Integrated Soil Fertility Management in Bean-Based Cropping Systems of Eastern, Central and Southern Africa

Common bean (Phaseolus vulgaris L.) is an important grain legume in Eastern, Central and Southern Africa (ECSA), where it is grown on over 3.7 million of hectares every year. In this region, bean consumption per capita exceeding 50 kg a year is perhaps the highest in the world, reaching over 66 kg in densely populated western Kenya (Wortmann et al., 1998). Bean provides the essential dietary protein, fiber and income to at least 100 million people in Africa (Kimani et al., 2001). Beans are grown primarily as a food crop and to generate income by smallholder, resource-poor farmers, in holdings that rarely exceed 1.5 hectares. They are grown in pure stands or in association with maize, bananas, or root or tuber crops, and in recent years, between rows of fruit crops, banana and coffee, especially in the early establishment phase of these traditional cash crops. About 22% of the production area is solecropped, 43% is in association with maize, 15% with bananas, 13% with root and tuber crops, and 7% with other crops (Wortmann et al., 1998). In southern Africa, beans are either grown in pure stands (42%) or in association with maize (47%), or to a lesser extent, with root crops (6%) or other crops (5%). Production is mainly rain-fed, except in Mauritius and the Nile Valley of Sudan where beans are grown as an irrigated crop. In lowland areas of Madagascar, Malawi, Mozambique, and DR Congo, beans are sown after another crop in order to use residual moisture and to take advantage of lower temperatures during the winter months.


Introduction
Common bean (Phaseolus vulgaris L.) is an important grain legume in Eastern, Central and Southern Africa (ECSA), where it is grown on over 3.7 million of hectares every year.In this region, bean consumption per capita exceeding 50 kg a year is perhaps the highest in the world, reaching over 66 kg in densely populated western Kenya (Wortmann et al., 1998).Bean provides the essential dietary protein, fiber and income to at least 100 million people in Africa (Kimani et al., 2001).Beans are grown primarily as a food crop and to generate income by smallholder, resource-poor farmers, in holdings that rarely exceed 1.5 hectares.They are grown in pure stands or in association with maize, bananas, or root or tuber crops, and in recent years, between rows of fruit crops, banana and coffee, especially in the early establishment phase of these traditional cash crops.About 22% of the production area is solecropped, 43% is in association with maize, 15% with bananas, 13% with root and tuber crops, and 7% with other crops (Wortmann et al., 1998).In southern Africa, beans are either grown in pure stands (42%) or in association with maize (47%), or to a lesser extent, with root crops (6%) or other crops (5%).Production is mainly rain-fed, except in Mauritius and the Nile Valley of Sudan where beans are grown as an irrigated crop.In lowland areas of Madagascar, Malawi, Mozambique, and DR Congo, beans are sown after another crop in order to use residual moisture and to take advantage of lower temperatures during the winter months.
Bean is produced essentially in the Eastern and Central African highlands, where the population density is the highest (Wortmann et al., 1998).It is produced by smallholder farmers with few resources to allocate to soil improvement.In the Great Lakes region for example, Dreschsel et al. (1996) report extremely low use of mineral fertilizer, only about 0.4 kg ha −1 .Moreover, because of the high population density, farmers are faced with rapid soil fertility decline as a result of continuous cropping and inappropriate cropping systems with very little or no external nutrient input to replenish soil fertility.Therefore, bean yield is generally low in most regions and is most likely to decline because of the ever increasing population density.In fact, under current farming systems in small holders' fields, soil nutrient balances are negative (Bationo et al. 2006), except in banana based systems (Wortmann and Kaizzi, 1998).Although bean grain yields are variable across countries and regions, they generally vary from 200 kgha -1 in less favorable environments to 700 kg ha -1 in more favorable environments when grown in pure stands, and about half of this when intercropped (Kimani et al., 2001).
In ECSA, low soil fertility is the most important yield-limiting factor in most of the beanproducing regions.The major soil fertility related problems are found to be low available phosphorus (P) and nitrogen (N), and soil acidity, which is associated aluminum (Al) and manganese (Mn) toxicity.According to the Atlas of common bean production in Africa (Wortmann et al., 1998), P is deficient in 65 to 80% of soils and N in 60% of soils in bean production areas of Eastern and Southern Africa, while about 45 to 50% of soils are acidic with a pH less than 5.2, containing high levels of either Al or Mn.The details on the importance of the edaphic stresses for common bean production are presented in table 1.Low soil fertility causes considerable losses in productivity as the estimated bean production losses due to edaphic stresses in the ECSA are about 1,128 million tons per year

Concept of ISFM
Integrated soil fertility management (ISFM) is an approach that stresses sustainable and cost-effective management of soil fertility (Sanginga and Woolmer, 2009;Vanlauwe et al., 2010).This soil fertility strategy relies on a holistic approach that embraces the full range of driving factors and consequences of soil degradation-biological, chemical, physical, social, economic, health, nutrition and political (Bationo et al., 2006).ISFM attempts to make the best use of inherent soil nutrient stocks, locally available soil amendment resources and mineral fertilizers to increase land productivity while maintaining or enhancing soil fertility.Vanlauwe et al. (2010) define ISFM as 'A set of soil fertility management practices that necessarily include the use of fertilizer, organic inputs, and improved germplasm combined with the knowledge on how to adapt these practices to local conditions, aiming at maximizing agronomic use efficiency of the applied nutrients and improving crop productivity.A conceptual diagram is shown in Figure 1.

Local soil fertility diagnostic tools
Soil fertility can vary drastically from one end of field to the other (Vanlauwe, 2006).Therefore, response to applied soil management and external inputs can vary accordingly and no single recommendation can be made for a whole farm.As ISFM technologies are generally complex, labour intensive and costly, a more accurate intervention and recommendation system is crucial to improve the chance of adoption of these technologies by farmers.To assure farmers get the maximum return from the investment in inputs for Fig. 1.Conceptual relationship between the agronomic efficiency of fertilizers and organic resource and the implementation of various components of ISFM, (Vanlauwe et al., 2010) soil improvement, it is important that recommendations well target specific local conditions.Hence the indicator of soil quality is important for local communities to better manage their soil resources though better decision making (Barrios et al., 2001).These have to be simple enough for use by farmers and extension personnel.Dominant plant species on a farmland have the potential to integrate changes in soil quality, reflecting changes in the physical, chemical and biological characteristics of the soil (Pankhurst et al., 1997).
There have been research initiatives from both national programs and PABRA to integrate farmers' perceptions of soil fertility in simple soil fertility assessment.In South Kivu province of D.R. Congo, Ngongo and Lunze (2000) tested bean response to fertilizer application on soils with varying levels of weed infestation.Field trials were conducted for two seasons to test the effect of compost applied at 20 t ha -1 rate.Where Gallisonga parviflora is the dominant weed species, soil nutrient levels were high, bean (Phaseolus vulgaris L.) yield was high and did not increase further compost application.Where Pennisetum polystachia is the dominant species, soil fertility and bean productivity were low.Bean yield on these fields was increased considerably with the application of compost.The results (Table 2) confirm the farmer's perception that Pennisetum polystachia is an indicator of low soil fertility.Conyza sumatrensis and Bidens pilosa indicate intermediate level of soil fertility and response to compost was also intermediate.
Similar results were reported by Ugen and Wortmann (2006)   A more detailed assessment of soil chemical properties based on weed species density was done in Uganda by Ugen, et al. (1999).In this study, they made observations on 39 fields with annual crops over 4 locations in Eastern and Central Uganda.Densities of weed species relative to the total weed population were determined and surface soil samples were collected and analyzed for organic carbon (OC), soil pH, available P, K, Ca and Mg and total N, P and K levels.Oxalis latifolia was correlated with OC content, as this weed was unlikely to be present when percentage C was less than 2.3, and increased in density as soil C increased.Soil pH was well correlated with both Sorghum halepense and Oxalis latifolia.Sorghum halepense was found to be a good indicator for soil pH, being generally absent when soil pH was less than 5.8, while Oxalis latifolia increased in importance when soil pH increased.Eleusine indica was not likely to occur when soil P availability was less than 13 mg kg -1 and increased with increased P availability, while the total P was positively correlated with Eleusine indica, but negatively correlated with Euphorbia hirta densities.Sorghum halepense occurred infrequently when exchangeable K was less than 0.6 cmol(+)kg -1 .Eleusine indica and oxalis increased as exchangeable K; Eleusine indica and Bidens pilosa tended to increase as total K increased.Relative densities of Digitaria scalarum, Euphorbia hirta and Cyperus esculentus L. (yellow nutsedge) w e r e n e g a t i v e l y r e l a t e d t o e x c h a n g e a b l e M g a n d accounted for 62% of variation in exchangeable Mg.Soil fertility assessment using natural weeds is the most convenient tool available to smallholder farmers and extension workers because it requires minimal training.Therefore, weed flora was used to develop a decision guide to aid farmers in the identification of areas in their fields with severe nutrient deficiencies in Uganda and Eastern DR Congo.Thus, farmers and extension workers have at their reach a quick and inexpensive way to assess the soil fertility levels and make decisions concerning soil management.

Genetic approaches
Genetic variation for abiotic stress tolerance exists within common bean germplasms in 15 national and regional bean programs (Singh, 1991;Aggarwaal, 1994;Lynch and Beebe, 1995;Rachier et al., 1999;Rao 2001;Beebe et al., 2009), and significant number of new lines bearing these traits were developed at CIAT headquarters in Cali, at CIAT Africa regional breeding program hosted by the University of Nairobi, Kenya and the national bean programs.Singh (2001) stated that development of high yielding cultivars adapted to low soil fertility and low input sustainable systems is essential to maximize yield of common bean to enhance food security, reduce production costs and generate income.PABRA considers development of low soil fertility-tolerant bean varieties as an option to increase bean yield at no additional cost (Kankwatsa et al., 2008).Thus, bean improvement for low soil fertility adaptation has become an important component of ISFM strategy for optimum bean production in low input systems of smallholder African farmers.
Screening common bean cultivars for low soil fertility tolerance is done under field conditions following a harmonized protocol (CIAT, 1994).The methodology consists of screening the same sets of bean genotypes at several locations under a single stress and combining results across sites with different stresses (Wortmann et al., 1995).Screening is done at two stress levels: moderate stress and no stress.The moderate stress corresponds to the stress level at which a well-adapted control variety under stress performs at 40 to 50% of its normal unstressed performance.The criterion for selection is essentially bean yield, but farmers' preferences and market preference were also considered.Five most popular market classes in ECSA are red kidney, red mottled, small red, white navy and pintos (Kimani et al., 2001).Such intentional market class choice was intended to link production to well-established markets and favor adoption by farmers.The potential genotypes identified are evaluated with farmers following the participatory varietal selection (PVS), a strategy developed within PABRA for heterogeneous environments where farmers have a range of preferences (Sperling et al., 1993).The selected varieties are then promoted through other participatory methods in on-farm trials.

Low soil fertility adapted bean lines or varieties of various market classes
The most popular bean types grouped in major market classes were screened and lines tolerant to edaphic stresses were selected to allow potential users to choose for their own market.Bean lines or varieties belonging to various market classes and types, grouped by www.intechopen.comtheir tolerance to low N, low P and low pH conditions are presented in table 5. Bean genotypes were evaluated for low N adaptation at two sites, Mulungu in DR Congo and Selian in Tanzania.The genotypes varied significantly in their grain yield under N deficient conditions and in their response to applied N. Without applied N, the yield varied from 695 to 1,789 kg ha -1 while with added N at the rate of 30 kg ha -1 , the grain yield varied from 1,258 to 3,139 kg ha -1 at Mulungu.Without N, most lines gave significantly higher yield than the local sensitive check Kirundo, and previously selected tolerant variety, MwaMafutala.Adapted from Lunze et al., 2007 Table 6.Bean lines and varieties resistant to Al toxicity in soil with a yield advantage over the check At Antsirabe, Madagascar, several lines outperformed the local check Soafianarana and improved check Goiano Precoce.Most genotypes selected as resistant to Al toxicity at Antsirabe showed the same at Mulungu.The following lines gave consistently better yield under acid Al-toxic soil at the two low pH sites: AND 1056-1, AND 932-A-1, ARA 8-1B, BZ 12984-C-1 and VTTT 920-26.At Gikongoro, Rwanda many lines gave higher yield than the Al-resistant checks 7/4, Acc and RAB 478.However AND 93-A-1, BZ 12894-C-1 and AFR 593-1 were also identified as resistant to Al at the other two low pH sites, Mulungu and Antsirabe (Lunze et al., 2007).
The overall evaluation across sites and stresses allowed selection of bean lines that had consistently high yield under different stresses and across sites, with identified tolerance to one, two or even all three soil constraints considered.BZ 12894-C-1, AND 932-A-1, DRK 137-1, www.intechopen.com Nm 12806-2A are outstanding across sites and in response to all stresses -high aluminum toxicity, low N and low P availability.Several other lines have manifested tolerance to two stresses: ARA 8-B-1, AFR 709-1, AFR 703-1 and AND 1055-1 are tolerant to low P and low pH: RWK 10, ARA 8-5-1, and T 842-6F11-6A-1 tolerant to low N and low pH.These appear to have multiple tolerances to edaphic stresses and good adaptation at all environments.The results show existence of tolerance to low soil fertility in market class bean types.The promising varieties identified provide opportunities for higher bean productivity on acid soils and those with limited N and P supply (Lunze et al., 2007).

Dissemination and impact
Efforts have been made to promote all the promising and potential cultivars in all regions with problem soils as their yield advantage in similar environments has been confirmed.The benefit of low soil fertility adapted bean varieties has been demonstrated in Eastern DR Congo (Njingulula, 2003;Mastaki, 2006), Malawi, Kenya and Uganda (Kankwatsa et al., 2008).In Eastern DR Congo, with very low fertility soils, Ubusosera and RWR 382 have become the main varieties grown by farmers, replacing previously grown local or improved varieties.Kimani (2005) reported that different countries have selected and widely promoted bean genotypes that are adapted to their own environment (table 7).More than 10 years later, most farmers still kept BILFA varieties because of their adaptation to marginal soil conditions (Musungayi et al., 2008).Njingulula (2003) conducted a study to assess the impact of two low soil fertility adapted varieties in Eastern Congo and has indicated that farmers who have adopted those two varieties (Ubusosera named MwaSole, resistant to Al and RWR 382, locally named MwaMafutala adapted to low N) have noticeably improved their socio-economic conditions.In this study, (1) 32% of respondents noted that bean quantity for consumption and sale was increased, (2) 28.4% of respondents said that bean become more permanent throughout the year, (3) 30% of respondents mentioned that nutrition was improved as they could eat twice a day compared to only once previously, (4) and other farmers responded they purchased livestock and various household items such as bicycles, radio, etc and paid medical care and school fees.

Lines
Countries where adopted

Use of organic resources
Organic matter based soil nutrient management is a traditional practice that continues on smallholder farms.Among the organic resources used are animal manure, compost, crop residues for soil incorporation, natural fallowing, improved fallows, relay or intercropping of legumes, and biomass transfer (Place et al., 2003).Organic manure, compost and farmyard manure are the most common inputs used to improve soil fertility by small scale farmers (Musungayi et al. 1990;Kankwatsa et al., 2008).The need for both organic and mineral inputs to sustain soil health and crop production through ISFM has been highlighted due to their positive interactions (Vanlauwe et al. 2010).

Compost and farmyard manure
FYM use is the only possible practice for many resource poor farmers to improve crop and soil productivity.FYM and compost application are the most common practices used to improve soil fertility on smallholder farms although the quantities available are inadequate to meet crop nutrient demand.Organic manure available on farm estimates on dry matter basis varied from to 3.1 to 18.9 t ha -1 in Central Kenya (Opala, 2011) and 0.6 t ha -1 in Eastern DR Congo, sufficient to cover only 25% of their land under crops every season (Musungayi et al, 1990).Thus, scarcity of organic materials on farm limits their use at recommended rates.Because of scarcity of these resources, most farmers have developed localized application rather than broadcasting.This involves applying FYM or compost in the furrow or planting hole and covering it with soil before placing bean seeds, to avoid direct contact of manure with seeds.Considerable yield increase is achieved using compost alone, and the increase is dependent on initial fertility level in soil (Table 8).
Bean yield (kg ha  The potential of organic resources, such as FYM and compost to improve bean productivity, either as source of nutrients or by improving mineral fertilizer efficiency is well established in the region and various fertilization recommendations have been formulated.However, any substantial effect of organic amendments requires very large quantities that are not readily accessible to the majority of bean growers who are smallholder farmers (Thung and Rao, 1999).

Biomass transfer
Search for alternative sources of organics and more economical resources have always been a concern of the national bean programs.Many shrubs and trees, such as Lantana camara, Tephrosia vogelii and Tithonia divesrifolia which are common on smallholder's farms in ECSA have been studied in different countries..Among all shrubs, Tithonia is the most common shrub with substantial biomass production in most countries of the region.It is grown for land stabilization along the road and for erosion control on cropland, and as plot or compound boundary, and as an ornamental plant.Tithonia use and popularization as soil improving resource has gained interest with the increasing need of intensifying bean-based production system with the recognition of the potential of this species to accumulate nutrients.Jama et al., (2000) estimated Tithonia biomass available after nine months for transfer to fields at 2 t ha -1 kg of dry matter in Western Kenya.Tithonia has very high shoot vigour with relatively high nutrient concentrations in its biomass.Concentration of N ranges from 3.0 to 4.1%, P from 0.24 to 0.56% and K from 2.7 to 4.0% (Jama et al. 2000).
Besides this advantage in nutrient concentration, the biomass of Tithonia is also known to be rapidly decomposing (Buresh and Tian, 1998) due to proper balance in lignins, polyphenols and N (Palm et al., 1997).In fact, it has the ability to extract relatively high amounts of nutrients from the soil, a property which may make the practice not sustainable in the long run due to nutrient depletion.On soils with high natural nutrient stock however, such as soils of volcanic origin of the Great Lakes region, Tithonia use has shown considerable potential in improving crop production.In most cases, as in the productions systems of a growing number of small scale farmers without livestock, Tithonia remains the only and essential source of organic material available on the farms (Rabary, 2001).
Tithonia biomass transfer is a practice extensively studied in Rwanda, Kenya, Tanzania and DR Congo for its integration into bean-based production systems.In Rwanda and DR Congo, considerable bean yield increase is obtained, 227% in Rwanda (Ruganzu and Nabahungu, 2002) and 68% in DR Congo (Ngongo, 2002).

Rate application of Tithonia biomass
Determining the appropriate rate of biomass transfer is essential for integration of Tithonia biomass use in farmers' cropping systems, either to beans grown as monocrop or as a mixed crop.The effect of the rate of application on bean yield of monocrop was evaluated in DR Congo and Rwanda (Ngongo, 2002;Ruganzu and Nabahungu, 2002;Nabahungu and Ruganzu, 2001).In Rwanda, the study was conducted on soils with pH varying from 3.8 to 4.9.The rate of application of Tithonia biomass varied from 0 to 12 t ha -1 on dry matter basis (Table 9).The optimum rate of application was determined to be 6 t ha -1 in Rwanda, and 4 to 6 t ha -1 depending on the initial soil fertility in DR Congo (Table 10).This slight difference is due to the difference in soil fertility at experimental sites, Rubona in Rwanda and Mulungu in DR Congo.In light of these results, the appropriate rate of application of Tithonia biomass on climbing bean may vary between 4 and 6 t ha -1 , which is slightly lower than the recommendation on maize of 7.5 t ha -1 in Tanzania (Ikerra et al. 2007).In the later study, they found that the optimum rate of application on maize could be set at 5 t ha -1 in the long rainy season while in short rainy season, the recommendation could be higher at 7.5 t ha -1 .The rate of application in bean-maize mixed cropping, a more common cropping system in farmers' field, was evaluated by Ngongo (2002) in DR Congo (Table 11).In this system, the best yield of both crop either sole crop or intercropped was obtained at the rate of 8 t.ha -1 ofbiomass.

Method of application of Tithonia biomass
The technique of Tithonia use was fine-tuned by Ngongo (2002), who studied the appropriate time for biomass application.The time of application prior to bean sowing varied from 0 to 7 days.The best bean yield was obtained when biomass application was made at least 4 days prior to sowing.It was concluded that there was no reason to apply Tithonia biomass 7 days before planting (Table 12).The mode of application involved applying the fresh biomass on soil surface and incorporating it into soil 24 hours later, leaving it to decompose for 4 to 7 days before sowing bean.

Biomass application 6 t ha -1
Bean yield (kg ha

Green manure and cover crops
Research and extension efforts were made to identify legumes that present farmers with as many options as possible so that they can choose cropping systems that are best suited to their needs.Many plant species have been selected and their potential benefit demonstrated using both on-station and on-farm experiments.More detailed field investigations have been done in Uganda and DRC.The green manure and cover crops are primarily used to enhance land productivity by improving N economy in the production systems.Several legume species have been evaluated and found to be well adapted to the region and produce substantial amounts of biomass.Among the potential green manure species identified were Crotalaria ochroleuca, Mucuna pruriens (L.) DC, Lablab purpureus (L), and Tephrosia vogelii Hook, which were tested extensively in on-farm experiments for their potential to improve productivity and to obtain feedback based on farmers' perception (Wortmann et al., 1994;Esilaba et al., 2001).High N concentration in Crotalaria biomass grown under N-limiting conditions indicated that large quantities of N were biologically fixed.Mean maize grain yields following Crotalaria sole crop were 180% and 240% of maize grain yields following maize in two on-station trials and nine on-farm trials, respectively.In the study done by Fishler and Wortmann (2001), grain yields of maize and bean following one season of Crotalaria fallow were 41% and 43%, respectively, and the yields were more than following a two-season weedy fallow.Grain yields of maize following a one-season fallow with Mucuna and Lablab were 60% and 50% higher, respectively, compared to maize www.intechopen.comfollowing maize.Maize and bean yield were more, although effects were small, during the second and third subsequent seasons, indicating additional benefits from the residual effects of the green manures.
In spite of the demonstrated advantages of green manure and cover crops, the adoption by farmers has always been very poor, hampered by several factors such as availability of seeds, difficulties in planting and establishing the crop, and labor intense practice such as mowing and incorporating the cover crop (Esilaba et al. 2005).Alternative methods of green manure integration on farm were studied, particularly simultaneous cultivation of green manure with main crop, bean or another crop.In fact, in densely populated regions, leaving land under green manure for the whole season is not always acceptable to farmers.Therefore, intercropping with main crop has been developed.In Eastern Uganda, Mucuna and Lablab were successfully produced by intersowing into maize at three weeks after sowing maize, although the yields of the associated maize crop were reduced by 24% to 28%, and farmers estimated the labor requirements for Mucuna and Lablab to be less than for Crotalaria (Fishler and Wortmann, 2001).In Uganda, the yields of the green manure species were reduced by 40-70% when intercropped with a food crop as compared to sole crop production and that yields of food crops were reduced by 61-87% when intercropped with Crotalaria ochroleuca G. Don.In contrast, maize grain yield response in the first season following sole-crop of green manure production ranged from 0 to 240% (Fishler and Wortmann, 2001).Production of the green manure by intercropping Crotalaria with either maize or beans was found to be feasible with little reduction in food crop yield with a mean Land Equivalent Ratio of 1.3.
Additional benefits of cover crops reported in Uganda are weed suppression, soil erosion control and control of mole rats (Fishler and Wortmann, 2001).Results from these trials also indicated that Mucuna pruriens (L.) and Lablab purpureus (L) were best for weed suppression and to control soil erosion.Also the requirements for tillage and weeding could be lower following green manure and cover crops.In addition, maize can often be planted directly in the holes left from uprooting Mucuna and Lablab, thus reducing labor requirements during the following season.Tephrosia vogelii Hook.f. was effective in controlling mole rats and there was significant adoption by neighboring farmers.

Climbing bean rotation effects
Climbing bean is by far the bean type with high biomass production and probably with high BNF, and therefore considerable green manure effect is expected.Its integration in the production system can improve productivity and sustainability of bean-based cropping systems.In fact, climbing bean develops extensive nodulation, up to three times more than bush bean (Van Schoonhoven and Pastor-Corrales, 1994).This is an indication of higher capacity for N fixation.It has been suggested that the contribution from N fixing legume in rotation could be responsible for most of the beneficial rotation effect observed in the subsequent maize crop (Sanginga, 2003;Bado, 2002;Balldock et al, 1981).Evidence of net positive soil N balance by climbing bean has been reported (Kumarasinghe et al. 1992).In this study, they reported that at the late pod-filling stage the climbing bean had accumulated 119 kg N ha -1 , 84% being derived from fixation, 16% from soil, and only 0.2% from the 15 N fertilizer.In a long-term experiment, Wortmann (2001) reported improved sorghum yield in the rotation with climbing bean and estimated that N derived from the www.intechopen.comatmosphere was 40% to 57% of plant N, depending on the calculation used.It is clear that evidence of benefit of climbing bean cultivation exists, either as rotational effects or improved N nutrition.
On-station and on-farm farmer participatory trials were conducted by Lunze and Ngongo (2011) to assess the beneficial effects of the climbing bean on the subsequent maize crop in rotation, compared to bush bean and continuous maize cropping systems.The estimate of N contribution from the climbing bean to the system estimated as N fertilizer replacement values (Bado, 2002), and these values varied from 15 kg N ha -1 to 42 kg N ha -1 in the first season of rotation.Note that this method is believed to overestimate the amount of N supplied by the legumes in the systems (Bullock, 1992).Without applied N fertilizer, the average maize grain yield increase over three cropping seasons in response to the preceding climbing bean effect were 489 kg ha -1 and 812 kg ha -1 compared to bush bean and maize as preceding crops, respectively, which is 17.5% and 33.8% increase.However better yield advantage of climbing bean over continuous maize was obtained in the long rains cropping season (Season A), 43.2% compared to 24.2% in short rains season (Season B) (Table 13).
Preceding These studies in Uganda and DR Congo provide evidence for beneficial effects of climbing bean to improve N nutrition of the following maize crop.It is presumed that climbing bean promotion is an appropriate strategy for higher productivity and sustainability on smallholder farms, as substantial gain of N will be achieved by proper integration of climbing bean in the bean-based production systems in ECSA.Additional effects of growing climbing bean are expected, such as reducing erosion because it provides more extensive soil cover through its canopy compared to bush bean.These beneficial effects are considerable on relatively medium fertility soils where climbing bean effects were highest, because of the large quantity of biomass produced.

Biological nitrogen fixation in bean
The enhancement of the capacity of beans for biological fixation through symbiosis with Rhizobium phaseoli was recognized by biological nitrogen fixation(BNF) working group within PABRA as an important option to improve the productivity of bean crop on farms (Nyabienda, 1988).This was particularly true for regions where farmers have limited access to N fertilizers.However, improvement of bean BNF requires a multiplidisciplinary approach that will require the plant breeder to increase the host capacity to fix N 2 as bean is considered as a poor N 2 fixing legume (Giller, 2001), and selection of effective Rhizobium strains that can compete for nodulation with native populations of bacteria present in most soils.

N 2 fixation variability in bean germplasm
BNF is among the strategies extensively investigated as one of the avenues to improve bean productivity.Matheson (1997) measured N 2 fixation at Kawanda and Namulonge, Uganda using two common methods: 15 N abundance method and N-difference method using nonnodulating beans lines INIAP 404 and EXRICO.N2 fixed was lower at Namulonge than at Kawanda.It ranged from 10.6 to 35.1 kg Nha -1 and from 0.7 to 20.4 kg N/ha, respectively, at Kawanda and Namulonge.This study found that the genotypes identified as tolerant to low soil N had greater N fixing capacity.Table 14 summarizes the genotypic variation in N 2 fixation of common bean at Kawanda, Rwanda.
The same author calculated the net N balance and found a negative net N balance for all varieties, ranging from -23.6 kg ha -1 to -5.3 kg ha -1 , indicating that more N had been exported in grain at harvest than had been fixed and retained in soil.The yield was negatively related to net N balance: the higher the yield, the more negative the net N balances.However, unlike bush bean, climbing beans exhibit greater N 2 fixation capacities (Wortmann, 2001;Lunze and Ngongo, 2011).Lunze and Ngongo (2011) estimated N 2 fixation by climbing bean r from 16 to 42 kg ha -1 per season.N fixation is further enhanced by several cultural practices and agronomic management, i.e. inoculation, P fertilization and liming, that are discussed below.

Inoculation
BNF is extensively investigated as one of the avenues to improve bean productivity.Rhizobial inoculation is a common practice to promote BNF in bean, given that the Rhizobium strain is well adapted to local environments.Strain CIAT 899 is currently used for bean seed inoculation in all regions of ECSA.Inoculation with this strain in Burundi showed a yield advantage up to 59% in Gitega (Ruraduma, 2002).Evaluation by the same author of local Rhizobium strains so far has failed to identify a more efficient strain.

Liming and P fertilization effects
The efficiency of BNF is further improved by liming and P fertilization.This is not always affordable for smallholder farmers.Considering these factors, bean seed pelleting with lime and phosphate was evaluated in Burundi and Malawi.On-farm trials at Ikulwe, Iganga District of Uganda, the %N 2 fixation was estimated for a well-adapted high yielding variety  MCM 5001, considering the effects of inoculation and P application on N 2 fixation.Neither inoculation nor 100 kg ha -1 of TSP (triple super phosphate) application increased bean yield, but 100 kg ha -1 TSP with inoculation yielded significantly higher yields than these treatments individually.Inoculation had therefore greater effects on bean yield at higher P levels.The estimate of % N 2 fixation without addition of P fertilizer and inoculation ranged www.intechopen.comfrom 5% to 32%, while the addition of 50 kg ha -1 of TSP increased N 2 fixation by 50%.However, they did not observe any significant differences (P<0.05) in %N 2 fixation between treatments and farms.
In Malawi, Chilimba and Kapapa (2002) evaluated bean seed pelleting with dolomitic lime.On-farm trials conducted in Dedza where soils are strongly acidic showed significant responses of four bean varieties to inoculation and seed pelleting with lime on BNF and grain yield.Seed pelleting and inoculation showed beneficial effects in increasing plant N content, nodule numbers and even grain yield.In acidic soils, both inoculation plus seed pelleting significantly improved BNF and grain yield whereas in a normal non-acidic soil, inoculation only or pelleting only significantly improved BNF and grain yield.The on-farm evaluation was conducted in Bembeke Extension Planning Area where soil pH is 4.5 and the soil is fine, kaolinitic, thermic, Kandiudafic Eutrudox while at Chitedze research station the soil pH is 5.9 and the soil is classified as fine, kaolinitic, thermic, Udic Kandhaplustalfs.The results on the evaluation on soils with varying soil pH are presented in Adapted from Chilimba and Kapapa, 2002 Table 15.Effect of inoculation and seed pelleting on % N, nodule number and the yield of beans at three locations (Marieta, Ester and and Chitedze) in Malawi The results of the above studies confirm that both lime and P fertilizer application rate can be considerably reduced by coating bean seeds to achieve relatively higher yield.This technology is considered a good ISFM strategy.

Mineral fertilizers
Most smallholder farmers in the ECSA regions are well aware of the value of mineral fertilizers, but the rate of application remains low, below the recommended rates, except for commercial farmers.The essential reason for this is the high cost of mineral fertilizers, and low profitability.In most regions however, bean comes after cereals, which are commonly fertilized so that bean benefits from residual fertilizer effects.Nonetheless, fertilizer www.intechopen.com recommendations have been developed in several countries for bean, particularly in Kenya, Tanzania, Burundi and South Africa.
On Ferrarsols (Oxisols) of Burundi, mineral fertilizer applied alone on bean was found to be non-economical, and under certain conditions bean yield was depressed.The recommended rate (kg ha -1 ) for Burundi was 15 -30 for N, 50 -60 for P 2 O 5 and 30 for K 2 O, with 2 to 5 t organic manure (Ruraduma, 2002).The ISFM strategy suggests options that combine organic resources and mineral fertilizers to achieve higher yield, economically.Fertilizer application to bean crop is developed for those grown as a sole crop or intercropped with maize.In addition to determining the rate of application, the proper time of application is important to maximize N 2 fixation, as early application of N can inhibit nodulation of bean.
In Uganda, Wortmann, (1998) recommended applying a small amount of N at the planting time and the major part of fertilizer N at the second weeding time, to favor bean productivity and N2 fixation.That is an application of 5 kg N ha -1 and 20 kg P ha -1 at sowing time and applying 35 kg N ha -1 at second weeding time for both maize and bean that are intercropped.In on-farm trials with the objective of developing appropriate fertilization practice on snap bean, Ugen et al. (2009) (Liebenberg, 2002).

Nitrogen
According to Liebenberg (2002), inoculation of dry bean seed is regarded as ineffective in South Africa.Consequently, dry beans should be considered as incapable of satisfying all of their N requirements through N 2 -fixation.The application of all the N fertiliser at planting time is recommended, particularly where non-decomposed material has been ploughed in before planting.

Phosphorus
Commercial production in South Africa showed modest yield responses to P fertiliser application in dry beans and P is not normally a yield-restrictive factor.Under subsistence production, where small quantities of fertilizer are applied, P can be a yield-limiting factor.
Where the P content of the soil is lower than 20 µg g -1 (Bray 1), it is recommended that TSP be broadcasted and ploughed into the soil to a depth of 15 to 20 cm before planting.P fertiliser must still be band-placed at the time of planting.In low pH soils, P can be utilised efficiently by band-placing 3.5 cm to the side of the row and 5 cm below the seed.Table 17.Guidelines for phosphorus fertilisation

Potassium (K)
When dry beans are grown on soils with high clay content, K is not normally a limiting factor.Deficiencies are most likely to occur on sandy soils with an analysis of less than 50 mg kg -1 K.The optimum leaf K content is 2% potassium.

Effect of cropping systems
Wortmann et al. (1996) studied the relationship between maize monoculture to fertilizer N and P and the intercrop response from 62 fertilizer response trials conducted in Kenya.Their results indicate that intercrop was more productive than the monoculture with no fertilizers applied, but overall responses of the systems to applied nutrients did not differ.Maize both in monoculture and intercropped responded more frequently to applied N than did the intercrop bean.Frequency of response to applied P was similar for both crops and both production systems.

Combination with organic manure
It is well established that combining mineral fertilizer with organic resources improves fertilizer use efficiency.In Western Kenya's Vihiga District, the recommendation is 50 N kg www.intechopen.comha -1 and 50 P kg ha -1 with 5 t ha -1 of farmyard manure (Rachier et al., 2001).The N and P rates are reduced by 50% in the following season.If money or credit is not available, in addition to 5 t ha -1 FYM, only 25 kg ha -1 N and P are recommended.The following season, the N rate is reduced by 25% and no P is applied.If FYM is available on farm, 5 t ha -1 FYM together with 25 kg ha -1 N and P are recommended.This rate is reduced by 50% in the second season.If the green manure is grown, then N, P and FYM rates are reduced by 50% in the first season, and in the second season only 25 kg P ha -1 is needed.In the study conducted by Ngongo (2002) in DRC, the aim was to reduce the quantities of both mineral fertilizer and Tithonia biomass recommended.The author evaluated bean response to application of 4 t ha -1 Tithonia biomass with varying rates of mineral fertilizer (Table 19).

Treatments
Bean grain yield (kg ha The results confirmed that less fertilizer could be recommended when Tithonia is applied.The application of 4 t ha -1 Tithonia with 20 kg ha -1 of NPK fertilizer was as efficient as the recommended mineral fertilizer rate of 60 kg ha -1 .Thus a 3-fold reduction of inorganic fertilizer is possible when Tithonia is applied.

Mijingu phosphate rock
Bean growing environments are well described in all ECSA countries and P deficiency is widely reported as the most important bean production constraint (Wortmann et al., 1998).The low P availability to plants is explained by the nature of the soils that are highly weathered with low total P and/or high P fixing capacity (Rao et al., 1999).Rock phosphate (RP) which is a resource relatively common throughout ECSA countries is recognized as a fertilizer with high potential to improve bean crop productivity having low cost, compared to conventional mineral P fertilizers.The deposits of the RP are reported in many countries and exploited to some extent in Tanzania (Mijingu), Malawi (Tundulu), Zambia (Isoka), South Africa, DR Congo (Kanzi) and Burundi (Matongo).Minjingu Phosphate Rock (MPR), a sedimentary biogenic deposit which contains about 13% total P and 3% neutral ammonium citrate soluble P is reported to be highly reactive (Jama and Van Straaten, 2006). www.intechopen.com On acidic soils of Tonga in Rwanda, Nabahungu et al. (2002) studied the effects MPR associated with limestone and green manures (GM) (Tithonia and Tephrosia biomass) on P uptake and on maize yield.This study showed that MRP significantly contributed to P increase in the soil and resulted in increased P uptake and maize yield.Green manures in combination with MPR increased P uptake significantly.The results indicated that using a combination of limestone, MPR and GM is the best strategy in improving maize productivity on acid soils (Table 20).The effect of RP alone or GM alone is low, whereas the combination of MPR with organic resources improved its effect.A combination of RP and GM produces a demonstrated yield advantage over control or each of these treatments alone.The recommendation for acidic soils of Rwanda is to combine organic resources with RP and lime (ground limestone or burnt lime) allow maize yields to increase up to approximately 6 t ha -1 .The same recommendation is applied for bean.In Northern Tanzania, the recommended rate of application is 250 kg ha -1 MRP combined with FYM at 5 to 10 t ha -1 or a GM grown in the previous season (Mucuna pruriens or Vernonia subligera).
Other practices for MRP utilization have been developed in Kenya and Tanzania.The RPfortified compost technology is well accepted by farmers in Western Kenya (Odera and Okalebo, 2009).Another option which combines PR, liming, Rhizobium inoculation and bean seeds, developed by a Phosphate rock evaluation project (PREP), known as PREP-PAC.This product comprises PR (2 kg), urea (0.2 kg), legume seed, rhizobial seed inoculants, seed adhesive and lime pellet, packed to fertilize 25 m 2 of land.In Western Kenya the use of PREP-PAC and climbing bean package increased maize and bean yield by 0.72 and 0.25 t ha - 1 respectively, resulting in a 161% return in investment (Nekesa et al., 1999).PREP-PAC use on bean for the two seasons in seven districts in Easter Uganda showed significant increases of 881 kg ha -1 in bean yield, increasing from 1,316 to 2,197 kg ha -1 (Esilaba et al., 2005).This product is commercialized, and makes the PR more accessible to smallholder farmers, thus offers opportunity for easier handling and use for increased bean productivity.Another PR from Uganda, Busumbu RP (BRP) was studied by Nabahungu et al. (2007) who reported that composting increased P availability and P recovery from BRP to the extent similar to that of TSP, as well as the availability of Ca and Mg. www.intechopen.com

Liming
Lime application is a common soil improvement practice generally recommended on acidic soils.The rate of application is based on exchangeable soil Al concentration.For tropical soils, lime recommendation is 1.5 times the exchangeable Al in t ha -1 , the rate which is sufficient to neutralize toxic Al.One of the first studies to determine the optimum rate of lime for bean was done in Malawi by Aggarwal et al. (1994).In this study, bean response to applied calcitic lime at the rate of 0, 25, 50, 75 and 100% of the exchangeable Al was evaluated in 1992-93 cropping season using 15 varieties and in 1993 -94 season using 8 varieties.Additional treatment was lime applied at 100% of exchangeable Al plus P since the soil was P-deficient.The results indicated that although the performance of the varieties was poor due to low fertility, liming caused a linear increase in nodule number, nodule weight, and grain weight up to 75% level of Al neutralization (Table 21).Finely powdered ground limestone, cheaper than lime (neutralizing value of 90 to 100) evaluated in Rwanda, was comparable to agricultural lime found in the marketplace.Thus lime recommendation was devised for ground limestone.Table 22 presents the results of bean response to lime (agricultural lime) and limestone in two seasons (Nabahungu and Ruganzu, 2001).

Decision support tools
Several technologies have been developed and widely tested with successful results, resulting in development of soil fertility management decision guide, particularly in Uganda (Farley, 1998;Esilaba et al. 2001), Kenya (Rachier et al., 2001) and Eastern Congo (Ngongo, 2001).Soil fertility management research results have been translated into recommendations that take into account the soil physical conditions as well as farmers' socio-economic status.Decision support systems (DSS) to the use of inorganic and organic sources have been developed for different environments and socio-economic conditions in bean based production systems in the region at different locations.These guides are efficient tools for realizing better bean crop yield in areas where they are developed.However, the extrapolation to other sites and location with different soil conditions, climate and population density and socio-economic conditions is not guaranteed with acceptable results as they are site specific (Esilaba et al., 2001).A model decision tool for Uganda is presented in Table 23 while for Eastern DR Congo the tool was developed using soil fertility assessment based on farmers' perception (Table 24).Other DSS are similar, but the local conditions and available resources vary.

Promotion and use of ISFM options
PABRA has developed and disseminated different technologies to address farmers' production constraints and to increase bean productivity in sustainable ways (Kimani et al., 2001).These comprise simple technologies such as new bean varieties and more complex, knowledge intensive ones such as integrated pest and disease management (IPDM) or ISFM.
The technologies and their dissemination methods were based on low cost options that were in many cases developed through collaboration with multiple partners, and engaged and empowered end users in participation and adoption (Kankwatsa et al., 2008).The spread and adoption of new bean varieties have been very impressive; reaching several million www.intechopen.comfarming families, while the spread and adoption of ISFM technologies have been slow.Scaling up and disseminating information on recommended technologies was through Farmer Research Groups using a modified farmer field school approach.Because farmers were driving the experimentation and dissemination processes, all aspects of local culture were taken into consideration (Kankwatsa et al., 2008) and they often made recommendation for improvement to suit their production circumstances, both in the social and the technological aspects: this led to the development of the ownership by the farming communities and a high rate of adoption among participating farmers.

Bush
Bush beans sole crop (4 t ha -1 Tithonia or 10 t ha -1 compost) + 50 kg ha -1 (P,K) Bush beans-maize intercropped (6 t ha -1 Tithonia or 20 t ha -1 compost) + 50 kg ha -1 (N,P,K) Climbing Climbing beans sole crop (8-10 t ha -1 Tithonia or 20 t ha -1 compost) + 50 kg ha -1 (N,P,K) Climbing beans-maize intercropped (10 t ha -1 Tithonia or 30 t ha -1 compost) + 75 kg ha -1 (N,P,K) Table 24.Decision guide for ISFM in a Bean-based cropping system in Eastern DR Congo Awareness and adoption of ISFM technologies by farmers is found to be essential factor to their eventual widespread adoption in ECSA.Kankwatsa et al. (2008) found that farmer's awareness varied significantly with their participation to the technology promotional activities of farmers.They reported that participating farmers were more rational in their choice of technologies to solve specific constraints and more participating farmers adopted the IPDM and ISFM technologies than the non-participating farmers.To favor farmers' capacity to experiment and eventually to adapt technological options to their biophysical and socioeconomic conditions, they were exposed to a whole range of new soil fertility management options through farmers' field schools and dialogue with researchers in the process of farmers' participatory research.They have continued to experiment and ended to suggest modifications to the technologies.In Bushumba, Estern DRC, Njingulula and Ngongo (2007) reported the results of participatory evaluation of the rate of Tithonia application.They found that although the rate recommended by research was not much modified, farmers found the better manner to express the rate of application using a common volume measure.Farmers' recommendation of Tithonia biomass application was the rate of 3 to 4 basins per 4 m x 4 m portion of land, which is equivalent to 5 to 7.5 t ha -1 , in perfect agreement with the recommendation by researchers.Going through this process increased their confidence in their own ability to find solutions to different problems, and improved the rate of adoption of suitable technology.The impact studies were conducted by Kankwatsa et al (2008) in PABRA member countries to evaluate the fate of promoted ISFM technologies.

Conclusions
African smallholder farming conditions are worsened by declining soil fertility as a consequence of population pressure on a limited landbase.PABRA, as an African research www.intechopen.comfor development program has made outstanding efforts in establishing partnership with numerous stakeholders, farmers, rural communities, non-governmental and governmental organizations, private sector, traders and research organizations (national, regional and international).In spite of complexity of ISFM technologies developed, bean productivity has been improved under low soil fertility conditions by developing strategies and technologies that enhance resilience to environmental stresses and by enhancing farmers' access to adapt and use cost effective integrated environmental stress management options (PABRA, 2008).The strategy of PABRA on ISFM has two broad avenues for achieving the objective.These include soil fertility management and deployment of resilient bean germplasm.
Genotypes with improved performance on low fertility soils is valued by farmers and readily adopted, and have made it possible to improve bean production in regions where they are adopted at no additional cost.Identification and use of cultivars tolerant to mineral deficiencies and toxicities are essential for reducing production costs and dependence of farmers on soil amendment inputs.The resilient bean cultivars have been shown superior to existing popular varieties developed without consideration of tolerance to soil infertility tolerance.Many bean genotypes have been selected with tolerance to single or multiple edaphic stresses (Wortmann et al., 1995;Rao, 2001).Therefore, as stated by Lynch et Beebe (l995), efficient cultivars would have several benefits, including increased food production, increased farm income, increased bean availability and consumption, and thereby improved nutrition for low income producers and consumers.Indeed, improved variety is considered as a fast adoption technology (Kankwantsa et al., 2008).The nurseries of these genotypes are maintained with the PABRA regional networks with characteristics of market preferences, besides tolerance to low soil fertility.
Bean productivity is further enhanced by various soil management options developed in participatory manner.
The use of farmer participatory approach to technology dissemination enabled farmers to be familiar with, understand, experiment and adapt ISFM technologies.Through capacity building through farmers' groups and farmer research groups, farmers have been empowered to obtain various technologies and services from appropriate partners.Various locally available resources are used, as limestone and phosphate rock, alone or in combination with mineral fertilizers, and the effect of organic resources in improving nutrient use efficiency is well demonstrated.Considerable and beneficial bean yield increase is achieved.However, the low production of organic resources on farms calls for biomass transfer.Tithonia biomass is readily available and adopted by farmers and contributes considerably to bean productivity enhancement as an essential source of nutrients and organic amendment.Maximal use of locally available nutrients through low-external input technologies, techniques, combined with optimal use of external nutrients appears to be the most appropriate strategy in the existing economic environment (de Jager, et al., 2004).
Based on research data on soil fertility management in major bean growing areas, a comprehensive decision support system (DSS) was developed for use by farmers and extension workers in different agro-ecological zones and under diverse socio-economic scenarios.In contrast to standard guidelines, the DSS provides farmers with ISFM options under different scenarios, thereby allowing farmers the ability to choose the option(s) that best suit their needs and socio-economic conditions.Soil fertility conditions are assessed www.intechopen.comusing criteria and indicators that are easy to measure, such as dominant weed species.The desired outcome is that the tool will be used by researchers, extension workers, and farmers for assessing and implementing options of using scarce resources for maintaining soil fertility and improving crop yields in bean-based cropping systems.
PABRA places special emphasis on increased access to cost effective and environmentally friendly integrated bean production, especially by female bean growers.Currently, two approaches are used to facilitate and increase farmer's access to ISFM technologies: a deliberate promotion and delivery of improved varieties and ISFM technologies as a single package; and policy advocacy and harnessing of enabling policies (including input support system) to deliver ISFM technologies to bean farmers.The Alliance (PABRA) feels confident that these approaches will greatly enhance the adoption of ISFM technologies by bean growers and overall bean production in eastern, central and southern Africa.
+C: with compost and -C: without compost

Table 3 .
Positive (+) and negative (-) relationships of the relative densities of weeds species with soil properties which may be useful in diagnosis of soil fertility status (Wortmann et al., 1995)n to local environments(Wortmann et al., 1995).At Mulungu, with no lime, most test lines gave significantly higher yield than the sensitive check variety Kirundo, while only two lines VTTT 923-6-1, HM 21-7 and AFR 593-1 outperformed the tolerant check MwaSole (Table6).The yield advantage of Al-resistant genotypes could be as high as 300% compared with the check.

Table 7
. Bean lines/varieties adapted to low soil fertility in different countries www.intechopen.com

Table 8 .
Effect of compost application to fields dominated with different weeds on bean yield

Table 10 .
Climbing bean yield response to applied Tithonia biomass in DRCongo

Table 11
. Bean and maize yield response to Tithonia biomass in sole cropping and intercropping systems in DR Congo.

Table 12 .
Effect of different application of Tithonia diversifolia on bean yield

Table 14 .
Nitrogen fixation, biomass and yield of 30 varieties grown under low N conditions atKawanda (1996)

Table 15 .
evaluated rates and timing of the mineral fertilizers CAN, NPK 17-17-17 and DAP (diammonium phosphate), which are the common fertilizers in the marketplace.In this study, application of DAP at planting and topdressing with NPK-17:17:17 at 21 days after emergence appeared to be appropriate for bean production.More complete mineral fertilizer recommendations used in South Africa are presented in Tables16, 17 and 18

Table 19 .
Effects of Tithonia biomass in combination with mineral fertilizers on bean yield in DR Congo.

Table 21 .
The yield declined at higher rate of lime application due to nutrient imbalance, which might have been induced by lime.Lime recommendation is therefore soundly based on neutralizing the exchangeable soil Al concentration, rather than increasing soil pH Effect of lime application on seed yield, nodule number, root weight and shoot weight of common bean varieties grown during 2 seasons at Bembeke, Malawi

Table 22 .
Rutunga et al. (1998)cultural lime and ground limestone at Rubona, Rwanda www.intechopen.comIntegratedSoilFertilityManagement in Bean-Based Cropping Systems of Eastern, Central and Southern Africa263Bean yield was very low without lime and doubled with lime application, but was still considered low.For better yield, lime application should be associated with organic resource application, which according toPalm et al. (1997)improves productivity on acid soils by its interactions with the mineral soil in complexing toxic cations and reducing the P sorption capacity of the soil.Examples from DR Congo, Rwanda and Burundi indicate that combining lime with organic manure results in considerable bean yield increase.Rutunga et al. (1998)studied crop rotation system of maize and beans, established at Rubona (Rwanda) from 1984 to 1992.They evaluated the effects of different types and rates of fertilizers in improving the productivity of acidic Oxisols.Continuous cropping of maize followed by beans for a period of 8 years gave no yield in control plots.A single application of 2 t ha -1 of lime increased significantly (p = 0.01) the soil pH, Ca 2+ content, cationic exchange capacity, and decreased the level of the exchangeable Al.However, this quantity of lime, when applied every two years for a period of eight years, led to overliming.The application of more than 8 t ha -1 of FYM annually, combined with 300 kg ha -1 of NPK (17:17:17) fertilizer (every six months) significantly improved soil organic C and crop production at Rubona, Rwanda.The high rate 35 t ha -1 of FYM or the combination of "lime, FYM and NPK fertilizers" gave the best crop performance.

Table 23 .
Tentative guide to fertilizer use for maize and bean in Uganda