Inheritance and production of multiple small fruits per node, in Abelmoschus species, to meet consumer’s demand, in the West African region

Inheritance and fruit production studies were carried out with an Abelmoschus esculentus cultivar, Okpa mkpe (P1), which expresses a mutant trait that produces multiple small fruits per node (msfpn) and two elite A. esculentus cultivars, Awgu early (P2) and Mpi ele (P3), as well as a high yielding A. callei cultivar, Ogolo (P4); all of which produce the conventional solitary fruit per node (sofpn). The aim was to meet consumer’s need for small sized fresh marketable okra fruits in the region. The inheritance studies showed that the mutant trait was controlled by a pair of dominant genes. The fruit morphometric studies showed that the A. esculentus cultivars and their hybrids differed significantly in length of fruit (LOF), diameter of fruit (DOF), circumference of fruit (COF), area of fruit (AOF) and volume of fruit (VOF). The direct cross of the mutant parent to the two elite A. esculentus cultivars showed that all the F1 hybrids produced small sized fruits. On the average, multiple small sized marketable fruits were produced on 61% of fruit-bearing nodes of P1, 58% of P1 x P2 hybrid and 52% of P1 x P3 hybrid fruitbearing nodes, respectively. The percentage reduction in fruit parameters of the msfpn fruits in comparison with the sofpn fruits ranged from 14.28 to 23% for LOF, 10.47 to 15.12% for DOF and 11.18 to 18.89% for COF fruit parameters. Attempts to cross the msfpn parent (P1) with an A. callei (late okra) elite cultivar, P4, proved inconclusive. The discovery of the msfpn trait on an A. callei cultivar, Ojo ogwu, creates a possible opportunity for transferring the trait among A. callei cultivars. It is concluded that exploitation of the msfpn mutant trait could result to meeting okra consumers’ need for small sized fruits in the region and pave the way for exporting small sized fruits; which are in great demand by okra canning industries overseas.


INTRODUCTION
The two edible okra, A. esculentus (early okra) and A. callei (late okra), which are characterized by the production of solitary flowers at the nodes, are short day plants with critical day lengths (CDL) of 12½ and 12¼ h respectively.The implication of their different CDL is that while early okra can flower at any time of the year in Nigeria, late okra only starts flowering around September when day length shortens considerably (Njoku, 1958;Oyolu, 1977;Nwoke, 1980;Siemonsma, 1982).
The fact that Africa is held as the centre of origin of okra (Purseglove, 1968;Karakoltsidis and Constantinides, 1975;Martin and Ruberte, 1978;Vickery and Vickery, 1979), notwithstanding, not much has been done to improve okra cultivars, especially in the West African region, to meet consumers demand.High yielding exotic okra introduced into the region fail to meet local consumer's needs; hence in situ improvement efforts becomes a research imperative (Okonkwo, 1981).Similar challenges have been documented in some okra producing regions of the world.Abdelmageed et al. (2012), reported that despite the increase in okra and other vegetable production in Turkey, yet the production does not meet the demand of the population.Consequently, any attempt to increase productivity like seeking better cultivars than those presently grown would certainly be of value.
Okra consumers in the region prefer small sized fruits necessitating research effort in this direction.Fatokun and Oken'ova (1979) observed that fruit size is an important yield component because an average Nigerian consumer prefers small-sized fruits.Also the processing and packaging industries in the developed countries, where okra is consumed, prefer small podded varieties (Sistrunk et al., 1960).Incidentally, in the region, emphasis had been placed on increasing yield by breeding for larger fruits rather than breeding for higher number of small sized fruits per plant.
A mutant early okra type that bore msfpn was isolated from our pool of local okra germplasm during screening.The (msfpn) mutant isolated looked quite different from the supernumerary inflorescence mutant described by Fatokun et al. (1979), because it produces normal shaped small sized fruits directly from the axil of the node and not as an extension of the petiole.Additionally the report of one-year old A. callei cultivars, Ojo Ogwu, which produces msfpn on sympodial-like branches on its trunk, is new to the author's best of knowledge.Since Fatokun et al. (1979) reported the existence of supernumerary inflorescence in A. esculentus, scarcely any follow-up research work is known to have been directed at exploring the potentialities of exploiting this mutation for the improvement of fresh okra fruit production in the region.Reawakening research interest in this neglected but important mutant trait (which is suspected to have more undiscovered variants) which can boost the production of small sized okra fruit in the region is a long overdue step in the right direction.
This present study reports on the inheritance pattern of this mutant trait and the production of small sized hybrid fruits, through crosses, between the mutant parent and the two elite A. esculentus and one A. callei cultivars.It Udengwu 1685 equally reports for the first time, to the author's best knowledge, the occurrence of this mutant trait in A. callei.

Inheritance of msfpn mutant trait
Three local A. esculentus cultivars P1, P2, P3 (Table 1 and Figures 1  to 3) and one local A. callei cultivar, P4 (Table 1 and Figure 4), were used in the separate crosses.Three pre-germinated seeds of each of the three parents were sown in medium sized black polythene bags measuring 12cm in diameter and 25 cm deep, filled with a mixture of top garden soil, poultry manure and river sand in the ratio of (3:1:1).The bags were placed on wooden benches in the screen house in the Botanic garden, University of Nigeria, Nsukka.Ten days after the emergence of the two juvenile leaves, the seedlings were thinned down to one plant per bag.Watering was done twice daily, morning and evening.Mature flowers were emasculated very early each morning when they were due to open, adopting nondestructive emasculation technique (NDET), as described by Udengwu (2007).Table 2 gives the details of the direct and reciprocal crosses.
When the hybrid seeds were dry they were harvested and stored in desiccators, using anhydrous Calcium chloride pellets as dehydrant.Three pre-germinated seeds, from the stored F1 hybrid seeds, as well as the parental seeds, were planted per stand, in holes about 1 cm deep in the Botanic garden, on flat beds measuring 3 x 3 m with 15 cm as the within row spacing and 30 cm as the between row spacing, following standard cultural practices.The hybrids were completely randomized and there were three replicates with the three parents serving as guard rows.Chicken manure was applied at the rate of 10.75 kg per plot.Ten days after the emergence of the two juvenile leaves, the seedlings were thinned down to one per stand.The plants were rain fed.When the plants were in bloom, the hybrid plants were selfed to obtain the F2 seeds.The F1 hybrid plants were also backcrossed to the recessive parents to study the segregation in the backcross generations (Table 3).The selfed and backcross flowers were tagged.
The beds for the backcross were similar to that of the F1 while that for the F2 measured 10 x 2 m with plant spacing similar to that of the F1.There were three replicates in a randomized complete block design.When the first fruits produced attained the age of 9 days, counts were made with respect to expression of the msfpn mutant trait, treating the replicates as a single population.Chisquare test was used in the analyses of the data.However as a result of the small population size of the experimental plants, Yates correction formula for small numbers and continuity, according to Steel and Torrie (1960) was used.The formula used was:

Expression of msfpn mutant trait in hybrid plants and fruit size studies
One hundred and twenty seeds of each of P1, P2, P3, F1 of P1 x P2 and F1 of P1 x P3, from the inheritance studies, were soaked in tap water overnight and all the seeds that floated were discarded.All the plantings were done on flat beds measuring 4 x 3.0 m.The cultural practices were similar to the one used for the inheritance of msfpn mutant trait.There were 10 experimental plants per row giving rise to 50 plants per block in a randomized complete block design.The plants were sprayed weekly with Vetox 85 for the control of both the leaf and fruit borers.The plants were rain-fed throughout the period of the studies.Harvesting of fresh marketable   fruits from the parents and the two hybrids was carried out every 5 days.As the flowers opened and formed fruits they were tagged to distinguish the expression of msfpn fruits from the sofpn fruits.Number of nodes that expressed the msfpn mutant trait was counted from 30 randomly and evenly selected plants from all the blocks and the mean expression of the trait was determined for each of the germplasm.Harvesting of fresh fruits for morphometric measurements were restricted to 6 day old fruits.Thirty fruits randomly and evenly selected from the three blocks were used for each of the measurements for all the germplasm.ANOVA was used for the analysis of collected data, while LSD (5%) was used to separate the means.

RESULTS
Inheritance pattern of the msfpn trait and number of genes governing the trait.
The results of the various crosses and their tests for significance, using the modified Chi-square statistic, have been summarized (Table 6).In the cross between P 1 and P 2 , the F 1 showed that all the plants produced msfpn fruits for both the direct and reciprocal crosses.The segregation of the F 2 plants showed that for the direct cross out of the 349 plants raised, 250 plants produced msfpn flowers while 98 produced sofpn inflorescence.For the reciprocal cross, out of a total of 385 plants raised, 272 produced msfpn flowers while 113 produced sofpn inflorescence.This is in agreement with a ratio of 3 msfpn flowers: 1 sofpn inflorescence.The backcross of the msfpn flower F 1 parents to the sofpn inflorescence parent produced 256 msfpn flower producing offspring and 225 sofpn with sofpn inflorescence; for the direct cross.The backcross involving the reciprocal cross produced 228 plants with msfpn flowers and 198 plants with sofpn inflorescence.Both backcrosses produced a ratio of 1 msfpn flower: 1 sofpn inflorescence.
For the second cross P 1 x P 3 , the F 1 showed that all the hybrid plants produced exhibited msfpn flowering pattern.The segregation of the F 2 plants showed that for the direct cross, out of the 341 plants raised, 243 were of the msfpn flower type while 98 were of the sofpn inflorescence type.For the reciprocal cross out of the 303 plants raised in the F 2 , 215 produced msfpn flowers while   type while 212 were of the sofpn inflorescence type.These two results also conformed to a ratio of 1 msfpn flower type: 1 sofpn inflorescence type.
In the third cross P 1 x P 4 and its reciprocal P 4 x P 1, all the F 1 produced plants that expressed the msfpn mutant trait.To produce the F 2 plants, the F 1 selfed plants gave very poor germination rates which necessitated termination of the studies.

Expression of msfpn mutant trait and fruit size studies
The results showed that the four cultivars used for the studies-P 1 , P 2 , P 3 and P 4 differed in their fruit bearing characteristics (Table 1; Figures 1 to 4).The five fruit morphometric characteristics, LOF, DOF, COF, AOF and VOF measured for the four parents as well as the hybrids( P 1 x P 2 and P 1 x P 3 ) differed significantly from each other(LSD .05)(Table5).The ANOVA for the five fruit morphometrics (Table 6) showed that very highly significant differences were recorded for the five characteristics studied for both the parents and the hybrids.
Table 7 shows the mean expressions of the msfpn mutant trait on P 1 , the mutant parent and the direct cross hybrids-P 1 x P 2 and P 1 x P 3 .The percentage expression of the trait on the mutant parent was 61%, while it was 58% in the direct P 1 x P 2 hybrid and 52% in the hybrid, P 1 x P 3 .The details of the comparison of the fruit size of    the msfpn parent plant with those of sofpn parents are presented (Table 8).Reduction in fruit size due to the msfpn mutant fruits produced on the parent plant was 23.70% for LOF.For DOF it was 15.12% while it was 18.89% for COF.For the direct cross hybrid (P 1 x P 2 ), the reduction in fruit parameters were 17.62%, 11.02% and 11.42% for LOF, DOF and COF respectively.In the second direct hybrid cross (P 1 x P 3 ), the reductions were 14.28%, 10.47% and 11.18% for LOF, DOF and COF, respectively.For the F 1 hybrid of the third direct cross, P 1 x P 4 , only 20% of the seeds germinated while for the reciprocal cross, P 4 x P 1 , the germination rate was merely 5%.The very few F 1 plants that were raised, all expressed the msfpn mutant flowering pattern, though the few seeds they produced were generally small and without embryos.For the F 2 generation, reduced size of the seeds coupled with very poor seed germination necessitated discontinuation of the studies.
Towards the end of this study, the msfpn mutant trait was identified on the fat trunk of a one year old A. callei cultivar, Ojo Ogwu (Figs. 5a and b), which had lost all its leaves, with signs of the branches drying up; while the trunk still appeared fresh, with new small leaves forming around.The plant was found growing wild and alone in a relatively moist soil in December, when relative humidity is known to be at its lowest level in the area.This is probably the first report of this mutant trait on late or West African, okra.The germplasm of this isolated A. callei cultivar has been preserved in our seed bank.Its full study and characterization is currently going on and will form part of a separate report.

DISCUSSION
The results of the inheritance studies suggests that a pair of genes dominant to the normal genes governing solitary flower production governs the inheritance of this msfpn mutant trait in early okra.Fatokun et al. (1979) had studied the inheritance of supernumerary inflorescence in A. esculentus and reported that it was governed by a pair of genes (SiSi) which were dominant over the genes governing the conventional solitary fruiting habit (s i s i) .Incidentally the mutant type described by Fatokun et al. (1979) produced more than one fruit as an extension of the petiole.Some of the fruits produced showed distorted features which may compromise their ready acceptance by okra consumers.On the other hand the msfpn mutant reported in this study consistently formed normal shaped fruits from the axils of the leaves at each of the multiple fruit bearing nodes.As the older initiated fruits were harvested, the younger ones developed.With regular harvesting of developed fruits at the age of six days, all the initiated fruits could be harvested, resulting to the production of increased number of smaller sized fruits.However, without regular harvesting, most of the other initiated flower buds fail to form new fruits and abscission follows.
The genes controlling the msfpn mutant trait may be identical to the ones already described by Fatokun et al. (1979) or they could be allelic to it.One is more inclined to suggest that a multiple allelic set of genes may be involved in the inheritance of this trait.This will become obvious when many more mutant types are isolated and multiple crosses are carried out among them.It is interesting that this mutant trait is governed by a pair of Mendelian genes which is characteristic of many other genes controlling other traits in okra which have been reported to be simply inherited.Martin et al. (1981) noted that a surprisingly large number of characteristics in okra are inherited in a simple fashion with high heritabilities; which suggested that they are controlled by relatively few genes.In their opinion the large chromosome number of okra provides an excellent opportunity for very wide recombination.Obviously hybrid plants expressing this mutant trait could be grown by okra farmers.With the production of small sized fruits, consumer's demand can easily be met and the door opened for export since canning industries overseas equally prefer small sized fruits (Sistrunk et al., 1960).Reduction in the size of fruits produced on the nodes expressing the mutant trait on both the mutant parents as well as the hybrids could be attributed to competition for assimilates by many fruits on a node.Kress (1981) and Stephenson (1981) observed that resource limitations as well as sibling competition are responsible for the reduced fruit size in Angiosperms.The reasons for the sterility observed in the F 1 hybrids and F 2 plants from the direct and reciprocal crosses (P 1 x P 4 and P 4 x P1) are not unconnected with the barrier to gene flow known to exist between A. esculentus and A. callei (Martin, 1982;Siemonsma, 1982).This is attributed to the wide variation in chromosome number among the two species.Whereas A. esculentus has a genomic chromosome number of 2n=130, A. callei has 2n= 194 (Singh and Bhatnagar, 1976).Hamon and Hamon (1991) observed that interspecific hybrids between A. esculentus and A. caillei can be obtained artificially, but at experimental stations and in the field very low rates of cross fertilization are observed.In addition, the sterility of the F 1 hybrids makes their genetic participation in subsequent generations unlikely.
The very highly significant results of the fruit morphometric studies are indicative of the genetic diversity that exists among the cultivars and the hybrids.This can be exploited for the further advancement of okra with respect to fruit size to meet consumers' demand in the region.Many reports equally indicate that a good understanding of the genetic variability that exists among cultivated species of the genus Abelmoschus, in the different characters, is a useful tool in the genetic improvement of the crop (Ariyo, 1990;Bisht et al., 1995;Kiran-Patro and Ravisankar, 2004;Omohinmin and Osawaru, 2005).For the hybrid small fruit production studies, observations in the field showed that the selfing of P 1 over 4 generations always produced plants with msfpn, but only on 61% of the flower bearing nodes.For the direct hybrids, P 1 x P 2 , it was produced on 58% and for P 1 x P 3 on 52%, of the flowering nodes.Though the reasons for the observed diverse floral behaviour of P 1 are not known, they may not be unconnected with some of the complexities involved in floral initiation and development in Angiosperms which are still poorly understood (Prusinkiewicz et al., 2007;Hake, 2008;Lippman et al., 2008;Vollbrecht and Schmidt, 2009;McKim and Hay, 2010;Feng et al., 2011;MacAlister et al., 2012;Park et al., 2012).The role of cadastral genes may not be ruled out.
For P2, P3 and P4 the flowers and fruits produced over the 4 generations were always of the solitary types.From the evolutionary point of view, the sofpn habit may be considered primitive in comparison to the msfpn type; whose occurrence in okra is to be a recent development based on the knowledge that members of the Malvaceae family produce more of solitary flowers, unlike members of the Solanaceae family, that commonly produce multiple fruits per node.
According to Dimech (2011) the advantages of the inflorescence mode is all about reproduction compared to a single primitive flower.There may be dozens or even hundreds of flowers in an inflorescence, with many seeds or fruits for each flowering.Increased pollination is an important bonus.Massing flowers together makes them more visible to pollinating insects and birds.In their report, Fatokun et al. (1979) noted that the flowering pattern of okra whereby only one flower and eventually one fruit is produced per node places a limit on the number of fruits that can be produced per plant.The existence of supernumerary inflorescence, wherein more than one flower is borne per node, provides a means of increasing total yield.
The expression of this mutant trait in A. callei, the edible okra which is indigenous to West Africa is reported for the first time (to the author's best of knowledge).This trait was found only on the trunk of very robust one-year old plants.Simpson (2006) explained that plants with sympodial growth have a specialized lateral growth pattern in which the apical meristem is terminated.The apical meristem can either be consumed to make an inflorescence or other determinate structure, or it can be aborted.Growth is continued by a lateral meristem, which repeats the process.While it is still pre-mature to draw conclusions about the expression of this trait in this cultivar, the observation further points to some of the yet to be studied aspects of flower and fruit development in the genus Abelmoschus; which are germane to increasing fresh fruit production in edible okra.
Obviously, when the reported occurrence of this mutant trait on a cultivar of A. callei is fully studied, it could facilitate ready transfer of the trait among late okra cultivars, resulting to possibly overcoming the barrier to gene flow known to exist between A. esculentus and A. callei hybrids (Martin, 1982;Siemonsma, 1982;Hamon and Hamon, 1991).This could equally result to production of small sized A. callei fruits, which will facilitate meeting okra consumer's demand for small sized fruits during the dry season; when A. callei (dry season okra) becomes readily available as A. esculentus (rainy season okra) is winding up fruit production, under natural growing conditions, in the region.The production of small sized fruits through successive planting of small fruited, A. esculentus and A. callei cultivars, could result to the availability of small sized okra fruits for the greater part of the year, under rain-fed conditions in the region

Conclusion
The current study confirms the monogenic pattern of inheritance of the Mendelian genes controlling the msfpn mutant trait.Variants of the trait could exist in the population, and these could be identified when more extensive screening are carried out, for incorporation into okra improvement program in the region.The results indicate that the dominant msfpn inflorescence mutant trait in A. esculentus can be easily transferred to other A. esculentus cultivars that produce large fruits which consumers discriminate against.With the production of small sized fruits, consumer's demand can easily be met and opportunity created even for export, since canning industries overseas equally prefer small sized fruits.The identification of the mutant trait on an A. callei cultivar is reported for the first time (to author's best knowledge) and when fully studied could facilitate the transfer of the mutant trait to A. callei cultivars that produce large fruits.The expression of the mutant trait in both esculentus and A. callei cultivars could result to the production of small sized okra fruits all the year round in the region.

Table 1 .
Fruit bearing characteristics of the 4 okra cultivars.

Table 2 .
Details of direct and reciprocal parental crosses for the inheritance of msfpn in okra.

Table 3 .
Details of F1 and backcrosses for the inheritance of msfpn inflorescence trait in okra.

Table 4 .
Phenotypic expression of msfpn in Parents, F1, F2 and backcross generations in okra. of the sofpn inflorescence type.The proportion of plants produced for both the direct and reciprocal crosses agreed with a ratio of 3 msfpn flower type: 1 sofpn inflorescence type.The back cross of the F 2 plant from the direct cross with the sofpn inflorescence parent gave rise to 214 msfpn flower producing plants and 255 sofpn inflorescence plants.For the reciprocal cross, 259 plants were of the msfpn flower

Table 5 .
Means of five fruit morphometrics for the four okra cultivars.Means on each column bearing the same letters do not differ significantly at LSD 0.05, LOF= Length of Fruit, DOF=Diameter of Fruit, COF=Circumference of Fruit, AOF= Area of Fruit, VOF= Volume of Fruit.

Table 6 .
Summary of Analysis of variance for five fruit morphometrics.
LOF= Length of Fruit, DOF=Diameter of Fruit, COF=Circumference of Fruit, AOF= Area of Fruit, VOF= Volume of Fruit.

Table 7 .
Mean Expression of multiple small fruits per node (msfpn) on mutant parent and F1 hybrids.

Table 8 .
Percentage reduction in fruit parameters based on expression of mspfn mutant gene.