MATERIALS AND METHODS

This experiment was mounted during the 2010-2011 farming season. Seeds of four landraces of cowpea-Akidi, Olaudi, Ileje ajaka and Ileje were obtained from dealers in Enugu and Kogi States, Nigeria, respectively. A plot of land measuring 10x10 meters was manually cleared in the University of Calabar Experimental Farm, Calabar. Five beds were made with a spacing of 2 m between beds. Three seeds were sown in a hole of 4 cm deep per variety (Center for New Crops and Plants Production, 2002). The 4 varieties were randomized on each bed with 8 replications per variety using Randomized Complete Block Design (RCBD). Spacing was 50x75 cm. After seedling emergence, each stand of individual variety was thinned down to 2 stands. Weeding was done 3 and 5 weeks after planting while staking was done 4 weeks after planting. Days to seedling emergence was noted while data on number of leaves, leaf area and vein length were collected 5 and 10 weeks after planting. Days to 50% flowering and days to 50% maturity were also recorded while number of flowers per plant; pod lengths, inter-node length per plant, number of seeds per pod, number of pods per plant, seed yield per plants and 100 seed weight were obtained at maturity.

Statistical analyses: Analysis of variance was computed using statistical software PASW 18 at probability level of 95% (α = 0.05). Phenotypic and genotypic coefficients of variation (PCV and GCV), phenotypic, genotypic and environmental variations, genetic advance and broad sense heritability were estimated according to the methods of Singh and Chaudhary (1985) while genotypic variance was calculated as mean difference between genotypic mean square and error mean square (environmental variance) (Uguru, 1998).

RESULTS

Performance of morphological traits of the four cowpea landraces: The mean morphological and yield performance of four landraces of cowpea showed that there were significant differences (p<0.05) in all the parameters studied except days to 50% seedling emergence that showed no significant difference (p>0.05) among the varieties screened (Table 1). Result obtained showed that Ileje had the smallest leaf surface area at 5 and 10 weeks, while Olaudi had the highest leaf surface area. This trend however, was followed for the number of leaves per plant at 5 and 10 weeks. Importantly, it was observed that Olaudi produced the highest number of seeds per plant (seed yield) while no significant difference (p>0.05) on seed yield was observed in Akidi and Ileje varieties. There was no significant difference (p>0.05) on the number of seeds per pod between Akidi and Ileje ajaka but differed considerably with those of Olaudi and Ileje. Significant differences were observed in the number of pod per plant among the varieties with Olaudi having the highest number of pods. Pod length per plant almost showed the same trend as the number of seeds per pod. Days to 50% flowering showed that Akidi’ variety flowers faster than other landraces screened. Present result revealed that there was no significant difference in the days to 50% flowering between Olaudi and Ileje varieties.

Table 1:	Mean yield and yield related morphological traits of four cowpea landraces
Means followed with the same superscript along horizontal array indicates no significant difference (p>0.05)

Table 2:	Variance estimates, genetic advance and heritability of morphological traits of cowpea landraces
PCV: Phenotypic coefficient of variation; GCV: Genotypic coefficient of variation; GA: Genetic advance; h2: Broad sense heritability.

Ileje ajaka slightly differed from Ileje in their time of flowering. Additionally, days to 50% maturity showed no significant difference (p>0.05) between Olaudi and Ileje ajaka but significantly differed from those of Akidi and Ileje. The result obtained revealed that Ileje variety matures late when compared to other landraces studied. Olaudi was observed to performed best in most morphological traits studied, most importantly, seed yield.

Variance estimates, genetic advance and heritability of morphological traits investigated: The results presented in Table 2 showed variance estimates, genetic advance and heritability of morphological traits after planting. Results revealed that there were wide genotypic, phenotypic variances among cowpea landraces screened except for days to 50% seedling emergence with very low variance estimates as compared with other parameters. Leaf area per plant, number of leaves per plant, number of flower per plant, vein length at 5 and 10 weeks and seed yield showed very high phenotypic and genotypic variances. Phenotypic and genotypic coefficients of variations of the aforementioned traits were also correspondingly high. Other morphological attributes such as days to 50% flowering, number of pod per plant, days to 50% maturity and number of seed per pod showed high phenotypic variance, phenotypic coefficient of variation but rather low genotypic variance and genotypic coefficient of variation. Generally, our result also showed that all the cowpea landraces studied had low genetic advance while broad sense heritability were comparatively high. Conversely, the magnitude of the genetic variances in the yield and yield-related traits were not proportional to the heritability estimates. Seed yield per plant had the highest variance estimates, genetic advance and high heritability.

DISCUSSION

Creation of genetic variability and subsequent selection for important traits is undoubtedly crucial activities that any plant breeder should apply to achieve better yield and other desirable traits. Selection of superior genotypes in any crop is proportional to the amount of genetic variability present in the population and the degree to which the traits are inherited. The magnitude of these estimates obviously suggests the degree to which improvement can be made possible through selection (Ragsdale and Smith, 2003; Udensi et al., 2011a).

Though all morphological traits studied in this research contribute either directly or indirectly to grain yield which is the focal point for selection and breeding of cowpea, it is our thinking that greater attention should be paid to days to flowering, days to maturity and most importantly grain yield. It might prove logical to think that since akidi reaches flowering time first, it would have matured first. This however, was not the case as olaudi and ileje ajaka matured faster than akidi variety. The mechanism underlying this is presently unknown though it might probably be assumed that the variety to reach anthesis first does not necessarily imply that the variety will mature first.

One would have expected that the more the number of leaves, the higher the seed yield courtesy of photosynthesis. This was not the case as olaudi having 111.52 leaves per plant at the 10th week produced 911.0 seeds. The broader the leaf area, the more surface exposed for photosynthetic activities, which could have been the underlying reason behind the seed yield of olaudi variety. This corroborates the earlier reports of Udensi et al. (2010, 2011a). Going by the outstanding performance of olaudi on days to flowering, days to maturity, pod length and seed yield obviously suggest that probably olaudi had superior genes (Udensi et al., 2011a). Present result seems to suggest a direct proportional relationship between the lengths of pod and number of seeds therein. It implies therefore that the longer the pod, the higher will be the seed number. This succinctly, points out to the fact that selection and subsequent breeding of cowpea landraces should take cognizance of varieties with long pods.

The high and wide variability observed among the cowpea landraces screened in the study, especially on some traits (Table 2) agrees with earlier reports by Tyagi et al. (2000) and Sarsamkar et al. (2008) in pigeon pealy (Idahosa et al., 2010; Udensi et al., 2011a). There were high genotypic and phenotypic coefficients of variation for most of the parameters studied. This is in conformity to the reports of Makeen et al. (2007) and Farshader and Farshader (2008).

Variance estimates in our work, including the broad sense heritability suggest that additive gene effects played paramount role in influencing traits performance. Udensi et al. (2011a) opined that since phenotypic coefficient of variation was greater than genotypic coefficient of variation, so yield improvement through phenotypic selection becomes cardinal. This submission presents itself in our present study. Heritability estimates is the measure of phenotypic variance attributable to genetic causes which has a good predictive function of breeding crops. Present result on broad sense heritability suggest high breeding value, implying more additive genetic effect which is pivotal for crop improvement (Dabholkar, 1992; Rohman et al., 2003; Udensi et al., 2011a). Seed yield should be an important index during cowpea selection, breeding and improvement. This is the case with our present study as there was corresponding high heritability estimate for seed yield and genetic advance, although Amin et al. (1992) reported that for better genetic gain through selection, high heritability will not always be associated with high genetic advance.

CONCLUSION

Giving that these indigenous crops have high nutritive value, high productivity, high variability, high adaptive potential and reservoir for economic genes to withstand changing climatic conditions, they need intensive and holistic research, even in the humid lowland conditions of Calabar, Nigeria. Present results explicitly suggest that these cowpea landraces should be improved, especially olaudi variety owing to their morphological attributes, high variance estimates and high heritability.