On-farm diversity of common bean (Phaseolus vulgaris L.) varieties in east Hararghe Zone, Eastern Ethiopia

Common bean (Phaseolus vulgaris L.) is an important source of food and income. However, its on-farm diversity and conservation by smallholder farmers is not known in the east Hararghe Zone of eastern Ethiopia. Thus, field survey was conducted from November 2018 to February 2019 to assess the on-farm diversity, cropping system and the role of gender in production and management of common bean varieties. Common bean producing districts were selected purposively whereas sub-districts (kebeles) were selected randomly. Three kebeles from each of the two major agro-ecological zones, two key informants and twelve general informants from each kebele were selected to constitute a total of 84 informants. The selection of general informants was stratified based on gender and wealth status. Structured interview guide was used to interview general informants whereas semi-structured interview guide was used to interview key informants. Descriptive and inferential statistical analyses were conducted in R (version 3.5.2). Seven farmers’ varieties of common bean were identified and their diversity was higher in tepid, moist mid highlands (M3) whereas cropped area (ha) was significantly (p < 0.05) higher in tepid sub-humid highlands (SH3) agro-ecology. Most activities of common bean were undertaken by male adults. Insect pests were the prominent constraints of common bean production. Farmers’ varieties were used as food and feed, and soil fertility management. Therefore, their on-farm and ex situ conservation, maintenance of endangered varieties, and shortage of cropping land due to chat plantation need special concern to promote their cultivation.


Introduction
Common bean is among major food legumes and ranks third as one of the most important worldwide food crops next to soybean and peanut in terms of production (Jibril et al. 2015). It is cultivated in a wide range of agro-ecologies and farming systems including well-watered and drought-stressed areas (Asrat 2011). Common bean is a widely adaptable, warmseason herbaceous plant with distribution in tropical, subtropical, and temperate environments (Heuzé et al. 2015). It can be cultivated in diverse types of soils, but ideally in well-drained silt loam, sandy loam or clay loam soils with high organic matter content and a pH between 4 and 9 (Ecoport 2013).
P. vulgaris L. (2n = 2x = 22) belongs to the Fabaceae family, which comprises species displaying a wide variety of forms: trees, shrubs and herbs, including many with a climbing growth habit (Porch et al. 2013). It is one of the most important legume crops grown in all continents of the world except Antarctica, because of its high protein, fiber, and complex carbohydrate content. Beans (Phaseolus spp.) are extremely diverse crops in terms of cultivation methods, in diverse environments and elevations ranging from sea level to 3000 m height, morphological variability, and utilities (dry as pulse and green as vegetable) (Pathania et al. 2006).
Its production contributes both as food for human being and fodder for livestock, export commodity and serves as a source of income and employment to a large supply chain and for risk aversion strategies to poor farmers during drought due to early maturity and moderate degree of drought tolerance (Darkwa et al. 2016). As a legume, it provides nitrogen and other soil health benefits under cropping system to subsequently grown crops.
Common bean production has been practiced in all regions of Ethiopia (Menbere 2017). However, the major common bean producing areas are central, eastern and southern parts of the country (Hirpa et al. 2013). White and red types are commonly cultivated in the country. They occupied 78,910.13 and 211,212.30 hectares of land and produced 125,980.175 and 357,942.475 tonnes of grain, respectively, during 2016/17 main cropping season (CSA 2017). These two types of common bean share 18.72% of land area and 17.2% of yield contributed by all pulse crops in the country (CSA 2017).
In eastern Hararghe Zone, common bean is 1st in area of production (3593.21 ha) and productivity among the pulse crops (CSA 2016). It is an important source of protein, source of cash, emergency crop, and well suited for multiple cropping systems with maize and sorghum.
Among common bean growing countries in Africa, Ethiopia is considered to grow many varieties of the crop (Tura et al. 2018). The immense genetic diversity of landraces of crops is the most directly useful and economically valuable part of biodiversity (Jose et al. 2014). Unlike high yielding varieties, the landraces maintained by farmers are endowed with tremendous genetic variability, as they are not subjected to subtle selection over a long period.
Even though the productivity of farmers' varieties is lower than improved varieties, their local demand is often high because they are required for the preparation of cultural foods and they are more reliable under a wider range of conditions (Bethelem et al. 2020). In addition, Farmers' traditional varieties, or landraces, have been selected for specific agro-ecological niches and in some cases have been shown to better resist insect pests, diseases and climate shocks, as well as meet the nutritional and cultural needs of Ethiopia's diverse farming communities (Tigist et al. 2018). Despite the significance of Ethiopia's legume diversity, it remains largely unstudied and unmonitored (Ruelle et al. 2019) in general and in east Hararghe Zone in particular. Studies done on common bean so far in the country focused on adoption, agronomic aspects, breeding, and disease resistance aimed at increasing yield (Agete 2014).
Therefore, this study was undertaken to assess the on-farm diversity and cropping system of common bean, and to assess traits, values and the role of gender in production and management of common bean varieties in east Hararghe Zone.

Description of study area
A field survey was conducted in east Hararghe Zone to collect basic information on common bean production and cropping systems adopted by the local farmers. The geographical location of east Hararghe Zone ranges from 7°30 0 -9°45 0 N latitude and 41°10 0 -42°50 0 E longitude with altitude ranging from 500 to 3040 m above sea level (Mengesha 2015). The zone has two rainy seasons, the short rainy season and the main rainy season, with a mean annual rainfall ranging from 400 mm in the lowlands to nearly 2000 mm at higher altitudes. The area has a temperature ranging from 10 to 25°C (Tesfaye 2013). The major annual crops grown in the zone include sorghum, maize, common bean, lentil, wheat and barley. The zone has the population of 2,723,850 (CSA 2007) and an area of 22,622.6 km 2 (Titay et al. 2021). The selected Woredas and Kebeles of the study area are also indicated in the Fig. 1.

Sampling method and sample size
Purposive selection of potential common bean producing districts (Woredas) and random selection of peasant associations (kebeles) were done to collect the valuable information. Accordingly, Haramaya, Qersa, Kombolicha and Meta districts were used for the study since they are major producers of common bean in M3 (tepid moist mid highlands) and SH3 (tepid sub-humid highlands) agro-ecological zones (EHZAO 2017).Three kebeles from each major agro-ecological classification, two key informants and twelve general informants from each kebele were included which constituted a total of 84 informants (72 for household survey and 12 for key informant).The selection of general informants was randomly made whereas the selection of key informants was purposive based on their experience in production of common bean. Twelve households (general informants) from each kebele were stratified based on gender and wealth status and thus 3 men from high income households, 3 men from low income households, 3 women from high income households and 3 women from low income households were randomly selected.

Data collection
Data were collected through structured and semistructured interview guides. Structured interview guide was used to interview general informants by ODK (Open Data Kit) installed on smart phone whereas semi-structured interview guide was used to interview key informants. Language translators were used to interview the local farmers.
Data of basic information, market access, interspecific diversity of legume crops, uses of common bean, cropping practices, intra-specific diversity of common bean, key attributes of common bean varieties, rating of attributes, and gender roles in production and management of common bean were collected through structured and semi-structured interview guides.

Data analysis
Interview data were compiled and cross-tabulated for graphic visualization in MS Excel. Descriptive and inferential statistical analyses were conducted in R (version 3.5.2). A two-factor analysis of variance (ANOVA) was conducted to determine whether there Fig. 1 Map of selected Woredas and Kebeles in east Hararghe Zone of Oromia Regional state, eastern Ethiopia. Woreda = district, Kebele = sub-district were significant differences in the area planted to agroecological zone and closest market time to Woreda, followed by a post hoc Tukey test to identify pair-wise significant differences. Shannon Diversity Index (SDI) analysis was employed to observe the diversity of common bean varieties across the study kebeles and agro-ecological zones following the diversity analysis method (Shannon and Weaver 1948) as follows: where p i is the proportion of the total number of varieties in the ith class and n is the number of households producing common bean.

Results and discussion
Variety identification and diversity Farmers identified seven common bean varieties based on their own criteria such as seed color and size ( Table 1). The richness and diversity of varieties varied across kebeles and agro-ecological zones. Shannon Diversity Index (SDI) analysis was employed to observe the diversity of common bean varieties across the study kebeles. Accordingly, it was higher at Damota (1.00) in 2017 cropping season (Table 1). Similarly, it was higher (1.03) at Yabeta Selama in 2018. In contrast, its values were nil at Duresitu Bilisuma kebele for both cropping seasons and Egu kebele in 2017 cropping season because Adi and Dima were the only varieties reported to exist at former and later kebeles, respectively. This variation in diversity might be due to differences in seed accessibility, lack of awareness and income level of farmers. Menbere (2017)  The three diversity indices of common bean varieties were higher in M3 (tepid, moist mid highlands) agro-ecology as compared to SH3 (tepid, subhumid mid highlands) agro-ecology ( Table 2). The variation of diversity in varieties might be due to differences in market access and seed availability. Contrary to this result, Bethlehem et al. (2020) reported highest value of Shannon diversity index in H3 (tepid, mid-highlands) agro-ecology as compared to others.

Wealth
Two sample t-test showed that the cropped area of common bean was not significantly (p = 0.05436) affected by the wealth group of the farmers in 2017 cropping season. Contrary to this, it was significantly (p = 0.004988) affected by wealth group in 2018 cropping season (Table 3). Cropped area of common bean was higher in mid-to-high income group (0.25 ha) as compared to low income group possibly due to more land availability and income level to buy the seed by mid-to-high income group farmers. In line with this result, Betelhem et al. (2020) reported that the area planted to common bean by smallholder farmers in Bench Maji, Sheka, Kafa, east Gojjam and south Wollo zones was significantly affected by the relative wealth categories in 2015/16 cropping season in which mid-to-high income households planted significantly more common bean than low income households (0.39 ha compared to 0.29 ha).

Agro-ecology
Common bean growing area was not significantly (p = 0.063) different in the M3 and SH3 agroecological zones in 2017 cropping season, but it was significantly (p \ 0.001) varied by the agro-ecologies during 2018 cropping season and higher value (0.28 ha) was recorded in SH3 (Table 4). This higher value of cropped area might be due to good growing conditions for common bean in this agro-ecology. It was also significantly (p \ 0.001) varied among kebeles in 2017 and 2018 cropping seasons (Table 4). The highest area of production was recorded at Yabeta Selama kebele both in 2017 (0.34 ha) and 2018 (0.39 ha) probably due to variation in land availability, income level and soil fertility status. In agreement with this result, Menbere (2017) reported significant difference in cropped area due to administrative zones and agro-ecologies in which higher area was recorded in east Gojjam zone (0.45 ha) and M3 agro-ecology (0.54 ha). Betelhem et al. (2020) also reported significant differences in cropped area in which on average, households in East Gojjam zone planted more land to common bean (0.57 ha), followed by South Wollo (0.43 ha), Kefa (0.20 ha), and Bench Maji and Sheka (0.14 ha).
Time required to the nearest road (by walking) was not significantly (p [ 0.05) different among the Woredas and agro-ecologies in 2017/18 (Table 5). But the time required to the closest market (by vehicle) was significantly (p \ 0.05) different among the   Woredas as well as agro-ecology (Table 5). The highest time to the closest market (47.2 min) was recorded in Meta while the lowest time (19.3 min) was recorded in Haramaya Woreda. Likewise, significantly higher time (23.2 min) was recorded in tepid, sub-humid highlands (SH3) agro-ecology over tepid, moist mid highlands (M3) agro-ecology. This might be due differences in topography and road facility across the Woredas and agro-ecologies.

Seed source
Different seed sources of common bean varieties such as market, family, food-aid, formal and neighbors were reported by the respondents (Table 6). About 37.5 and 34.7% of the respondents reported that market was the main source of Dima and Adi varieties, respectively. Only few respondents (13.9%) mentioned that formal seed sources such as University and Office of Agriculture served as seed source for Adi variety. The higher dependence of farmers on market as seed source might be due to shortage of land and low production of the local varieties. In contrast to this finding, Tura et al. (2018) reported that many farmers did not purchase common bean seed from market.

Growing experience
The identified varieties of common bean were grown by the respondent farmers for varied years (Fig. 2). Accordingly, greater number of respondents (37.5%) used to grow Dima variety for more than ten years. Few respondents also reported that varieties such as Jiner, Dima ashongore and Bure were grown for more than ten years. However, some key informants reported that the growing area of most common bean varieties since last five years was getting low due to land shortage and large area covered by chat plant. They also informed that varieties such as Bure, Jiner and Dalecha were endangered and their cultivation was less common in the community currently possibly due to less demand as a source of food and market. In addition, the adaptability of these varieties grown in the study area might be less. Furthermore, less attention might have been given for biodiversity conservation for these endangered varieties. In agreement with this result, Menbere (2017) reported variability in market demand among the varieties of common bean. Similarly, Tura et al. (2018) also reported that the dominance of a given variety is associated with market and use values.  Agronomic management Time of sowing The response of respondent farmers showed that common bean varieties were sown in different months of the year (Table 7). Greater number of the respondents (16.6 and 12.5%, respectively) reported that Adi and Dima varieties were sown at the end of June. The preference of farmers sowing in this time might be due to the dependence attributed to the onset of the rainfall. According to the majority of key informants (83.3%), Dima variety can be grown two times in a year based on the availability of rainfall. Zelalem (2015) also reported that the sowing seasons of common bean varieties vary following the amount and onset of rain. Asrat et al. (2013) also indicated that cropping practices in common bean production vary  depending on the reliable rainfall in which many farmers of Amaro and Konso plant larger areas with common bean during 'Belg' while many farmers in Boricha and Loma plant larger areas in the 'Meher' season.
Time of harvesting Sixteen different harvesting times were indicated by the respondents of which the majority (20.8%) reported that Adi variety was harvested at the beginning of October (Table 8). Some of the respondents mentioned that, sometimes based on the onset of rainfall, they practice sowing Dima variety at the end of March or beginning of April and harvest beginning from mid-June to mid-July. This might be due differences in sowing time and differences in maturity of the varieties. Teame (2017)  Crop rotation and intercropping Of the rotation crops with common bean varieties mentioned by the respondent farmers, sorghum was given the highest value (Table 9). Conversely, the least values were reported for faba bean and fenugreek as rotation crop with common bean varieties. The preference of common bean to rotate after cereals might be to restore the depleted soil nutrients especially nitrogen which is mostly exhausted by the cereals. Betelhelm et al. (2020) also reported that almost all farmers that grow common bean rotate common bean with other crops such as maize, teff, barley and wheat.
Maize and sorghum were equally important as an intercrop (64%) whereas only 18% of the respondents mentioned chat as an intercrop with common bean (Table 9). The preference of common bean as an intercrop might be due to its compatibility to grow as an understory crop with other crops even in a small piece of land. Betelehem et al. (2020) also reported that majority of farmers in Benchi-Maji, Sheka and Kefa Zones use maize to grow in intercrops with common bean and sorghum in south Wollo and east Gojjam Zones.

Marketing practice
Respondent farmers in the study area practiced to sell the seed of common bean varieties they produce in the market (Fig. 3). The seed of all varieties they grew were sold in the market. Most respondent farmers sell the seed of Dima and Adi varieties (59.7 and 58.3%, respectively). Only few respondents used to sell the seed of Jiner, Dima ashongore and Dalecha in the market. This might be due differences in market demand and low amount of seed production from local varieties. The report of Tura et al. (2018) agrees with this finding that many farmers produce small quantity of seeds from landraces targeting only their home consumption.

Growing
Majority of the respondents (48.6%) reported that Adi variety was grown once a year from end of June to beginning of October (Fig. 4). On the other hand, Dima variety was grown twice a year beginning from mid-April to mid of June and from end of June to the end of October. All varieties under study were grown at least once a year probably because of lack of conducive weather condition to double or triple the cultivation of common bean varieties in a year. In line with this result, Menbere (2017) also reported that all common bean varieties were grown once a year in main growing season (Meher) in east Gojjam and south Wollo zones.

Yielding status
The majority of respondents perceived that the yield of both Adi (40.3%) and Dima (33.3%) varieties were low in 2017 and 2018 cropping seasons as compared to previous years (Table 10). The low yield of these varieties might be due to the shortage of moisture, change in climatic condition and prevalence of diseases and insect pests. Similar results were reported by Menbere (2017) that lower yields of common bean varieties were obtained in 2015 than in 2016 growing season which was attributed to the shortage of moisture (drought) in the latter. Amanuel et al. (2018) also reported significant difference in seed yield of common bean cultivars due to the interaction effect of cultivars by P rate in 2016 cropping season in which Tatu gave highest seed yield whereas Remeda gave the lowest seed yield (kg/ha).

Tolerance to drought
According to respondent farmers, the identified varieties of common bean showed differences in response to drought during 2017 and 2018 cropping seasons (Table 11). About 55.5% of respondents reported that Dima variety showed low tolerance to drought. The susceptibility of Dima variety to drought might be due to the lack of minimum amount of soil moisture to support its growth. Similarly, Menbere (2017) reported that all common bean varieties under study in Amhara region were highly affected by drought.

Response to water logging
Majority of respondents reported that Dima (55.5%) and Adi variety (47.2%) gave low yield in response to water logging, respectively (Table 11). This result might be due to differences in tolerance level of common been varieties to the stress of waterlogging since common bean is a lowland pulse crop. Contrary to this result, Menbere (2017) reported that waterlogging was not a problem on the yield of common bean (Table 12).

Resistance to frost
Majority of respondent farmers (55.5%) reported that Adi variety was never affected by the prevalence of frost in the study area (Table 13) which might be due the adaptation of varieties to tolerate frost or due to early maturation of the varieties before the onset of the frost which normally occurs in October in east Hararghe Zone. Contrary to this finding, Menbere (2017) reported that none of the varieties under study was affected by frost.

Resistance to diseases
The response of common bean varieties to disease occurrence was differently reported by the respondent farmers in the study area (Fig. 5). Accordingly, a greater number of respondents replied that Adi variety was often and sometimes affected by disease. On the other hand, only few respondents reported that Dima variety (15.3%) was never affected by the disease.
Based on their long-time experience and deep knowledge, key informants mentioned the diseases affecting common bean production include Sharuf (Oroomifa) (white mold) and Wag (Amharic) (rust). This might be due to differences in adaptation and tolerance of varieties to disease. Menbere (2017) also reported disease as one of the common problems of common    bean production in Amhara National Regional state. Asrat et al. (2013) also reported diseases (rust and angular leaf spot) as an important constraint of common bean production in Boricha, Loma, Amaro and Konso area of southern Ethiopia.

Resistance to insect pests
Majority of the respondent farmers (36.1%) reported that Dima variety was often affected by pest. On the other hand, few respondents (6.9%) reported that Adi variety was never affected by pest (Table 14). In addition, four key informants (33.3%) also mentioned that Adi variety was more resistant to insect pests as compared to other varieties which agreed with the response of general informants. Furthermore, some key informants mentioned pests such as Raammoo Bili (Oroomifa) (bean bruchid), and green colored pest (bean fly, Ophiomyia spp.) of common bean varieties that affected the performance of the varieties. On the  other hand, Menbere (2017) reported statistically nonsignificant response of common bean varieties to insect pests. The difference in response to pest of common bean varieties in this study might be due to variations in tolerance to pests and their adaptations to pest stress.

Seed shattering
The ability of common bean varieties to retain the seed in their pods during maturity was explained by the respondent farmers as shown in the Table 14. Greater proportion of respondents (33.3%) reported that Dima variety was the first one to often shatter which might be due to differences in genetic makeup of varieties to retain their seeds in a pod at maturity. In contrast to this result, Asrat et al. (2013) reported differences in seed shattering property of common bean varieties in which small-seeded white variety showed more shattering.

Seed story
About 59.7% of respondents reported that Dima variety was the common traditional one (Fig. 6). Similarly, about 23.6% of respondents reported that Adi was newly introduced variety. According to the respondents, varieties such as Bure, Dalecha, Dima, Dima ashongore and Jiner were not newly introduced to the study area. This difference in seed story might be attributed due to fewer introductions of newly released varieties, less awareness of local farmers, and low income of farmers to buy the seed of new varieties. Furthermore, adaptability, resilience and

Ease of harvesting
Majority of the respondents (55.5%) reported that Dima variety was easy to harvest at its maturity (Table 15). Contrary to this, only very few respondents reported that varieties such as Adi, Bure, Adi bosolia and Dima were difficult to harvest. This difference in ease of harvesting of varieties might be due to differences in the morphological structure and their growth habit; the determinate bush as in the case of Dima variety is easier as compared to bush climbing one as in the Adi bosolia and Dima ashongore. Similar result was reported by Rediet et al. (2017) in which bush type genotypes of common bean performed better under intercropping situation.

Ease of food preparation
More number of respondents (13.8%) reported that the grain of Dima was very easy to prepare food such as 'Nifuro' in Amharic or 'Shoomo' in Oroomifa (the boiled grain) as compared to Adi variety (6.9%). This difference in ease of food preparation might be due to lesser time required to boil the grain of Dima as compared to others. Similarly, Betelhem et al. (2020) also reported that common bean varieties such as Chele Gobo in south west and Key Boloqe in the north east Ethiopia were preferred to prepare food as compared to other varieties.

Food taste
More number of respondents (18.1%) reported that Dima variety followed by Adi had very good taste when used as food (Fig. 7). Some key informants also mentioned that Dima variety was preferred to make Shoomo (Oroomifa) (boiled grain) due to its color mix with maize grain as a local food while they mentioned that the grain of Adi variety was preferred to make 'ShiroWot' (Amharic) (a sauce made from the flour of the grain) due to its aroma. The preference of the grain of different varieties for different types of food preparation might be attributed to the food culture, the appearance and the aroma of the boiled grain. Similar to this result, Abraham et al. (2016) reported that red seed beans are preferred because the red color imparts to the food after cooking. In contrast, Betelhem et al. (2020)   colored seed bean) was preferred by farmers due to its good taste and color.

Soil fertility requirement
Majority of respondents (56.9 and 51.4%, respectively) replied that Dima and Adi varieties require medium soil fertility for their optimum growth and development (Fig. 8). Only few respondents also reported that Adi bosolia, Bure, Dima ashongore and Jiner varieties require medium soil fertility. All key informants also reported that common bean requires medium soil fertility in black soils. This might be due to its ability to fix nitrogen which satisfies its nitrogenous fertilizer demand with less or no application of chemical fertilizer. Menbere (2017) also reported that common bean requires low input but it gives high yield. The report of GRDC (2014) also shows that the average amounts of nitrogen fixed range from 60 kg N per hectare for lentils to 130 kg N/ha for lupins.

Importance as food
Majority of respondent famers (40.3%) considered the grain of Adi variety very important as food (Fig. 9). On the other hand, 23.6% of the respondents reported that the grain of Dima variety was extremely important as food. The extremely importance of Dima variety might be due to its attractive color when cooked with maize grain, its less time requirement when cooked and its property to mature earlier. Moreover, most key informants (75%) reported that Dima variety was used to make Nifro (boiled grain) because of its attractive mixture color appearance with maize grain and Adi variety was used to make Wot (sauce of the grain flour) because of its aroma. Similarly, Betelhem et al. (2020) showed that red varieties (Chele Gobo in the south west and Key Boloqe in the northern Ethiopia have an appealing color and good taste, mature in a shorter time, and can be processed for food more quickly than others. Importance as a source of income More number of respondents (22.2%) replied that the grain of Adi variety was extremely important as a source of income (Fig. 10) which might be due to the ability of Adi grain to fetch the premium price in the market and its demand as an export commodity in the country. In conformity with this result, Betelhem et al. (2020) also reported that in the northern Ethiopia, the white variety (Nech Boloqe) is grown primarily for export markets being an important source of income. Likewise, Menbere (2017) also reported that white variety of common bean is mostly produced for export purpose to the European Union (EU) and United States of America (USA).

Importance as fodder
Greater number of respondents reported that all common bean varieties were used as a source of fodder (Fig. 11). Only few respondents (2.7%) considered Adi variety as extremely important source of fodder. Majority of key informants (75%) also reported that Dima variety gives more residues which are commonly fed to domestic animals such as cattle, sheep, goat and donkey after threshing. The high production of residues might be due to its high biomass production and more palatability as compared to the other varieties. Menbere (2017) also reported that the residues of common bean were highly used as a source of livestock feed. In addition, Betelhem et al. (2020) reported the importance of common bean as fodder is greater than grass pea (Lathyrus sativus) in the north eastern Ethiopia.

Soil fertility management
All general and key informants reported that growing common bean varieties had positive impact on the soil fertility status. According to their perception, growing of common bean in rotation, as a sole or intercropping with other crops such as sorghum and maize improves the fertility status of the soil. The report of Menbere (2017) also agrees with this result that crop rotation with common bean was practiced by the majority of farmers to improve soil fertility of the farmlands. Betelhem et al. (2020) also reported that through intercropping and rotation, common bean contributes to soil fertility enhancement and therefore plays an essential role in agro-ecological intensification.

Grain price
Respondent farmers in the study area used to sell the grain of common bean varieties and the price for each variety is indicated in the Table 16. Higher number of respondents (36.1%) reported that the grain of Adi and was sold at a price ranging from 10 to 15 birr per kg. This might be due its high demand for export market. Tura et al. (2018) also reported similar results in which some landraces of common bean (Abbajolle, Wajjo and Walensu) were sold at high price than others (Bushay, Dima and Gumaray). Betelhem et al. (2020) also reported that both white and red varieties are more common in the market and sold at high prices in south western and north eastern parts of Ethiopia.

Production constraints of common bean varieties
As shown in Fig. 12 below, insect pests were reported as the leading (81.9%) production constraint of common bean. On the other hand, the problem of wild animals and flooding were mentioned as equally important. Three of the twelve key informants also reported that the common insect pests in the study area include Raammoobilli (bean bruchid, Acanthoscelides obtectus)) and bean fly (Ophiomyia spp.). They also mentioned some common bean diseases such as Sharuf (white mold, Sclerotinia sclerotum) and Wag (rust). Those key informants also mentioned some weed species such as Namale (Parthenium hysterophorus), Hoolagabis (Commelina benghalensis), Qoresufi (Argemone ochroleuca) and Hadda (Guizotia scabra), as constraint to the production of common bean in the area. There is lack of pesticides and herbicides, and low awareness of farmers how to protect and control pests and weeds at their threshold. Moreover, erratic and low rainfall and lack of integrated disease management system might have reduced the productivity of the crop. Similarly, Menbere (2017) reported insect pests as a major constraint of common bean production. Girma (2016) also reported low soil nitrogen and phosphorus levels, and acidic soil conditions as important constraints for bean production in most cropping areas of Ethiopia. Contrary to this finding, Katungi et al. (2010) reported that drought was by far the most important common bean production constraint, with a probability of occurrence estimated at 38% in Ethiopia and 60% in Eastern Kenya.   Gender roles in production and post-harvest management of common bean According to the response of respondent farmers in the study area, most agronomic practices such as plowing, hoeing, land leveling, seed sowing and weeding were undertaken by male adults (Table 17). Only few numbers of male elders, female adults, female and male children were involved in aforementioned agronomic practices. Likewise, majority of respondents reported that harvesting (85%), threshing (71%), cleaning (65%) and storing (65%) practices were also performed by male adults (Table 17). On the other hand, all of the respondents mentioned that the activity of preparing food (cooking) from common bean was done by female adults. The number of female farmers (36%) was also mare than the number of male farmers (4.2%) in seed marketing. This might be due to the tradition and culture adapted in the area in which most indoor activities are carried out by female farmers while most outdoor activities are undertaken by male farmers. In line with this result, Betelhem et al. (2020) reported clear division of labor between male and female in the family in which male farmers were primarily responsible for ploughing and female farmers were responsible for food preparation. Nahusenay and Tesfaye (2015) also reported that about 90% of food processing, water fetching, fuel and wood preparing for household consumption are mostly done by women due to the norms developed as a culture in the community. Conversely, Mukasa and Salami (2015) reported that women are more likely to work in agriculture than their male counterparts in Tanzania and Uganda.

Conclusion
Through field survey, seven farmers' varieties of common bean were identified. Farmers mostly grow common bean and other legume species either by rotation or by intercropping with cereals and other crops. Farmers of mid to high income allocate more area to grow common bean than low-income farmers. Most agronomic and post-harvest activities of common bean are carried out by male adults whereas seed marketing and food preparation activities are undertaken by female adults. The diversity of farmers' varieties is more in tepid, moist mid highlands (M3) agro-ecology as compared to tepid, sub-humid mid highlands (SH3) agro-ecology. Adi and Dima varieties of common bean are the main sources of food and income, respectively. These varieties are also more tolerant to production constraints such as drought, disease, pest and others. In general, the diversity of farmers' varieties is low and should be maintained for agro-ecological intensification and soil fertility management. goes to language translators during survey work. We would also like to thank the McKnight Foundation Legume Diversity Project for financial and technical support.