|Promotion of forage legume
seed production and processing on smallholder farms in Western Kenya
Forage legumes are important in the smallholder farming systems in Kenya in terms of improving the quality of the pastures and agricultural by-products fed to livestock, as well as improving soil fertility. An inadequate supply of affordable seed has however, limited the use of forage legumes in the farming system. A study was undertaken with the aim of (1) evaluating the production potential of Desmodium uncinatum (Silverleaf desmodium), Stylosanthes guyanensis (Stylosanthes), S. scabra (Stylosanthes), Dolichos lablab (Lablab bean) and Macrotyloma axillare (Axillare) and (2) assisting smallholder farmers to start producing seeds for sale and utilizating them on their farms. A total of 832 farmers in Uasin Gishu and Bungoma district and 231 in Migori district were exposed to legume seed production through formal and informal training’s during the study period. Sixty two percent of the farmers in Bungoma obtained seed yields in the region 50 - 137 kg/ha for Silverleaf desmodium , 71% in the region of 300 - 750kg/ha for S. guyanensis and 33% for S. scabra in Bungoma district. These yield levels were economically profitable. In Migori district Silverleaf desmodium was a profitable enterprise while Stylosanthes guyanensis and Macrotyloma axillare was not. Farmer evaluations indicated that most of their seed was lost through harvesting and threshing for Silverleaf desmodium and Stylosanthes spp and through number of harvests for Dolichos lablab. Macrotyloma axillare failed to produce seed at Bungoma. Sustainable marketing was of major concern to farmers if they have to undertake commercial seed production.
Forage seeds play a vital role in livestock development as they are the foundation for increased feed production per unit area. The demand for seed, however, depends on the emphasis given to livestock production by farmers and individual Governments. In Kenya, dairy cattle are highly ranked by smallscale farmers and the Government as an important enterprise for income generation, food security and manure production (National Development Plan 1997-2001, Rees et al. 1998 and Stall et al. 1997). Milk yields per cow are generally low and this is attributed to lack of provision of adequate quality feed. Napier grass and cereal by-products which form the bulk of roughage for dairy cows are generally low in crude protein especially during the dry season. Commercial protein sources that could supplement these roughages are too expensive for smallholder dairy farmers to afford on a regular basis and in adequate amounts (FAO 1985, Valk 1990). As a result, the Government has given priority to research and promotion of technologies that can encourage on-farm production of such feeds so as to reduce the cost of milk production (KARI 1991). Production and utilization of forage legumes is one of the low cost methods for enhancing both the quantity and quality of livestock feeds on smallholder farms since they are high in protein content (Tothill 1986). The legumes can also concurrently enhance soil fertility for companion fodder grasses and subsequent cereal crops (Tothill 1986, Mwangi et al. 1999 and Synders 1992) there by reducing the cost of livestock feed and crop production for the resource poor farmers.
Recent participatory rural appraisal reports from KARI regional research centres (Rees et al 1998 a,b, Stall et. al. 1997, Ramadhani et. al. 1999) and previous reports from the National Dairy Development Project (NDDP 1994) indicate that farmers are a ware of the importance of forage legumes and are willing to plant them on their farms, but unavailability of low cost legume seeds is a major constraint. The imported legume seed is scarce and expensive for farmers to purchase. Work carried out on research centres, indicate that Kenya can produce its own seed for some of the legumes if the existing research findings are transferred through trainings and on-farm research to small scale farmers and interprenuers (Muyekho and Wandera 1996). The objectives of this study were:
The study was carried out on smallholder farms in Turbo, Uasin Gishu district; Kibabii and Bukembe (Kanduyi), Bungoma District; Rongo and Central Kanyamkago, Migori district. Turbo lies in the Lower Highland zone 3 (LH3) at an altitude of 2000m a.s.l. with an annual rainfall of 900mm. The soils are predominantly cambisols and farm sizes range from 10 to 50 acres per household. Kanduyi is in Lower Midland zone 1 and lies at an altitude of 1211m a.s.l. with an annual rainfall of 1250mm. The soils are ferrasols with farm sizes per household ranging from 1 to 10 acres. Rongo and Central Kanyamkago are in the Lower Midland zones 2 and 3 (LM2 and LM3) respectively of the Lake Victoria region. Soils are predominantly Phaeozems. The altitude and annual rainfall are about 1200m a.s.l. and 1250mm respectively. Average farm sizes per household is 5 acres. Turbo and Kanduyi has one cropping season while Rongo and Central Kanyamkago has two reliable cropping seasons.
At all the sites identification, sensitisation and formation of groups of smallholder farmers to start legume seed production was done using Participatory Rural Appraisal tools. Detailed analysis of the problem and potential benefits of forage legume seed production and herbage utilization were discussed with farmers. Farmers to participate in the verification of the seed production potential of Desmodium uncinatum (Silverleaf desmodium), Stylosanthes guyanensis (Stylo guyanensis) Stylosanthes scabra (Stylo scabra), Dolichos lablab (Dolichos) and Macrotyloma axillare (Axillare) were selected by fellow farmers during community meetings. The groups at Rongo and Kanyamkago evaluated seed production potential of Silverleaf desmodium, Stylo guyanensis and Axillare. The group at Kibabii, Kanduyi division in Bungoma district and at Turbo, Uasin Gishu district evaluated Silverleaf desmodium while the one at Bukembe in Kanduyi division evaluated Stylo guyanensis, Stylo scabra, Dolichos and Axillare. The selected farmers prepared land ready for planting, routine management and participate in all project training activities and evaluations. The project provided seeds, fertilizer, chemicals for spraying against pests and technical back up.
Training of group members on forage legume seed production and utilization were done stepwise through formal workshops and practical involvement at each farm. Neighbours were encouraged to attend. Data on germination, flowering, seed yield was collected. Data for the cost of various operations from planting to seed harvest for legumes and maize was also collected from farmers at Kibabii and Bukembe in Bungoma for the economic evaluation. Farmers also carried out qualitative evaluation of the various activities involved in legume seed production through matrix ranking.
To promote production and marketing of seeds, the members of the seed producing groups in collaboration with researchers and extension staff organized farmers meetings (barasas) and field days where neighbouring dairy farmers, Kenya seed company and non-govermental organizations attended. Researchers, Extension and Kephis staff regularly visited and provided advice to farmers.
Results and Discussion
Group formation and training
Five groups were formed by the communities at the 5 sites to undertake participatory research - extension on legume seed production. Kanduyi division had 2 separate groups, while the other 3 sites had one group each. The membership ranged from 8 to 12 farmers at each site. The groups at Bukembe in Bungoma, Turbo in Uasin Gishu, Rongo and Kanyamkago were formed for research purposes, while at Kibabii an existing registered women group was used. Farmers were trained on legume seed production and utilization at all the sites. In Turbo and Kanduyi, 150 farmers (90 women and 60men) consisting of atleast 2 members from each household and neighbours were given formal and practical training and the topics covered included: site selection, crop establishment, general management, harvesting, processing, storage and marketing for seed. Also included in the training were aspects of forage legume utilization as a protein supplement and group dynamics and formalization of groups for effective marketing of their farm produce. The training included a visit to Kenya Seed Company where farmers learned of commercial seed production, processing and market opportunities for their seed. In the Rongo and Kanyamkago sites about 50 farmers consisting of 29 women and 21 men were trained together with their area extension staff. The farmer research groups also visited and exchanged their experiences on legume seed production with the Kibabii and Bukembe farmer research groups.
In order to reach more members of the community field days were organized at each site. A total of 600 people consisting of 250 women, 150 men and about 200 students attended at the Turbo site. In Kanduyi, a total of 362 farmers attended of which 232 were women and 130 men. In the lake region 231 farmers attended of which 156 were women and 75 men. The field days generated interest in seed requirements and participating farmers sold most of their seed.
Seed yields of the 5 legume species varied greatly from farm to farm within a site and between sites. For Silverleaf desmodium seed yield at Turbo was extremely low while farmers in Kibabii recorded some of the highest yields (Table 1). In Turbo out of 3 acres on Mr David Sang and Daniel Cheruiyot’s farms only 100 kg of pods were harvested and this gave 5 kg of seed. The low yields were attributed both to dry spell that caused moisture stress during the flowering and seed setting phases and to the heavy rains that occurred when pods were ripe. Severe moisture deficit during flowering is known to reduce seed yield of many forage species (Humphreys and Riveros 1996). The reduction is mainly through reduced number of flowers, pods and seeds per pod. At Turbo, most of the pods were visibly having little seed, a sign of poor seedset due to moisture stress. The greatest amount of seed was however, lost through shedding of the ripe pods as a result of the rains. Silverdleaf desmodium usually has a heavy mat of vegetative cover at seed maturity and its difficult to recover the pods unless dessicants are used to reduce the vegetative cover. On Mr Elijah Kiptum’s farm Silverleaf desmodium failed to flower during the first year despite having been planted at the same time with those on Mr. David Sang and Daniel Cheruiyot’s farms. The reasons for the failure were unclear, although unlike other farms, the field was previously under wattle trees for many years and was feasibly poor in soil fertility. On Mr Peter Ngetichi’s farm the crop performed well but the farmer was more interested in grazing the crop than leaving it for seed. Appeals to keep the crop for seed were fruitless.
Silverleaf desmodium produced better yields in Kibabii, Kanduyi division of Bungoma district than in Turbo (Table 1). Seed yield across farms varied greatly but 62% of the farmers obtained yields in the range of 50 to 136.7 kg/ha. This yields are within the range of 50 to 330 recorded at Kitale under research managed conditions (Muyekho and Wandera 1996). Almost all the farmers agreed that they lost a lot of seed through harvesting and processing. The others either harvested late when the crop had shed most of the seed or did few harvests (Table 1). At Rongo and Kanyamkago sites of Migori district mean seed yields across farms were higher for the Kanyamkago site than for Rongo (Table 2). This could be attributed to the heavy rains experienced at Rongo site (1600mm per year) as compared to 1200mm per year at Kanyamkago. Most of the rains occurred both during the vegetative stage resulting in luxuriant herbage production at the expense of seed production and at harvesting hence resulting in great losses of the available pods. The mean yields especially for the first year were much lower than that achieved on most farms at Kibabii. Low yields in Migori was mainly attributed to the prevalence of insect pests which fed on the floral parts and pods during the flowering and seed setting phases. It is not clear whether insect pests in Migori are more than those in Bungoma.
Stylosanthes species produced significantly higher seed yields at Bukembe in Bungoma district than in Migori district (Tables 2 and 3). At Bukembe, more than 71% of the participating farmers obtained yields above 300 kg/ha (Table 3). This yield margin is comparable to those recorded at Alupe, Busia district by FAO in 1985, while for S. scabra only 33% of the farmers attained yields above 300 kg/ha. Most of the yields appears to have been lost through harvesting as indicated by the column on expected yields in Table 3 where the seed on the ground was recovered through sweeping. The farmers who obtained yields above 750 kg/ha also indicated that they tried to recover most of the fallen seed. At this site, hailstones fell when the seed was ripe and this could have greatly contributed to the amount of seed shed. The study suggests that Bungoma district is more ideal for Stylosanthes seed production than Migori and that timely harvesting and recovering of seed from the ground could greatly increase seed yields.
Seed yield of Dolichos lablab is shown in Table 3. The yield variation was mainly due to the number of harvests made by farmers. This is because Dolichos has prolonged flowering and hence the sequential seed setting and maturity. Macrotyloma axillare produced seed yields ranging from 30 to 75 kg/ha at the 2 sites in Migori district but failed to produce seed at the Bungoma district site. In Bungoma Axillare produced a lot of herbage with very few flowers. Furthermore, the flowers were heavily shaded and failed to form pods. Meadley and Milbourn (1971) while working with vining pea (Pisum sativa) and Muyekho (1993) with Medicago spp observed that dense swards or heavily shaded crops produce fewer flowers of which only a small proportion survive to form mature pods. These observations could partly explain the failure for Macrotyloma axillare to form seed. In Migori district, although Macrotyloma axillare produced a lot of herbage, use of trellis encouraged the crop to form more flowers and seed. The study suggests that use of trellis to expose more flowering points to better micro-environment encourage seed set in axillare.
uncinatum seed yields on different farms at Kibabii, Bungoma district
Average seed yield for the various forage legume species at Rongo and Kanyamkago
in Migori district of South Nyanza, Kenya.
Means followed by the same superscripts within a column and year are not significantly different at P<0.05 according to Duncan multiple range test.
Three scenerios based on the cost of inputs and on the price of seed obtained by farmers in Kibabii for Silverleaf desmodium and Bukembe for Stylo guyanensis were calculated to determine if the enterprise is profitable. Silverleaf desmodium was more profitable than maize at yield above 50kg/ha of marketable seed while Stylo guyanensis was profitable at 150kg/ha (Tables 4 and 5). The greatest net benefits and benefit cost ratio relative to maize and Silverleaf desmodium were from Stylo guyanensis. Based on the seed yields in Tables 1 and 2, this results suggest that over 72% of farmers growing Stylo guyanensis made profit, while for Silverleaf desmodium only 40% made profit. However, if the same calculations are applied to the average yield obtained by farmers at Rongo and Kanyamkago (Table 3), Silverleaf desmodium would be a profitable enterprise for farmers in Rongo during the second year while at Kanyamkago site it would be profitable for the first and second year. Stylo guyanensis was unprofitable because the yields were low (less than 150kg/ha of marketable seed). The two species are perennial in nature and as a result the costs and benefits are spread over the productive life cycle of 3 to 4 years. Data on seed yields obtained from farmers at Rongo and Kanyamkago in this study and previous studies indicate that seed yields of perennial forage legumes increases after the establishment year. The economic benefits calculated in this study for Kibabii and Bukembe are therefore only partial. The ideal situation should have been for seed to be harvested for more years. This could have enabled us to use a Net Present Value Approach to determine the discounted gross benefits less discounted total costs over the years in order to give conclusive economic evaluation.
Economic benefits of growing Desmodium uncinatum for seed when compared
to maize production at Kibabii, Kanduyi division of Bungoma district.
Economic benefits of growing Stylosanthes guyanensis for seed when compared
to maize production at Bukembe, Kanduyi division of Bungoma district.
Qualitative evaluation of the forage legume seed production technology was carried out with farmers at Kibabii and Bukembe in Bungoma district. Of the four legume species, Dolichos lablab was the easiest to grow and harvest by farmers and hence was not considered in the matrix ranking. Tables 6 and 7 show a summary of matrix ranking of the different activities involved in seed production technology of Desmodium uncinatum and Stylosanthes species. Seed harvesting and processing were generally considered to be the most difficult activities for D. uncinatum (Table 6). For S. guyanensis marketing and threshing were the activities that were considered to be difficult by most gender groups (Table 7). Majority of the gender groups did not find Desmodium seed marketing a problem because of the high local demand. However, they argued that local market alone may not be sustainable since Desmodium is a perennial crop whereby same customers may not come for seed each year. Stylosanthes were less known in the area and as a result the local market was initially low. However, based on the experiences of those farmers who fed the legumes to dairy cows and realised increases in milk yield, neighbouring farmers were convinced of its usefulness and purchased seed to establish their own plots for feeding. In Migori district, market out-lets for the produced seed was found to be in abundance as most of the neighbouring farmers were interested in establishing their own plots. Many non-governmental organizations with interest in dairy production such as Heifer International, CARE and KWAP purchased some of the seeds and distributed to other farmers in the region. Although the demand for seed appears to be there, further senzitization of farmers and stakeholders in different regions is required in order to create a sustainable market for the small-scale seed producers.
These results suggests that Kenya as a country can produce its own seed through farmer groups. However, simple technologies on harvesting and processing and sustainable market need to be addressed if farmers are to consider these as commercial enterprises.
As part of the exit strategy for farmers who had been participating in legume seed production and stakeholders involved in dairy cattle production and seed marketing, a workshop was organized (1) to give feedback on seed yields achieved by different farmers and the economic implications (2) to evaluate with farmers the key issues that made them achieve high yields (3) to present results of their own qualitative evaluation of the seed production technology and (4) to sensitise the potential seed buyers about the existence and the ability of Bungoma farmer research group in producing legume seeds.
(ii) poor management of forage legumes results in low seed yields and leads to economic losses.
(iii) forage legumes seed enterprise pays more than maize per unit area both at Bukembe and Kibabii.
(ii) The animals remain in good body health condition when fed the forage legumes.
(iii) Feeding lactating dairy cows increases the cream of milk produced.
(iv) The dung from animals fed with legumes is richer in nutrients thus improves soil fertility.
Stakeholder analysis of Stylosanthes seed production technology components
through matrix ranking (1=most difficult, 12 Easiest to handle)
Conclusion and recommendations for future research
This study indicates that Kenya can (1) produce its own legume seed through farmer groups, (2) legume seed production may be more profitable in Bungoma district than maize. However, in order to institutionalise small-scale seed production future work should concentrate in the following research-extension areas:
1. Increase the demand for legume seed through awareness creation using participatory research and technology transfer on the roles of legumes in increasing milk yield when used as protein supplements and improvement in soil fertility when used in crop intercrops.
2. Marketing research to quantify the demand for legume seed, the target species and the areas where seed is needed to enable seed producers and consumers know where to sale and purchase seed.
3. A newsletter on seed production technology and supply and marketing be started in order to promote seed production and utilization in Kenya of legume seeds within Kenya and the East African region.
4. On-farm research and training of groups of smallholder farmers on legume seed production in the sites where certain legume species produce high seed yields.
5. Development of simple and affordable harvesting and threshing equipment’s that small-scale farmers can use so use to minimise losses and also save labour for other activities.
6. In places where legumes can produce seed as well as high forage yields onfarm seed production research and training should be conducted alongside the forage evaluation to encourage farmers to produce their own seed.
FAO (1985). Forage plant development and seed production in Kenya. Project findings and recommendations. Rome AG.DP/KEN/80/003, Terminal Report. 61pp.
Humphreys L R and Riveros F (1986). Tropical pasture seed production and protection paper 8. 203pp.
KARI (1991). Kenya’s Agricultural Research priorities to the year 2000. 110pp.
Meadley J T and Milbourn G M (1971). The growth of vining peas. III. The effect of shading on abscission of flowers and pods. J. Agric. Sci. Camb. 77: 103-108.
Muyekho F N (1993). Environmental and agronomic factors affecting seed production in annual medics. PhD Thesis. The University of Adelaide Australia.
Muyekho F N (1996). Forage legume seed production: Opportunities and challenges for research and on-farm production in Kenya. In: Focus on Agricultural research for sustainable development in a changing economic environment. Proceedings of the 5th KARI Scientific Conference/Edited by P.O. Fungor and G.C.O. Mbadi, Nairobi, Kenya. pp645-657.
Mwangi D M, Giller K E, Thorpe W and Thomas D (1998). Effect of intercropping with herbacious legumes on the yield and quality of Napier grass in Central highlands of Kenya. Paper presented to NARP II KARI/DFID end of project conference. Pp421-422
National dairy development project (NDDP, 1994). Review report of the survey with dairy evaluation and Advice Form (DEAF).
National Development Plan (1997-2001), Government of Kenya
ReesD J, Nkonge C, Wandera J L, Mason V and Muyekho F N (1998). Participatory rural appraisals of the farming systems of the north of the Rift Valley Province, Kenya, 1995-7. Kitale: Kenya Agricultural Research Institute. 169pp.
Staal S, Chege L, Kenyanjui M, Kimari A, Lukuyu B, Njumbi D, Owango M, Tanner J, Thorpe W and Wanbugu M (1997). A cross-sectional survey of Kiambu district for the identification of target groups of smallholder dairy producers. KARI?ILRI Collaborative Research Report. ILRI, Nairobi Kenya.
Snyders P J M, Nahuis A, Wkesa F and Wouters A P (1992). Yield and quality of mixture of Napier grass and greenleaf Desmodium at two cutting regimes. Kenya Agricultural Research Institute (KARI) Report. National Animal Husbandry Research Centre, Naivasha. 38pp.
Tothill J C (1986). The role of legumes in farming systems of sub-saharan Africa. Proceedings of a Workshop held at ILCA, Addis Ababa Ethiopia, 16-Sept. 1985 ‘on the potentials of forage legumes in the farming systems of sub-saharan Africa. Pp. 162-182.
Valk Y S (1990).
Review report of the Death-surveys during 1989. Ministry of Livestock Development,
Contact details for Dr. Muyekho: