluster II with 3 cultivars and Cluster III with 2 cultivars, whereas Cluster IV had one cultivar (mono genotypic).The highly varied cultivars, P7, P8, P9, and P10 were obtained from Cluster I.These clusters outline showed that geographical variation had an indirect association with genetic diversity.Genetic diversity in okra germplasm with the use of cluster analysis had earlier been reported by several researchers (Akotkar et al., 2010;Das et al., 2012;Umrao et al., 2014;Seth et al., 2016).In general, the cultivars distribution patterns from diverse geographical regions into discrete clusters were at random.This result might be attributed to the free exchange and recurrent genetic constitution efforts by the farmers and plant breeders of the diverse agroecological zones in Nigeria.Besides, the disparity of selection pressure owing to regional okra favorites could improve the resemblance of the cultivars.Lack of relationship between the genetic diversity and geographical distance indicated that forces like natural and artificial selection, the exchange of genetic material, genetic drift and spontaneous mutation could lead to genetic diversity instead of geographical origin (Pradip et al., 2010;Seth et al., 2016).Therefore, selection for outcrossing of okra cultivars ought to base on genetic diversity instead of geographic diversity.Widely varied inter-cluster distances in the four clusters of okra cultivars are shown in Table 6.The inter-cluster distances were higher than the intra-cluster distances, signifying high genetic diversity among the cultivars of the varied groups.The intra-cluster distance in the ten studied cultivars indicated a low value between cluster IV and I (4.57), revealing similar relationships in the cultivars of these clusters (Pradip et al., 2010).The greatest intra-cluster estimate was observed between clusters III and I followed by between clusters IV and III.This depicts that the cultivars in these clusters diverged greatly.Therefore, outcrossing the cultivars featuring in these clusters could enhance transgressive segregations and generation progress.This has been as earlier opined by Umrao et al. (2014) and Seth et al. (2016).Conclusively, an involvement of highest yielding cultivars (Salkade, Y'ar gagure and Kwadag) could be exploited directly or hybridized to enhance novel recombinants and exploit transgressive segregates with high genetic yield potentials.

Table 1 :
The morphological descriptions of ten okra cultivars studied

Table 2 :
Combined ANOVA for eight studied okra characters in Maiduguri, Nigeria across two years

Table 3 :
Mean performance for eight studied okra characters in Maiduguri, Nigeria across two years

Table 4 :
Estimates of genotypic coefficient of variation (GCV), phenotypic coefficient of variation (PCV) and environmental coefficient of variation (ECV) for eight studied okra characters in Maiduguri, Nigeria across two years

Table 5 :
Clustering form of 10 okra cultivars by Tocher's method

Table 6 :
Mean intra (bold) and inter-cluster D2 values of 4 clusters for 10 okra cultivars formed by Torcher's method