Morphological Characterization of Some Cowpea Accessions from Africa

Background and Objective: Cowpea is a pantropical and highly variable crop plant with diverse varieties. The importance of morphological characters in resolving complex differences that coexist among this genus cannot be overemphasized. This study aimed at assessing the qualitative and quantitative characters of ten cowpea accessions collected from ten African countries. Materials and Methods: Cowpea seeds were planted in planting pots with 5 replicates and arranged in an incomplete randomized block design. A total of 37 morphological quantitative and qualitative characters were studied for 12 weeks. Results: There was high variability in the quantitative characters assessed and only TVu-13840 had 100% germination. The TVu-10847 recorded the highest average plant height, mean rachis and petiole length while Tvu-1236 had the highest leaf surface area with no flower pigmentation. Growth and leaf shape varied greatly and the twining tendency was more pronounced in Tvu-10847 and Tvu-11825. All the accessions developed a straight pod shape with the exception of Tvu-14316 which was slightly curved. High variability was observed in raceme position, flower color, seed shape and other distinctive qualitative features. Cluster analysis resolved the characters studied and grouped the accessions into four clusters. Conclusion: The morphological characters of the studied cowpea accessions is an important step in the management of cowpea genetic diversity and also a prerequisite towards selecting desirable traits or improved varieties for breeding purposes.


INTRODUCTION
Vigna unguiculata (L.) Walp. commonly known as cowpea is a grain legume belonging to the family Fabaceae. It comprises ten wild perennial subspecies and one annual subspecies (ssp. unguiculata) 1,2 . It is an herbaceous annual which grows at a temperature of 18-28 throughout all its stages of development 3 . Cowpea grows well both under favorable and unfavorable growing conditions being able to survive water, temperature and other abiotic stresses 4 . The dry grain is the most commonly used part of the plant however the fresh or dried leaves 5,6 , fresh peas and fresh green pods have also been found to be of use in certain cultures. The grain is one of the cheapest sources of proteins which make it extremely valuable to people who cannot afford protein-rich food such as meat and fish in the West and Central Africa, where cowpea is also an important crop 7 . It is truly a multifunctional crop, with high income generating potential for grain traders and farmers in growing locations.
Several authors [8][9][10][11][12][13][14][15][16] have characterized cowpea accessions using various markers however, morphological and agronomic traits remain valuable to plant breeders 16 . This is particularly so as identification of key traits of the germplasm enhances the more efficient selection of parents for breeding. Similarly, variation studies in cultivated crops provides valuable information for intraspecific hybridization towards the mapping of traits for the establishment of new populations 17 .
The need for proper understanding of the variability that exists in cowpea accessions is highly necessary to formulate and accelerate the conventional breeding programme. Collection, evaluation and quantification of diverse variability existing for different characters in cowpea accessions would help to classify cowpea into groups/clusters and identify desirable traits. This would in turn assist breeders in efficient selections. In view of this, the present study aimed to evaluate the extent of variability existing for different morphological characters in ten cowpea accessions collected from ten African countries which will be of immense use for breeders and farmers to choose cowpea accessions of interest for different breeding program.

MATERIALS AND METHODS Plant materials:
All the cowpea seeds from African countries that were used for this study were obtained from the Genetic Resources Unit of the International Institute of Tropical Agriculture (IITA), Ibadan, Oyo State Nigeria. The denomination of the collected accessions was assigned by IITA. The assigned denomination (accession number), cultivar name and country of origin were presented in Table 1.

Experimental site:
The study was carried done in the Botanical Garden screen house of the University of Lagos, Nigeria (Latitude 6°30.8800 N: Longitude 3°23.9440 E). The temperature ranged from 34.7°C during the daytime to 28.7°C at night, while maximum and minimum humidity were 75% and 55%, respectively.
Experimental design: Clean plastic bowls filled with loamy soil were used for the planting in an incomplete randomized block experimental set up of five replicates each. Four rows were marked per plot with intra row spacing of 15 cm and inter row spacing of 50 cm. The planting method used was drilling. The spacing was done within the screen house in other to give equal chances of survival to the crop and also avoid overcrowding. The germination period was observed for all the samples and documented. The plants were watered with an equal volumes of water when needed all through the period of the study. The entire screen house and the surroundings were kept clean and all weeds around were handpicked. The experiment was observed from 0 to 12 weeks. Data collection: Cowpea data were collected following the morphological descriptors of the International Board for Plant Genetic Resources 18 at different growth stages. Morphological evaluation was done using a modified method of Padulosi and Ng 19 . A total of thirty-seven characters were assessed comprising both qualitative and quantitative characters of the germinated plants. To avoid being biased, five representatives of each of the ten accessions studied were chosen and observed along the two diagonals of the observation square. On each of the studied accessions, both quantitative and qualitative traits were observed, measured with a meter rule and then coded appropriately for data entry.
Data analysis: Descriptive statistics of the mean and standard error of the data collected were calculated using Microsoft Excel 2013. Pair wise analysis was carried out by computing distance matrices for the characters assessed following Onuminya et al. 20 . Also, using NTSYS-pc 2.02j software package, a sequential, hierarchical and nested (SAHN) cluster analysis was carried out. Based on Nei genetic distances, dendrogram was generated following Quesada et al. 21 .
Almost all the cowpea accession exhibited globose leaf shape with the exception of Tvu-144 and Tvu-12995 which displayed sub-globose and sub-hastate leaf shape respectively (Fig. 2, Table 4). Most of the studied cowpea accession showed a slight tendency to twine although, Tvu-10847 and Tvu-11825 displayed a pronounced twining tendencies ( Table 4). The leaf surface area of the studied cowpea varies between accessions. Tvu-1236 (278 cm 2 ) recorded the largest surface area, while Tvu-16071 (70 cm 2 ) had the lowest surface area. The rachis length and petiole length recorded were not within the same range.
The raceme position observed among the studied accession indicates that two (TVU-12191 and TVU-14976) accessions had above canopy, four (TVU-1236, TVU-11825, TVU-144 and TVU-14316) accessions had throughout canopy and three (TVU-10847, TVU-12995 and TVU-16071) accessions had upper canopy ( Table 5). The flower color was also observed which include violet (n = 6), white (n = 2) and mauve pink (n = 1) among the studied cowpea accession (Fig. 3).  Other qualitative characters evaluated among the studied accession were presented in Table 5. All the cowpea accession developed straight pod shape except Tvu-14316 whose pod shape was slightly curved (Fig. 4). Variation in seed shape was also observed, accession (TVU-144, TVU-11825, TVU-12191 and TVU-12995) developed kidney-shaped seeds, TVU-1236 and TVU-14316 had globose-shaped seeds, TVU-14976 and TVU-16071 showed rhomboid-shaped seeds, while TVU-10847 showed ovoid-shaped seeds (Fig. 5, Table 5). Among the ten studied accession, only Tvu-13840 neither developed to flowering nor developed pods. Therefore, pod characters, seed characters and floral characters were not evaluated for the aforementioned accession.  The seed weight also varies and range between 726-2381 mg per ten seed. Other quantitative characters evaluate among the studied cowpea accession were presented in Table 6.
Cluster analysis of morphological characteristics: The cluster analysis (Fig. 6) resolved the morphological characters of ten sampled accession at about 0.40 similar coefficient while the accessions were grouped as one at similarity level of 0.14. There are various stages between the two extremes which

DISCUSSION
Previous studies on cowpea indicate that morphological traits influences the potential yield of cowpea [8][9][10]22 . Also, a number of morphological characters were found to be of great importance to assess the level of genetic variability and have led to a better classification of cowpea species 18,[23][24][25][26][27] . This study compared ten cowpea genotypes originating from various countries in Africa at morphological level. Both quantitative and qualitative traits were evaluated. The wide range of data obtained from majority of the traits suggested that genetic variability exists in the accessions hence they can be improved through plant breeding programmes. High variation was observed in the quantitative characters evaluated in the cowpea genotype while a few qualitative traits such as leaf shape, pod shape and pattern of pigmentation on pod showed less variation. The influence of these traits on cowpea contributes directly or indirectly to the variation that exist among these cultivars. The high variation observed in peduncle length, pod weight, pod length, number of locules per pod, number of pod per peduncle and number of pod per plant could be linked to genetic factors including those control time-factor for assimilates. These variations are of prime importance to framers, consumers and plant breeders, e.g. peduncle length of cowpea helps to determine pods position on the plant, which would in turn ease the visibility of pods on the crop canopy. Thereby, becoming an important character with respect to harvesting 28 . The study of Doumbia et al. 29 also reported that pod length and number of locules/pods has effects on seed yield and these traits can also be genetically controlled.
The weights of ten seeds of the ten accessions varies between 726-2381 mg, this finding also corroborate the values obtained for seed weight by Gbaguidi et al. 30 which was found to be between 400 mg and 2375 mg. The observed flower color of the accessions evaluated in this study showed that 66.7% of the genotypes developed violet flowers, 22.2% white flowers and 11.1% developed mauve-pink flowers which is in agreement with the observation of Yasin et al. 28 . Leaf color intensity also varied from pale to dark green. The study showed 50% of the accession having green leaves while dark green and pale green leaf colors were found in 30% and 20%, respectively of the cowpea genotypes. Also, 77.8% of the accession did not develop pod pigmentation while 22.8% had pigmentation on their pod. The variation observed in the intensity of green coloration on leaves, flower pigmentation, leaf shape, seed color, seed eye color and twinning tendency highlighted the presence of a diversity of morphological characters between the cowpea accessions. The results of this study was also consistent with the reports of Amusa et al. 31 .
Majority of the genotypes displayed globose leaf shape making up 80% of the cowpea accession while sub-hastate and sub-globose were each found in 10% of the genotypes. The kidney, ovoid, globose and rhomboid seed shape was observed in this study. Furthermore, variation in coefficient of diversity was recorded for various stages of the plant development, this is in agreement with the results reported by Amusa et al. 31 . In some genotypes (TVu-1236) early flowering was observed (54 days after germination) while in others (TVu-11825 and TVu-16071) flowering was late (95 days after germination). According to Omomowo and Babalola 32 short flowering period is an advantage at high temperatures and this also helps in avoiding low air humidity.
The dendrogram resulting from the hierarchical clustering of the quantitative and qualitative characters revealed four major clusters (A, B, C and D). This dendrogram showed a perfect correspondence with the morphological classification of the accessions studied using the quantitative and qualitative characters. the accessions collected from the ten African countries. Furthermore, the use of molecular method to elucidate and support the morphological evidences established in this study is also of prime importance.

CONCLUSION
Genetic diversity based on morphological characters are affected by variations in environmental conditions as well as limited number of morphological markers thus accounting for the relatively low similarity index recorded amongst the accessions studied. However, significant variations in several morphological traits were detected. Also, the late flowering and date to maturity of some of the studied cowpea accession suggest that a number of these genotypes are photoperiod sensitive.

SIGNIFICANCE STATEMENT
Owing to the need for proper understanding of the variability that exist in cowpea accessions to formulate and accelerate conventional breeding program, this study aimed to evaluate the extent of variability existing for different morphological characters in ten cowpea accessions collected from ten African countries. Both quantitative and qualitative vegetative and reproductive characters of the plant were assessed. Results showed great diversity in both characters which assisted in classifying the accessions into four clusters. The variations recorded would assist breeders in efficient selections for their breeding program.