1 In ternationa l Scholars Journa ls African Journal of Food Science Research ISSN 2375-0723 Vol. 9 (4), pp. 001-012, April, 2021. Available online at www.internationalscholarsjournals.org © International Scholars Journals Author(s) retain the copyright of this article. Full Length Research Paper A study of intraspecific hybrid lines derived from the reciprocal crosses between wild accessions and cultivated cowpeas (Vigna unguiculata (L.) Walp.) B. Lelou1,2* M. Diatewa2 and P. Van Damme1 1 Laboratory of Subtropical and Tropical Agriculture and Ethnobotany, Faculty of Bioscience and Engineering, Ghent University, Coupure links 653, BE - 9000 Gent, Belgium. 2 Faculté des Sciences de la santé; Institut de Développement Rural, Université Marien Ngouabi, B.P. 69, Congo-Brazzaville. Accepted 08 July, 2020 Cowpea (Vigna unguiculata (L.) Walp.) is an essential food legume in the (sub) tropical areas. Reciprocal crosses were performed using wild accessions for transferring genes of interest to cultivars (524-B/IT84S-2049). Of these, a low number of seeds were obtained because of crossing barriers. A seed sample of 17 F1 hybrids germinated after germination was 69.09%. A subdivision into seed sterility, partial seed fertility and seed fertility was observed. Seed sterility was possibly due to chromosomal disturbances that occur in endosperms and embryos during early seed development. In partial seed fertility, vigorous plants flowered about 128 DAE but no pods were formed because of floral abscission at anthesis. In plants of (524-B (♀) × tenuis (♀)) combination, partial sterility was caused by incomplete male sterility. In 13 F1 hybrids, viability among populations was reduced (67.25%). Eleven F1 hybrids were grown in the greenhouse to produce F2 seeds by natural self-pollination. Plants were characterised through using cowpea descriptors. Variability, in terms of morphological characters in adult F1 plants, was established. Differences in vegetative, inflorescence and fruit characters were described after full plant development. The results show that the adult F1 plants appear to be dominant for wild vegetative and inflorescence characters expressed by one or two of the parents in bi-parental and reciprocal crosses, for exception plants derived from crosses in which alba wild forms served as male parent showing the same morphological characteristics as the cultivated female parents, as evidenced by the traits inherited from cultivars. The heterotic status exhibited by emerged plants, supposes that the latter types inbred at the first generation confirm that the wild parents involved in these hybrids represent a wide diversity of germplasm. Seed characteristic studied in F2 has not shown segregation because of recessive type absence, suggesting that morphological traits should be monitored. Key words: Reproductive barriers, seed fertility, introgressive hybridisation, wild character inheritance. INTRODUCTION Cowpea (Vigna unguiculata (L.) Walp.) (2n = 22) is an essential grain legume crops in tropical and subtropical *Corresponding author. E-mail: blelou@hotmail.com. Tel: +242/04-420-17-77. regions (Rachie and Roberts, 1974). It provides inexpensive proteins and can be prepared easily in a number of edible forms, such as tender green shoots and leaves, immature pods, and green and dry seeds (Fery, 1980; Ehlers and Hall, 1997). In small-scale agriculture, the major problem is the extremely severe damage (up to 90%) caused by 2 post-flowering insect pests in cowpea, including the legume pod borer Maruca testulalis Geyer and the complex of pod-sucking bugs, including Clavigralla tomentosicollis Sthal, Anoplecnemis curvipes F. and Riptortus dentipesF. (Ehlers and Hall, 1997; Olufajo and Singh, 2003). Wild taxa of V. unguiculata identified within the primary gene pool are known as a potential source of resistance to major insect pests (Padulosi and Ng, 1990). Reciprocal crosses were performed between wild accessions and cultivated cowpeas for transferring post- flowering insect pest resistance to cultivars (524-B/IT84S- 2049). According to the total number of 487 flowers that were pollinated during the hybridisation programme in 1999, compatible cross number (that is, mature pod number/total number of reciprocal crosses) produced 78 (16.01%) mature pods in 20 to 25 days after pollination (Lelou and Van Damme, 2006). The number of incom- patible crosses was about 391 (80.7%), revealing the presence of reproductive barriers after fertilisation/seed development, as reported by Barone and Ng (1990); Lelou and Van Damme (2006). Viability is assayed using the tests of germination (Scott et al., 1998). These present attempts were conduc-ted to estimate seed fertility yielded in F1 hybrid lines derived from intraspecific reciprocal crosses between wild accessions x cultivars in order to identify some hybrid lines with excellent response after the experimental tests in 2000/2001 crop seasons. The incorporation of genes of one species into the gene pool of another is termed introgressive hybridisation (Yoshinari, 2007). To illustrate multi-step processes by which genetic materials from wild accessions infiltrate into cowpea cultivars and vice-versa, the characterisation in each of F1 cowpea adult-hybrids was conducted in order to examine the extension of variability range after different morphological character inheritance. The former characters are primitive and the latter are derived types. During hybridisation programme, the derived types are expressed in these hybrids and in order to make favourable genetic material. Thus three morphological types (vegetative, inflore- scence and fruit characters) were characterised amongst F1 cowpea hybrid lines. From a number of the F1 cowpea hybrid lines that emerged, the expression of characters was variable, indicating that environmental conditions as well as the segregation of minor genes influence pene- trance. It is known that characters are not phenotypically identical. It is possible e.g. that leaf texture type can be possessed less advantage in cultivation under green- house conditions than it apparently possesses in nature. The objective of this experimental study was, with regard to post-zygotic barriers that build up after success- ful fertilisation, to determine whether the expression of different genes derived from cultivars and wild accessions grown in the greenhouse could provide an excellent opportunity to study evolutionary processes in all F1 hybrid populations. On the basis of these results, it can be expected to select new cultivars of cowpea by selection for plant resistant to insect-pest flowering and adaptations to variable environments over several generations. MATERIALS AND METHODS Hybrid seed materials The seed samples of the original F1 cowpea hybrid lines obtained during a hybridisation programme were executed in a tropical greenhouse between wild accessions and cultivated cowpeas during the winter season of 1999 at the Laboratory of Tropical and Subtropical Agriculture and Ethnobotany, Faculty of Bioscience and Engineering (Ghent University, Belgium). In Table 1, we charac- terised wild and cultivated parental accessions of cowpea manipulated in this breeding programme. Seed germination experiments The seed samples of 17 F1 hybrid lines of cowpea were subjected to laboratory germination tests, as described below during the spring season of 2000 to 2001. In Table 2, the different 17 recipro- cal F1 hybrid combinations of cowpea, pollination number/hybrid combination, aborted ovule number and their pod and seed production, are reported respectively. These seed samples collected in all F1 hybrid lines of cowpea exhibited seed coat dormancy, as wild parental accessions of cowpea. The methods of seed preparation and sowing were an adaptation of the techniques described by Lelou (1997). Physical dormancy is caused by water impermeable layers of palisade cells in the seed or fruit coat which control water movement. Prior to the conduction of the germination tests, these seeds were treated to break physical dormancy. Seeds were surface-sterilised in 70% ethanol for one minute and disinfected by (CLO3)2Ca (5%) for five minutes, rinsed three times in distilled water. To induce a better permeability, seed coats were mechanically scarified mildly with a scalpel in order to overcome dormancy during germination. To prevent contamination by micro-organisms, these above mentioned seeds treated were immersed for a minute in the fungicide Thiram (C6H12N2S4). After treatment, all seed samples were sown and randomly dispersed across moistened filter paper into Petri dishes. Water was introduced in sufficient and enough quantities into the closed Petri dishes and kept in a ‘Philips’ incubator: 24°C day and 20°C night temperatures at 12 h photoperiod/day. The results of germination test were recorded daily for 7 days during laboratory experiments. Seedling emergence The durations from sowing to emergence of the radicle through the seed coat of seeds treated, were observed among seed samples constituting the cowpea F1 hybrid lines. Petri dishes placed into an incubator were checked daily in order to detect: a) radicle elongation, that is, observation of radicle visible through seed testa, when it exceeded 4 or 5 mm out extended cotyledons; and b) seedlings destroyed, that is, seeds ungerminated in which seedlings were regarded to have died when it feel over a most of the seedling turned yellowish or brownish. Once radicle definitively emerges, seedling is subsequently potted in small plastic pots (diameter 5 cm) and placed into the growth chamber with same climatic characteristics that the incubator. At the two leave-stages, the young plantlets belonging to adult F1 hybrids of cowpea were individually grown in plastic pots of 3 Table 1. Origin, biological status and some morphological and pollen male fertility characteristics of parental genotypes of cowpea (Vigna unguiculata (L.) Walp.) used in our breeding programme conducted under greenhouse at Lab. of tropical agriculture in Faculty of Bioscience (Ghent University, Belgium) in 1999/2000/2001. Parental genotypes and Biological Biological Country Plant Terminal leaflet Flower Pod Testa Plant Pollen fertility Growth habit Leaf texture Leaf marking number/ accession number status cycle of origin hairiness shape colour colour texture (%) genotype subsp. burundiensis NI Wild Allogamous Burundi Indeterminate Glaberscent Cariaceous Sub-globose Absence Violet Pale tan Rough 8 79.0 456 and prostrate subsp. baoulensis Wild Allogamous Cameroon Indeterminate Glaberscent Cariaceous Sub-globose Absence Violet Pale Rough 8 83.5 NI 749 and prostrate Tan subsp. pawekiae Wild Allogamous Malawi Indeterminate Glaberscent Cariaceous Sub-globose Absence Violet Dark Smooth 8 86.0 MT 53 and prostrate brown to rough subsp. letouzeyi Wild Allogamous Cameroon Indeterminate Glaberscent Cariaceous Sub-globose Absence Violet Pale Rough 8 84.5 NI 1420 and prostrate tan subsp. alba wild Allo- Congo Indeterminate Glaberscent Cariaceous Sub-globose Presence Violet Pale tan Rough 8 88.0 SP 145 autogamous and prostrate subsp. tenuis wild Allo- South Indeterminate Glaberscent Cariaceous Hastate Presence Violet Pale Rough 8 92.5 sp 167 autogamous Africa and prostrate tan subsp. pubescens wild Allo- Kenya Indeterminate Pubescent Cariaceous Sub-hastate Presence Violet Pale tan Rough to 8 92.5 NI 979 autogamous and prostrate hairs wrinkled var. spontanea weed Autogamous Niger Indeterminate Glaberscent Cariaceous Sub-globose Presence Violet Pale Smooth 8 78.0 NI 945 and prostrate tan to rough var. unguiculata 524-B Cultivar Autogamous USA Indeterminate Glaberscent Membranous Globose Absence White Dark Smooth 4 87.0 and spreading tan var. unguiculata IT84S- Cultivar Autogamous USA Indeterminate Glaberscent Membranous Globose Absence White Dark Smooth 3 87.0 2049 and spreading tan diameter 20 cm containing the same standard potting soil in the greenhouse, in order to produce the F2 generation by the natural self-fertilisation. Data measurement The tests of germination were immediately carried out after incubation. The samples composed of 110 seeds consi- dered as well-developed in moistened derived from 17 hybrid combinations variant from 4 to 10 seeds/hybrid linewere manipulated, according to ISTA procedures (1976). Variability of seed number tested was due to the difference yielded in seed samples. The percentage of germination was calculated by use of the following formula: [ΣI N NI/ΣI N NI] × 100 (1) where: ni is the number of germinated seeds; and Ni is the total number of seed placed for germination. Pollen fertility was assessed by staining fresh pollen grains with aceto-carmine, according to Belling (1921), and calculating the percentage of deeply stained pollen grains for hybrid line. The sample for determining the pollen fertility was approximately 200 pollen grains from 2 flowers/adult hybrid plant. The pollen viability image was captured and counted with a ‘Zeiss Axioskop’ optical microscope (10 × objective). Pollen measurement was made on the F1 plants. Pollen size measurement was sub- jected through using the ocular scale on the microscope, according to microscopic observation techniques described 4 Table 2. Distribution of F1 materials obtained, according to the different 17 hybrid combinations derived from reciprocal crosses between wild accessions and cultivated cowpea (Vigna unguiculata (L.) Walp.). Hybrid combinations Pollinated flower Aborted ovule Harvested mature pod Yield of harvested seed hybrid ♀ parents × ♂ parents number/hybrid combination number/hybrid combination number/hybrid combination number/hybrid combination 524-B × pawekaie 18 16 1 15 pawekiae × 524-B 29 26 2 6 524-B × alba 35 30 5 26 alba × 524-B 19 14 3 15 524-B ×tenuis 20 18 2 4 524-B × pubescens 7 2 5 27 524-B × spontanea 3 1 1 5 IT84S-2049 × baoulensis 21 19 2 22 IT84S-2049 × pawekiae 12 7 5 63 pawekiae × IT84S-2049 30 20 10 53 IT84S-2049 × alba 47 36 10 38 alba × IT84S-2049 33 23 7 75 IT84S-2049 ×tenuis 21 16 4 14 IT84S-2049 × pubescens 33 22 11 73 pubescens × IT84S-2049 18 13 11 73 IT84S-2049 × spontanea 23 22 1 8 spontanea × IT84S-2049 12 7 5 42 by Rapilly (1968). Phenotypic expression of morphological traits To have an idea whether wild gene incorporation into cultivated forms of V. unguiculata was yielded after some artificial hybridisations in the F1 generation, the stable adult plants derived from reciprocal crosses between wild accessions and cultivated cowpeas were examined. An experimental study was conducted in the greenhouse. Nine morphological characters in which five vegetative, two inflorescence and two fruit characters were characterised, according to the descriptors for cowpeas. The cowpea F1 hybrid line constituting the sample-adult plant was composed of 3 plants, with exception for 3 F1 hybrid lines derived from (524-B (♀) × tenuis (♂) and 524-B (♀) × spontanea (♂) combinations in which 2 plants were formed and in IT84S-2049 (♀) × baoulensis (♂) combination 1 plant was obtained. Fourteen hybrid combinations con- stituting the biodiversity at the first generation susceptible to inherit the latter types were examinedduring vegetative and reproductive stages. Biodiversity range description Distribution of the F1 adult hybrids, five qualitative and morphological, two inflorescence and two fruit characters as following as: growth habit, plant hairiness, leaf texture, terminal leaflet shape, leaf marking, flower colour, gametophytic male, pod colour and texture of testa seed characters were studied. Amongst these selected charac- ters, the frequencies of distribution were used to determine for each continuously varying traits with exception for pod colour and texture of testa seed characters in which 36 F1 adult plants were controlled Growth habit This character was visually evaluated, 6 weeks after planting, taking into consideration in the indeterminate spreading traits of hybrid plants. The scale used was: (i) Indeterminate spreading and not climbing, that is, lower branches touch the ground. (ii) Indeterminate spreading and prostrate i.e. plants flat on ground; branches spread several metres. (iii) Indeterminate spreading and climbing. (iv) Indeterminate spreading and climbing It was observed that the adult vigour plants in hybrid accessions lines tended to have an indeterminate 5 conformation. Distribution of the hybrid accession lines over the three-growth habit groups was respectively 78.58 and 21.42% for Groups 2 and 3. Plant hairiness According to descriptors for cowpeas, stems, leaves and pods were either glabrescent, that is, lacking hairs or tendency to become hairless vs. pubescent short hairs. These both characteristics were assigned the Indexes 3 and 5, respectively. From 14 hybrid accession lines observed, only 1 hybrid line (3 sample-adult plants/F1 hybrid line of cowpea) was assigned the Index 5, that is, pubescent and short hairs. In other words, almost 92.59% of hybrid accessions had glabrescent stems, leaves and pods. Leaf texture Leaf texture was visually observed 6 weeks after planting, according to the following scale: (i) Cariaceous (ii) Intermediate (iii) Membranous The distribution of the hybrid accession lines over the three groups of leaf texture was respectively 78.58 and 21.42% for Groups 1 and 3. Terminal leaflet shape A visual evaluating of the terminal leaflet shape took place 6th week after planting, according to the following scale: (i) Globose (having shape as a globe) (ii) Sub-globose (iii) Sub-hastate (iv) Hastate (triangular, shaped like a spear point) The distribution of the hybrid accessions over the four groups of terminal leaflet shape was 71.42, 21.42 and 7.14%, for Groups 2, 1 and 3, respectively. Leaf marking Observations were done for this character for 6 weeks after planting and the presence vs. absence of V mark was observed on leaflets: 71.42% of samples are free of leaf marking and 28.57% are leaf marked. Flower colour Flower pigmentation was visually evaluated using the following scale: (i) White (ii) Violet (iii) Mauve-pink (iv) Other About 38 sample-adult hybrid plants were inspected from 45 to 150 days with the first flowers appearing 1 week earlier. The distribution of the hybrid accessions over the four groups of flower colour was 14.28 and 85.72% for Groups 1 and 2, respectively. Gametophytic male The results of pollen fertility from 14 cowpea F1 hybrid lines were presented in range of 60, 74 and 86.5% for Groups 1, 2 and 3 respectively. Pod colour A visual rating of pod colour took place after harvesting and drying, according the following scale: (i) Pale tan or straw (ii) Dark tan (iii) Dark brown (iv) Black or dark purple (v) Other The number of hybrid accessions was reduced to 36 hybrid plants. Because of 2 adult plants form (524-B (♀) × tenuis (♂)) combinations flowered and collapsed after emergence. Distribution frequency of hybrid accessions over the five groups of pod colour was respectively 46.15, 38.46 and 15.38% for Groups 1, 2 and 3. Testa seed texture Nine textures of testa seed were identified in the collection. These textures presenting in the cowpea descriptors are: (i) 1 – smooth (ii) 3 – smooth to rough (iii) 5 – rough (fine reticulation) (iv) 7 – rough to wrinkled (v) 9 – wrinkled (coarse folds on the testa) Distribution frequency for texture of testa concerned notably 13 hybrid accessions, shown for F2 seeds in 16.66, 33.33, 41.66 and 8.33% for Groups 1, 3, 5 and 7, respectively. EXPERIMENTAL RESULTS AND DISCUSSION Seed germination evaluation According to the texture of seed testa derived from genetic materials of the reciprocal crosses between wild accessions × cowpea cultivars, it was provided evidence that all hybrid seeds inherited wild female parent characteristics at the first generation. The texture of seed testa was phenotypically identical, according to the origin of wild parental strains. Genetically, the effects expressed in these F1 seeds allowed to show that the success of germination test depends in part on wild parent material source. From these data reported, the germination of dominant seed coat dormancy identified amongst 17 cowpea intraspecific F1 hybrids obtained is summarised in Table 3. Seeds germinated There was a wide variation in fertility level in the seeds, in accordance with the cross-results. The rate of germinated 6 Table 3. Results of germination tests, of F1 seeds derived from reciprocal crosses between wild accessions and cultivated cowpeas (Vigna unguiculata (L.) Walp.), according to ISTA procedure (1976), germination was conducted in Petri dishes containing a moistened filter paper and placed in a Philips incubator with specific characteristics (day and night temperatures (24/20°C), fluorescent illumination system (black and light, 365 nm, 4 watt) and photoperiod/day (12 h) Hybrid combinations Total number of F1 seeds sown / Number of germinated seeds at the beginning and Total number of seeds Germinated seeds F1 ♀ parents x ♂ parents hybrid combination final periods of the tests germinated after 7 days percent/hybrid combination - - At the beginning period At the final period of the - - of the test test 524-B × pawekiae 6 2 4 6 100 pawekiae × 524-B 6 3 3 6 100 524-B × alba 6 2 4 6 100 alba × 524-B 6 1 5 6 100 524-B × tenuis 4 0 2 2 50 524-B × pubescens 7 0 0 0 00 524-B × spontanea 5 0 2 2 40 IT84S-2049 × baoulensis 8 0 1 1 12.5 IT84S-2049 × pawekiae 6 2 4 6 100 pawekiae × IT84S-2049 8 2 6 8 100 IT84S-2049 × alba 7 3 4 7 100 alba × IT84S-2049 7 2 5 7 100 IT84S-2049 × tenuis 6 0 0 0 00 IT84S-2049 × pubescens 9 0 0 0 00 pubescens × IT84S-2049 5 2 3 5 100 IT84S-2049 × spontanea 4 2 2 4 100 spontanea × IT84S-2049 10 2 8 10 100 - 110 seeds 23 seeds 53 seeds 76 seeds (mean) X = 69.09% seeds after the test was 69.09 germinated percentage. The highest percent of germination (100%) was distinguished in 11 hybrid lines derived from crosses. Therefore, the seeds from (524-B (♀) × tenuis (♂)) combinations were second (50%) and these from (524-B (♀) × spontanea (♂)) combinations were third (40%). Finally, the percentage of germinated seeds of (IT84S-2049 (♀) × baoulensis (♂)) combinations was 12.5%. In this later case, the final counts made after 7 days showed that germination speed was absolute minimum. The results revealed that one seed germinated. However, the seedling was considered as normal in Petri dish. Seeds ungerminated The number of seeds ungerminated (30.91%) was determined after seedling counts. Final counts made after 7 days showed that any radicle elongation was observed or very short and turned yellowish in the following hybrid lines derived from these combinations: 524-B (♀) x pubescens (♂), IT84S-2049 (♀) × tenuis (♂) and IT84S-2049 (♀) × pubescens (♂). Since the results of the standard of germination test showed defects as several factors such as abnormal embryo development, physiological immature at the harvested period, seed quality and yield are directly influenced by the genetic composition of seeds (Scott et al., 1998). Manz et al. (2005) reported that the relationships between germination test and radicle emergence initiation 24 h after imbibition and finished germination before 96 h are excellent, when artificial 7 conditions provide favourable effect for viable seeds to germinate and to perform seedling growth. The F1 generation observation Hybrid viability Luo et al. (2005) reported that an individual is viable if it can survive the adult stage. Sahai (2000) indicated that the late stage embryo mortality is the main case of seed sterility. In these cases of sterility evidenced, the embryo- logical disturbances should be caused by genetic factors (Rieseberg, 2001). In other circumstances, the adverse climatic effects can probably disturb the normal embryo development which varies considerably in different localities, localities, trees and species (Robertson et al., 1996; Hall, 2003). Adverse climatic effects can cause injury in any (early or late) stage of embryo development. Embryo survival depends on the severity of the injury (Sahai, 2000). According to Walsh et al. (2003), viability testing through germination is essential for the maintenance of a seed gene bank collection and can be a rapid way of identifying problems with the seed storage condition. After sterility extraction among F1 seeds derived from reciprocal crosses between wild accessions and cultivated cowpeas, the development of the next generation was required. Fertility indicates chromosome pairing between wild accessions × cowpea cultivars. However, the taxonomic studies in cowpeas revealed the close phylogenetic relationships between cowpea parent materials (Maréchal et al., 1978; Pasquet, 1993; Coulibaly et al., 2001). The particular (524-B (♀) × tenuis (♂)) combination was studied; hence it was discovered in progeny line that vigorous plants flowered about 128 DAE (that is, days after emergence) (data no shown). This mutant was also characterised by failure of a majority of the flowers to develop beyond the bud stage with the result that it is nearly completely sterile. It is evidence that hybrid viability at the F1 generation was found to be reduced (67.25%). F1 cowpea hybrid line characterisation According to IBPGR (1982), the characterisation consists of recording those characteristics highly heritable that can be easily observed by eyes and expressed in all environments. These are usually represented by the morphological characters of plants (Table 4). Hybrids of indeterminate spreading and prostrate × indeterminate spreading and climbing grown: A total of 38 hybrid accessions derived from between indetermi- nate spreading and prostrate × indeterminate spreading and climbing parents was studied. The F1 was indeter- minate spreading and prostate in grown habit. With regard to comparative parental characteristics; data analysis revealed that two lines derived (IT84S-2049 (♀) × alba (♂) and 524-B (♀) × alba (♂)) had produced the same morphological and characteristics as their respective cultivated female parents. We also observed white flower colour after the propagation of reproductive organ. These confirmed the presence of two infertile or apomictic lines among cowpea F1 hybrid combinations. Hybrids of pubescent short hairs × glabrescent: From 32 F1 hybrid accessions examined, all hybrid lines were almost glabrescent stems, leaves and pods. In certain hybrids and particularly under greenhouse conditions the expression of the pubescence short hairs was observed in 3 hybrid lines derived from (pubescens (♀)×IT84S- 2049 (♂)) combination. Hybrids of cariaceous × membranous of leaf texture: Two different classes were studied at the first generation. One of the membranous parents was observed in these observed apomictic lines. Other F1 hybrids indeterminate and prostrate inherited the cariaceous characteristic as the wild material parents. The distinction between types was clear and no difficulties were encountered during identification stage. Hybrids of sub-globose and sub-hastate × globose of terminal leaflet shape: The particular sub-hastate character was discovered in allo-autogamous and perennial wild parental strain, that is, subsp. pubescens. The F1 hybrids derived from (pubescens (♀) ≤ IT84S- 2049 (♂)) combinations were completely inherited sub- hastate characteristics. A typical leaf trait was shown in Figure 1. In the F1 hybrids formed with allogamous and other allo-autogamous perennial wild parents, the type of sub-globose terminal leaflet shape was generated. On this other hand, the type of globose terminal leaflet shape characteristic was observed among apomictic lines. Hybrids of V mark on leaflets: All combinations derived from allo-autogamous perennial wild parents × cultivated parents, the F1 hybrids shown V mark on leaflets. In combinations with allogamous perennial and annual wild parents, the leaf surfaces were leaf marking free. Hybrids of yellow × purple flower colour: Flower colour of the F1 plants was completely purple in F1 hybrid lines. In fact, several researches published that in F1 plant flower colour both genes control this trait. According to Fery (1980) white and purple flower are codominant at the first generation. Hybrids of high × low percent pollen fertility: Pollen fertility in F1 plants was determined by staining. The data from male fertility parental genotypes was high over one 8 Table 4. Data on morphological character characterisation carried out among 14 hybrid combinations derived from reciprocal crosses between wild accessions and cultivated cowpeas (Vigna unguiculata (L.) Walp.), according to descriptors for cowpeas. Characters studied Growth habit Plant hairiness Leaf texture Terminal leaflet shape Leaf marking Flower colour Pod colour Testa texture Pollen fertility percent - Index Index Index Index Index Index Index Index > 70% < 70% Parental genotypes Wild accessions pawekiae 2 5 1 2 - 2 3 3 86.0 - baoulensis 2 5 1 2 - 2 1 5 83.5 - alba 2 5 1 2 + 2 1 5 88.0 - tenuis 2 5 1 4 + 2 1 5 92.5 - pubescens 2 3 1 3 + 2 1 7 86.0 - spontanea 2 5 1 2 + 2 1 5 78.0 - Cultivars 524-B 3 5 3 1 - 1 2 1 87.0 - IT84S-2049 3 5 3 1 - 1 2 1 78.0 - F1 hybrid combinations 524-B × pawekiae 2 5 1 2 - 2 3 3 71.5 - pawekiae × 524-B 2 5 1 2 - 2 3 3 72.5 - 524-B × alba 3 5 3 1 - 1 1 1 81.6 - alba × 524-B 2 5 1 2 + 2 1 5 - 59.4 524-B × tenuis 3 5 3 1 - 2 - - 70.0 - 524-B × spontanea 2 5 1 2 + 2 1 5 76.4 - IT84S-2049 × baoulensis 2 5 1 2 - 2 1 5 72.7 - IT84S-2049 × pawekiae 2 5 1 2 - 2 3 3 80.8 - pawekiae × IT84S-2049 2 5 1 2 - 2 3 3 73.8 - IT84S-2049 × alba 3 5 3 1 + 1 3 1 80.8 - alba x IT84S-2049 2 5 1 2 + 2 2 5 72.7 - pubescens × IT84S-2049 2 3 1 3 + 2 1 7 - 67.3 IT84S-2049 × spontanea 2 5 1 2 + 2 1 5 72.3 - spontanea × IT84S-2049 2 5 1 2 + 2 1 5 84.3 - Growth habit, Index 2: indeterminate and prostrate; Index 3: indeterminate spreading and climbing. Plant hairiness, Index 3: glabrescent; Index 5: pubescent and short appressed hairs. Leaf texture, Index 1: cariaceous; Index 3: membranous. Terminal leaflet shape, Index 1: globose; Index 2: sub-globose; Index 3: sub-hastate; Index 4: hastate. Leaf marking, index (-): absence; index (+): presence. Flower colour, Index 1: pale tan; Index 2: dark tan; index 3: dark brown.Testa texture, Index 1: smooth; Index 3: smooth to rough; Index 5: rough; Index 7: rough to wrinkle. Pollen fertility: percentage > 70 and < 70% over one; however, low pollen fertility percentage was observed in plants from infertile line as (alba (♀) ×524-B (♂)) combination and in F1 hybrid line of (pubescens (♀) × IT84S-2049 (♂)) combi- nation. High pollen fertility was identified in plants derived from two infertile lines of (524-B (♀)× alba (♂)) and (IT84S-2049 (♀) × alba (♂)) combina- tions, in F1 hybrid lines from (IT84S-2049 (♀)σ pawekiae (♂)) and (spontanea (♀) ×IT84S-2049 (♂)) 9 Figure 1.The F1 (pubescens (♀) × IT84S-2049 (♂)) hybrid line presenting the sub-hastate terminal leaflet shape and V marks on leaflet characteristics. combinations. Hybrids of dark tan and brown × straw pod colour: Pod colour derived from F1 hybrids at the first generation presented the same phenotypic configuration as female parents in crosses there was maternal dominance inheritance occurred in all of hybrids (Birky, 1995). Hybrids of rough × smooth seed testa texture segregated: In all of hybrids the F1 textures of seed testa inherited seed characteristic of wild parental strains. After F1 adult-plant development, all morphological charac- teristics belong to the first generation but seeds provided by these hybrids plants constitute the F2 genetic material. In the first generation character recombination amongst all hybrid individuals revealed that wild characters expressed by one or two of the parent in bi-parental and reciprocal crosses were dominant, except in pod colour in which maternal dominant effect encountered, has shown maternal dominance inheritance, according to Aliboth’s et al. (1997) works, wild character dominance in the fertile hybrid lines derived from reciprocal crosses between wild accessions and cultivated cowpeas showed a heterotic status for morphological, inflorescence and fruit traits. In the F2 generation, 16.66 smooth, 33.33 smooth to rough, 41.66 rough and 8.33 rough to wrinkled types occurred in a ratio of 2 smooth: 2 rough individuals. The deficiency in the recessive class in F2 may indicate lower viability of the mutant type. In seed testa texture character in which individuals were distinguished in 2 classes, the F2 ratio observed may indicate segregation 10 Figure 2. Case of vegetative characteristic occurred in the sterile F1 (524-B (♀) × tenuis (♂)) hybrid line regarded to be near to cultivated parent. absence because of progeny tests with them were not possible. In studying of evolutionary processes of traits, in mutants of (524-B (♀) × tenuis (♂)) during the distinction between vegetative and reproductive types, these obtained individuals were morphologically near to culti- vated parent, that is, the 524-B cultivar (Figure 2). The wild parent (tenuis subspecies) was implicated through purple flower colour, indicating probably that partial hybridisation was executed into this sterile hybrid line. Berville (2002) illustrated this kind of difference in Helianthus sp. studies in which hybrid plants derived from interspecific crosses between cultivated Helianthus (♀) × Helianthus mollis (♂) (that is, wild diploid and perennial species) through using molecular markers, the obtained F1 hybrids developed fragments from wild parental strains in flower characteristics, revealing partial hybridisation occurred in these interspecific crosses. In partial. 11 Table 5. Germination of F2 self-fertilised seeds produced by F1 adult-plants of cowpea derived from reciprocal crosses between wild accessions and cultivated cowpeas (Vigna unguiculata (L.) Walp.), according to ISTA procedures (1976). Hybrid combinations Harvested F2 seed Number of F2 seeds ♀ parents x ♂ parents number/hybrid combination sown/hybrid combination Number of F2 seeds germinated at the beginning and final periods of tests At the beginning of At the final period of period of the test the test Percent of F2 seeds Observations germinated (%) 524-B x pawekiae 395 143 75 68 100 High germination speed pawekiae x 524-B 756 168 98 70 100 high germination Speed alba x 524-B 1071 193 78 115 100 High germination speed 524-B x spontanea 1113 188 100 88 100 High germination speed IT84S-2049 x baoulensis 76 68 5 15 29.41 Low germination speed* IT84S-2049 x pawekiae 1032 168 60 99 94.64 high germination speed pawekiae x IT84S-2049 1780 188 80 108 100 High germination speed alba x IT84S-2049 652 118 55 63 100 High germination speed PUBESCENS x IT84S-2049 1603 193 75 118 100 High germination speed IT84S-2049 x spontanea 1928 168 62 106 100 High germination speed spontanea x IT84S-2049 1239 155 77 78 100 High germination speed - 11645 1750 765 928 (Mean) X = 93.09% - *Low germination speed in this (IT84S-2049 × Baoulensis) hybrid line was observed at the beginning and final periods of the tests. hybridisation, plants were nearly or completely sterile The F2 generation observation We examined 11 F2 hybrid populations derived from 11 parental different combinations. From seed number produced by natural self-fertilisation in the F1 adult plants, the total number of F2 plants studied in the green house was 33 individuals (Table 5). After germination of self-fertilised seeds produced by 33 F1 adult-plants, we detected that post-zygotic barrier mechanisms which take effect after successful fertilisation, include seed abortion and the weakness or sterility of F1 hybrids. We have seen that F2 plants derived from (IT84S- 2049 (♀)×baoulensis (♂)) combination exhibited abnormalities during growth and development stages. This hybrid failed to yield viable seeds by self-pollination. Sterility identified in F2 generation may be suggested as being due to genetic causes acting along with the effects of cryptic structural differences (Rieseberg, 2001). Seed number obtained in F2 generation was more important than in the F1 generation. The difference in seed production between populations of F2 shown that plants F2 were productive and provided of viable seeds per pod and able to produce the next generation by natural self- pollination. Conclusion In producing intraspecific hybrids derived from reciprocal crosses between wild accessions and cultivated cowpea through using conventional breeding systems, artificial hybridisations yielded several crossability barriers. Despite dysfunctions identified in embryo development during gene transfer, in intraspecific F1 production after the germination tests, some F1 plants obtained were fertile. Pollen fertility of these plants was normal (70%). Even with semi-fertile seeds from (524-B (♀) × tenuis (♂)) combinations, partial sterile plants were generated. The occur- rence was induced by incomplete expression of male sterility. The distribution of fertile plants was reduced. During characterisation, two lines derived from (IT84S-2049 (♀) × alba (♂) and 524-B (♀) × alba (♂)) restored the identical characters of the cultivars. These lines were discarded because of apomictic lines. In fertile hybrid lines, the domi- nance of wild genes was evident. Heterosis fixed 12 in plants at the first generation declines in the F2 genera- tion because inbreeding influence. Based on intraspecific F1 hybrids, we conclude that lines obtained in this study possessed different characters. In the second generation of segregation, it would be important to follow these works in open field under agro-ecological conditions appropriated to V. unguiculata species in order to elucidate inheritance of morphological characters in segregating hybrids. With regard to the persistence of incomplete male sterility expressed in this above indicated cross, it would be advantageous to develop suitable restorer lines with the desirable recombination of morphological characteristics for exploitation of hybrid vigour. ACKNOWLEDGEMENTS We thank J. Geinaert for taking care of the plants in the greenhouse and for technical help. This work was entirely supported by Gent Universiteit (België). REFERENCES Aliboth VO, Kehinde OB, Fawole I (1997). Inheritance of leaf mark, pod dehiscence and dry pod colour in crosses between wild and cultivated cowpeas. Afr. Crop Sci. J., 5(2): 1-4. Barone A, Ng NG (1990). Embryological studies on crosses between Vigna unguiculata and V. vexillata. Their collection and potential utilization. In: N.G. Ng and L.M. Monti (Eds;), Genetic resources, contributions in cowpea exploration. Evaluation and Research from Italy and the International Institute of Tropical Agriculture, pp. 137- 143. Belling J (1921). On counting chromosomes in pollen mother cells. Am. Nat. 55: 573-574. Berville A (2002). Structuration and composition of scientific teams. Sunflower and olive genetics. UMR - 1097. (Diversity of Cultivated Plant Crops Genomes). Report, pp. 1-9. Birky CW Jr. (1995). Uniparental inheritance of mitochondrial and chloroplast genes: Mechanisms and evolution. In Review Proc. Nat. Acad. Sci. USA, 92: 11331-11338. Coulibaly S, Pasquet RS, Papa R, Gepts P (2001). AFLP analysis of the phonetic organization and genetic diversity of Vigna unguiculata L. Walp. reveals extensive gene flow between wild and domesticated types. Theor. Appl. Genet., 104: 358-366. Ehlers JD, Hall AE (1997). Cowpea (Vigna unguiculata (L.) Walp.). Fields Crops Res., 53: 187-204. Fery RL (1980). Genetics of Vigna. Horticultural Reviews, Edited by Jules Janick, Purdue University. Avi Publishing Compagny, Wesport, Connecticut, USA, 2: 314-376. Hall AE (2003). Comparative ecophysiology of cowpea, common bean and peanut In: Physiology and Biotechnology Integration for Plant Breeding. Edited by Henry T. Guyen and Abraham Blum. CPL Scientific Publishing Services Limited, pp 271-325. ISTA (1976). Int. Rules for Seed Testing Assoc. Norway Seed Sci. Technol., 4: 2-49. Lelou B (1997). A study of genetics organisation of different wild forms of Vigna unguiculata (L.) Walpers. Master Thesis. Gembloux Universty of Agronomy, Belgium, 62p. Lelou B, Van Damme P (2006). Production of intraspecific F1 hybrids between wild and cultivated accessions of cowpea (Vigna unguiculata (L.) Walp.) using conventional methods. Comm. Appl. Biol. Sci., Ghent Univ., 71(4): 57-75. Luo L, Zhang Y-M, Xu S (2005). A quantitative genetics model for viability selection. Heredity, 94: 347-355. Manz B, Müller K, Kucera B, Volke E, Leubner-Metzger G (2005). Water uptake and distribution in germinating tobacco seeds investigated in vivo by nuclear magnetic resonance imaging. Plant Physiol., 138: 1538-1551. Maréchal R, Mascherpa J-M, Stainer F (1978). Complex group taxonomic study of species Phaseolus and Vigna (Papilionaceae) genus on morphological and pollinic databases through using informatic analysis. Boissiera, 28: 10-273. Olufajo OO, Singh BB (2003). Advances in cowpea cropping system research. In: Challenges of Opportunities for Enhancing Sustainable Cowpea production to farming system/agronomic improvement of cowpea production. Edited by C.A. Fatokun, S.A., Tarawali, B.B., Singh, P.M. Kormowa and M., Tamò. IITA (Ibadan, Nigeria), pp 267- 277. Padulosi S, Ng NG (1990). Wild Vigna species in Africa. Their collection and potential utilization. In: N.G. Ng and L.M. Monti (Eds;), Genetic resources, contributions in cowpea exploration. Evaluation and Research from Italy and the International Institute of Tropical Agriculture, pp 58-77. Pasquet RS (1993). Intraspecific classification of spontaneous forms of Vigna unguiculata (L.) Walp. (Fabaceae) by using morphological databases. Bull, Jard. Bot. Nat. Belg., 62: 127-173. Rachie KO, Roberts LM (1974). Grain legumes of the lowland tropics. Advances in Agron., 26: 2-118. Rapilly F (1968). Mycology techniques in vegetale phytopathology. Epiphytic Ann., 19(N.H.S): 1-102. Rieseberg LH (2001). Chromosomal rearrangement and speciation. Trends Ecol. Evol., 16: 17: 351-358. Scott RJ, Spilman M, Bailey J, Dickinson HG (1998). Parent-of-origin effects on seed development in Arabiodopsis thaliana. Development, 12: 3329-3342. Walsh DGF, Waldren S, Martin JR (2003). Monitoring seed viability of fifteen species after storage in the Irish threatened plant genebank. Biology and Environment. Proceedings of the Royal Irish Academy. 1033(2): 56-67. Yoshinari T (2007). Introgressive hybridization as the breakdown of postzygotic isolation: a theoretical perspective. Ecol. Res., 22(6): 929-936.