Participatory Varietal Selection of Common Bean (Phaseolus vulgaris L.) in Wolaita, Ethiopia

MATERIALS AND METHODS

This experiment was conducted on-farm at Waraza Lasho village, 8 km from Wolaita Sodo (6^o51’36"N and 37°45’41"E), Ethiopia, at 1961 m above sea level, during 2014 main cropping season. Wolaita Sodo receives average annual rainfall of 1390 mm with the monthly average maximum and minimum temperatures of 25.4 and 14.7°C, respectively. During the experiment duration, May-August, 2014, the monthly average rainfall was 200 mm and the monthly average maximum and minimum temperatures were 23.6 and 15.1^oC, respectively. The soil of the experimental site at the depth of 0-30 cm is clay loam (sand 32%, clay 32%, silt 36%) and has pH 6.25, organic carbon 0.98%, P7.4 ppm (Olsen), total N 0.10% and exchangeable Ca, Mg and K of 14.20, 7.80 and 1.56 cmol kg^–1 soil, respectively.

Fourteen common bean genotypes (12 released and two advanced lines) obtained from Hawassa Agricultural Research Centre, Hawassa, Ethiopia, were planted in randomized complete block design with three replications on May 12, 2014. The common bean genotypes were planted in four rows with 2 m long separated by 40 cm, with 10 cm between plants. The distance between plots was 80 cm and that between replications was 1.2 m. The 27 kg ha^–1 N and 69 kg ha^–1 P₂O₅ were applied in the form of diammonium phosphate at the time of planting. Weeds were controlled with frequent hand weeding throughout the experiment.

Visual evaluation of genotypes was conducted at flowering and maturity stages using 8-10 farmers (host and neighbors) based on the farmers’ knowledge on common bean and their interest in evaluating the experiment. The scores for the visual evaluation were 1-4 (1= poor, 2 = good, 3 = very good, 4 = excellent) for each of the selection criteria. The ranking procedure was explained to participating farmers and the final ranking was done on consensus where differences were resolved through discussion (De Boef and Thijssen, 2007).

Days to flowering and maturity, 100-seed weight (g), pods/plant (average of five random plants at maturity), seeds/pod (average for ten random pods) and grain yield (g m^–2) were recorded using the two central rows. These data were analyzed using Genstat software (VSN International, 2012).

RESULTS

The effect of genotype was significant for days to flowering and maturity, grain filling period, pods/plant, seeds/pod, grain yield (g m^–2) and 100-seed weight (g). Grain yield ranged from 183 (genotype Awash-Melka) to 342 g m^–2 (genotype Ibbado), pods/plant from 7.87 (genotype Tatu) to 13.73 (genotype SARI-1) and 100-seed weight from 16 (genotype Awash-1) to 51 g (genotype Ibbado). There was also considerable variation among genotypes for seeds/pod, days to flowering and maturity and grain filling period (Table 1).

Table 1:	Mean squares and mean values for seven grain yield and yield components of fourteen common bean genotypes grown during 2014 main cropping season
1Genotypes Awash-1 to Waju are released varieties, ETAW-01-L-1-25A and ETAW-01-L-3-15A are advanced lines, 2Numbers in the parenthesis are degree of freedom, GY: Grain yield, SW: 100-seed weight, DTF: Days to flowering, DTM: Days to maturity, GFP: Grain filling period, *’**Significant at p<0.05 and p<0.01, respectively, ns: Not significant

Table 2:	Simple correlation coefficients among seven grain yield and yield components of fourteen common bean genotypes grown during 2014 main cropping season
*, **Significant at p<0.05 and p<0.01, respectively, ns: Not significant

The correlation of grain yield with 100-seed weight (r = 0.72, p<0.01) and grain filling period (r = 0.68, p<0.01) was positive whereas its correlation with that of pods/plant (r = -0.59, p<0.05) and days to flowering (r = -0.71, p<0.01) was negative. The 100-seed weight was also positively correlated with grain filling period (r = 0.54, p<0.05) and negatively correlated with that of pods/plant (r = -0.73, p<0.01), seeds/pod (r = -0.90, p<0.01) and days to flowering (r = -0.74, p<0.01) (Table 2).

The mean scores for seven farmers’ selection criteria ranged from 1.71 (genotypes Awash-1 and Awash-Melka) to 3.86 (genotype Remeda). The highest average scores (3.71) were also obtained for genotypes Ibbado and Tatu. The highest scores (4) for early flowering and maturity were obtained for genotypes Ibbado, Omo-95, Remeda and Tatu. The late flowering (1) and maturity (2) scores and that of late maturity score (1) were obtained for genotypes Awash-1 and Awash-Melka and genotype Waju, respectively. The highest scores (4) for pod and grain yield loads were obtained for genotypes Hawassa-Dume, Omo-95, Red-Wolaita and Remeda. The highest scores (4) for seed size were obtained for genotypes Ibbado and Remeda and the lowest score (1) for genotype Awash-1. Genotypes with either white (Awash-1, Awash-Melka, Waju, ETAW-01-L-1-25A) or creamy (genotype Tabor) seeds were given the lowest score (1) and those with red speckled (genotypes Ibbado and Tatu) seeds were given the highest score (4).

Table 3:	Farmers’ scores and ranks of fourteen common bean genotypes using seven selection criteria
1Genotypes Awash-1 to Waju are released varieties, ETAW-01-L-1-25A and ETAW-01-L-3-15A are advanced lines, DTF: Days to flowering, DTM: Days to maturity, PL: Pod load, GYL: Grain yield load, scores: 1: Poor, 2: Good, 3: Very good, 4: Excellent

Genotypes Hawassa-Dume, Ibbado, Remeda and Tatu are said to be fast cooking (scores of 4), whereas Awash-Melka and Tabor are long cooking types (Table 3).

DISCUSSION

As to the present experiment, the existence of genotypic variation for grain yield and yield components has been reported in previous studies in common bean (Atuahene-Amankwa and Mechaels, 1997; Fageria et al., 2010; Balcha, 2010, 2014). This suggests the possibility of selection to improve grain yield in the materials studied. On the other hand, the positive correlation of grain yield with 100-seed weight (r = 0.72, p<0.01) and grain filling period (r = 0.68, p<0.01) would suggest the possibility of increasing grain yield by increasing the later traits. However, the progress in selection will be slow and complicated because of negative correlation of 100-seed weight with pods/plant (r = -0.73, p<0.01) and seeds/pod (r = -0.90, p<0.01) and that of pods/plant with grain filling period (r = -0.68, p<0.01). The negative correlation between seeds/pod and seed weight (Balcha, 2014) and positive correlation between the later and grain yield (Fageria et al., 2010; Balcha, 2014) have also been reported in previous studies in common bean.

Besides grain yield and fast cooking time, farmers most preferred large and red or red-speckled seeds which are expected to give high market values. The importance of seed size and cooking time for common bean (Assefa et al., 2005) and maize (Tadesse et al., 2014), seed size for faba bean (Mulualem et al., 2012) and seed color for barley (Ceccarelli et al., 2001), wheat (Workineh et al., 2014), maize (Mulatu and Zelleke, 2002; Tadesse et al., 2014), common bean (Assefa et al., 2005) and sorghum (Muui et al., 2013) have also been reported in previous studies as being farmers important selection criteria.

In present experiment, farmers preferred early flowering and maturity because it gives them early food security for the family in the season. Earliness has also been considered as the most important selection criterion of farmers particularly in drought prone areas. This has been reported for several crops such as maize (Mulatu and Zelleke, 2002; Tadesse et al., 2014), common bean (Assefa et al., 2005), faba bean (Mulualem et al., 2012), sorghum (Muui et al., 2013) and barley (Ceccarelli et al., 2001). Farmers mostly preferred genotypes which combined high yield, early flowering and maturity, large and red or red-speckled seeds and fast cooking time. Thus, genotypes Ibbado, Remeda and Tatu would be used to improve adoption and varietal diversity.

ACKNOWLEDGMENT

This experiment (SCM/002/06) was conducted by the financial support provided by Wolaita Sodo University, Wolaita Sodo, Ethiopia.