Variation , correlation and path coefficient analyses in seed yield and related characters in local accessions of African Yam Bean ( Sphenostylis stenocarpa ) from Southern Nigeria

Local accessions of African Yam Bean, a poorly studied and under-utilized but important food crop, from Abia, Akwa Ibom and Cross River states of Southern Nigeria were studied for variation in seed yield and pod characteristics. Correlations were carried out on the studied characters to determine the degree of mutual association among them while path coefficient analysis was used to partition the correlation coefficients into their direct and indirect effects on seed yield. Results revealed that significant differences (p<0.01) exist among the accessions with respect to the studied characters. Mean number of seeds per pod ranged from 13.27 to 18.87; seed weight per pod from 3.19 to 6.21g; weight of 100 seeds from 22.86 to 36.72g; pod length from 21.67 to 36.82 cm and pod width ranged from 0.79 to 1.05 cm. Number of seeds per pod and pod length had significant positive association with seed weight per pod. Pod width also correlated positively with seed weight in some accessions. Positive direct effects on seed weight per pod were obtained with number of seeds per pod and pod length. The results taken together revealed significant variations in the accessions and provide evidence for effective selection which is a prerequisite for genetic improvement.


INTRODUCTION
Food security and sustainability are currently of serious global concern and unfortunately, many indigenous African crops that show promise in providing nutritional securities are presently neglected and under-utilized (Adewale and Odoh, 2012).
African yam bean (Sphenostylis stenocarpa (Hoscht ex.Rich) Harms), despite its enormous nutritional potential, represents one of such under-utilized crops in Nigeria, Ghana and many tropical African countries (Amoatey et al., 2000;Olasoji et al., 2011;Akande et al., 2012).In Nigeria, it is cultivated mainly in the southern and middle belt regions, but is also found in other West African countries including Ghana, Togo, Cote'd Ivoire, Cameroon and Central Africa (Klu et al., 2001).
African yam bean is used as food or food components and provides two consumable products: The tuber which grows as the root source and the actual yam beans which develop in pods above ground (Olasoji et al., 2011).Like *Corresponding author.E-mail: annmaureeno@yahoo.com.Tel: +2348063603103.other grain legumes, African yam bean is an excellent food, low in fats and rich in protein, carbohydrate, fibre, minerals and vitamins with relatively high contents of anti nutrients (Akande et al., 2012).
There is a general dearth of literature on African yam bean research and development; however, some important studies have been documented.Olasoji et al. (2011) reported genetic variability in its seed quality; Akande (2009) and Popoola et al. (2011) reported on the morphological characteristics of the crop.Ajibade et al. (2005) reported on the nutritive and anti-nutritive factors in African yam bean while Uguru and Madukaife (2001) reported the protein content.
Genetic improvement of this crop would among others, require that variations in seed yield and related characters among local accessions be understood for effective hybridization.Yield however is a complex character which is determined by several factors and as such, it is also necessary to study the interrelationships between yield and these related characters so that important characters that could be used as selection criteria for crop improvement would be determined.In this wise, correlation studies are useful in measuring the mutual association between characters.Alongside correlation is the path coefficient analysis which is a veritable tool in a breeder's hand for partitioning the correlation coefficients into direct and indirect effects so that the contribution of each character to a dependent character like yield could be estimated (Cyprien and Kumar, 2011).In other words, both the direct and indirect effects of related characters on yield must be considered if we want to know the total effects a change in such characters will have on yield.
In view of the critical need to genetically improve African yam bean as well as contribute literature to this poorly studied but important food crop, this study was aimed at examining some local accessions of African yam bean from southern Nigeria for variations in pod and seed yield characters, measuring the degree of association (correlation) between seed yield and related characters as well as partitioning the computed correlation coefficients into their direct and indirect effects on yield.

MATERIALS AND METHODS
Mature and dry pods of African yam bean accessions were collected in situ from three (3) different states (Abia, Akwa Ibom and Cross River) of southern Nigeria.One accession (Umudike) was collected from Abia state, three accessions (Abak, Etim Ekpo and Ikot Abasi) from Akwa Ibom state and two accessions (Biase and Obubra) from Cross River state.A random sample of fifteen (15) pods from each accession were studied and data collected on the following pod and seed characters: Pod length (cm), pod width (cm), number of seeds per pod, weight of seeds per pod (g), weight of 100 seeds (g).Pod length was obtained by measuring from peduncle stalk end to pod beak end, pod width was measured with a vernier caliper while a sensitive electronic balance was used to take accurate measurement of weights.

Statistical analyses
Data obtained from the pod and seed yield characters for the different accessions were laid out in a completely randomized design (CRD) with fifteen replications and analyzed using a oneway analysis of variance.Significant means were separated using the least significant difference (LSD) test.The intensity of linear relationship between the characters studied was determined using the Karl Pearson's coefficient of correlation, while path coefficient analysis was estimated by solving sets of simultaneous equations indicating the basic relationship between correlation and path coefficients according to Cyprien and Kumar (2011).

RESULTS
Statistical analysis of the mean values in the pod and seed yield characters of the studied accessions revealed significant variations among them (Table 1).The mean number of seeds per pod ranged from 13.27 to 18.87 (p<0.01).Abak and Umudike accessions had the lowest mean number of seeds per pod with 14.20 and 13.27, respectively.All other accessions had a mean number of seeds per pod of above 15 with the highest of 18.87 seeds per pod observed in Ikot Abasi accession.Mean seed weight per pod and weight of 100 seeds ranged from 3.19 to 6.21 and 22.86 to 36.72 g respectively.Accession from Obubra had heavier seeds generally followed by Biase accession while other accessions appeared to be weaker in this character with Etim Ekpo and Abak accessions having the lightest seeds.Mean pod length and width for the accessions ranged from 21.67 to 36.82 and 0.79 to 1.05 cm respectively.Ikot Abasi accession had the highest pod length relative to other accessions.Pod lengths of the remaining accessions did not differ significantly from one another except for Umudike accession which had the least pod length.Pod widths were bigger for Obubra and Ikot Abasi accessions, followed by Umudike, Etim Ekpo and Biase, while Abak accession had the least pod width.
Correlation coefficients for some characters of the African yam bean accessions evaluated are shown in Table 2.For all the accessions, number of seeds per pod had significant and positive correlation with seed weight per pod.Pod lengths also correlated positively and significantly with seed weight per pod in all the accessions except for Biase where correlation was positive but insignificant.In the path coefficient analyses, seed weight per pod was considered as the artifact of all the causal factors (number of seeds per pod, pod length and pod width) and the correlation coefficient of each causal factor with the seed weight per pod was partitioned into their direct and indirect effects as presented in Table 3.

DISCUSSION
Genetic variability is useful in facilitating the development  of a large number of new genotypes through hybridization by transfer of useful genes, thus maximizing the use of such available genetic potentials in boosting food production for sustainable food security (Olasoji et al., 2011).Interestingly, all the studied accessions of African yam bean revealed such variability, but also some similarity.Results obtained from the mean seed yield indicate that Ikot Abasi accession would be useful in genetic improvement for seed quantity.The range of number of seeds per pod reported in this paper is,  2012) from western Nigeria with range of 16.00 to 25.0 cm.The significantly higher pod width reported for Obubra accession may be naturally expected since the accession had significantly heavier seeds than the rest and the bulk of each seed would require more space in the pod.The general trend of these results agrees with the reports of Akande (2009), Popoola et al. (2011) and Olasoji et al. (2011) that significant variations occur in African yam bean accessions for many characters including flowering and reproductive characters, and seed quality.The results obtained from the correlation analyses provide evidence that selection for number of seeds per pod could lead to simultaneous improvement in seed yield for African yam bean.Akande et al. (2012) reported this same trend for the accessions they studied from southern Nigeria.Pod length correlated positively with seed weight per pod in most accessions but one.This indicates that longer pods in African yam bean are more often than not, likely to result in higher seed weights.Pod widths also correlated positively and significantly with seed weight in some accessions.Taken together, it is evident that both pod lengths and widths are contributory to seed weight in African yam bean and could be selected in attempts to improve seed yield in the plant especially for the accessions where they show such promise.In Abak and Umudike accessions, pod lengths correlated positively and significantly with number of seeds per pod, but interestingly, this was not the case with the other accessions where such correlations were though positive but insignificant.This result seems to suggest that longer pods may not necessarily strongly translate to more seed content in African yam bean.Path coefficient analyses showed evidently that number of seeds per pod exhibited the highest direct positive effect on seed weight in Abak (0.729), Biase (0.586) and Ikot Abasi (0.437) accessions.This direct effect is along with all other indirect positive effects of low magnitude through pod length and pod width (negative for Biase accession).According to a statistical test for relative importance of path coefficients by Cramer and Wehner (2000a), the direct positive effect of number of seeds on seed weight for Abak accession would be considered strong, but weak for Biase and Ikot Abasi accessions.Pod length exhibited the highest direct positive effect on seed weight in Umudike and Etim Ekpo accessions, while pod width had the highest direct positive effect on seed weight in Obubra accession.

Conclusion
The present study revealed significant differences in all the pod and seed yield characters evaluated for the African yam bean accessions.Obubra and Biase accessions were identified as superior in terms of seed weight while Ikot Abasi, Obubra and Biase accessions were superior to others in seed quantity per pod.Number of seeds per pod, pod length and pod width were all closely associated to seed weight in varying magnitudes for the accessions.The order of magnitude of positive direct effects of these characters on seed weight per pod could be summarized as: Number of seeds per pod > pod length > pod width.The findings from this study revealed that the interesting variations in pod and seed characters of African yam bean as well as the relative contribution of each character on seed yield (weight) would assist in the genetic improvement of this crop.

Table 1 .
Variation in pod and seed yield characters in local accessions of African yam bean from southern Nigeria.

Table 2 .
Inter-character correlations between seed yield and related characters in six accessions of African yam bean from southern Nigeria.

Table 3 .
Path coefficients showing direct and indirect effects of yield related characters on seed yield in local accessions of Africa n yam bean from southern Nigeria.