D.M. Mwangi1 and C. Wambugu2
1. Corresponding author, Kenya Agricultural Research Institute (KARI)
P.O. Box 30148, Nairobi, Kenya
2. International Centre for Research in Agroforestry (ICRAF)
P.O. Box 27, Embu, Kenya
Poor persistency of the legumes when grown together with Napier grass
Experience with lain Calliandra calothyrsus in Central Kenya
Central Kenya constitutes 18% of the land area of the country; it holds about 64% of the human population. The population density range is wide, ranging from about 100 persons per kilometre square in the dry lowlands to over 1000 persons per kilometre square in areas with high agricultural potential (CBS 1994). Agriculture is the main economic activity with coffee (medium to low altitude) and tea (high altitude) as the major cash crops. The farming system is predominately mixed with livestock, especially dairy production being important. In a recent farm survey (Staal et al. 1998), farmers ranked dairy production second only to cash crops in economic importance.
In Kenya, it is estimated that 80% of the marketed milk comes from smallholder mixed farms (DANIDA/MoLD 1991; Mbogoh 1984) which are mainly family farms with less than ten hectares (ha) of land (Gitau et al. 1994; Mwangi 1994) and less than ten dairy animals (Anon 1985; Anon 1987). Due to the high human population pressure, farms are small with average holdings of 0.9–2.0 ha per household (Gitau et al. 1994; Mwangi 1994) and are rapidly decreasing in size due to subdivision. Animals are therefore confined in stalls and fed on Napier grass (Pennisetum purpureum) in a zero-grazing production system. Approximately 80% of the dairy animals in Central Kenya are kept in this system (Mwangi 1994; Staal et al. 1998). The importance of the dairy enterprise in smallholder farms has increased in recent years, due to liberalisation in the dairy subsector, which has resulted in the redistribution and increase of the overall social and economic benefits of market-oriented smallholder dairying (Omore et al. 1999). This coupled with low cash crop prices has made smallholder dairy production an important income earner in smallholder farms in Kenya.
Planted forages, the maize crop, cereal residues, natural pasture and grass harvested from public utilities (i.e. road reserves, school compounds etc.) are the major sources of livestock feeds in Central Kenya. As the size of land holdings declined due to subdivision, the contribution of pasture to livestock production has declined. Therefore, most livestock feeds come from planted forages and the cropped land. Apart from the maize crop, which is discussed below, bean haulms, weeds and fodder crops planted on soil conservation terraces are a major source of livestock feeds on many farms. Even on farms where animals are grazed, fodder including crop residues gathered from the farm is usually the main source of feed rather than pasture (Staal et al. 1998).
Napier grass is the main fodder crop in Central Kenya and is grown by over 70% of smallholder farmers in the area (Stotz 1983; Potter 1987;Bayer 1990; Mwangi 1994; Staal et al. 1998). Stotz (1983) estimated that 240 thousand ha or 4% of the total arable land on smallholder farms in Kenya was under Napier grass. A recent survey in Kiambu district in central Kenya showed that on average 0.2 ha were planted with Napier grass in households keeping cattle (Staal et al. 1998). This represents approximately 15% of all arable land on these smallholdings. Data from longitudinal recording of 21 farms in Kiambu indicate that over 40% of the dry matter available to dairy cows in the area comes from Napier grass (Table 1).
Table 1. Sources of livestock feeds in Central Kenya.
Source of feed |
Proportion of overall dry matter (DM) available |
Napier grass |
40.9 |
Dry maize stover |
17.1 |
Concentrate feeds |
12.2 |
Grass |
7.6 |
Weeds from cropped land |
6.0 |
Maize thinnings |
3.3 |
Green maize stover |
3.0 |
Banana pseudostems |
2.6 |
Poultry litter |
1.7 |
Banana leaves |
0.8 |
Banana thinnings |
0.6 |
Source: Smallholder Dairy Project (longitudinal recording in Kiambu district) unpublished data.
Maize is a staple food crop in Central Kenya and has become increasingly important as source of fodder in smallholder farms. The maize crop supplies approximately 23% of the dry matter (DM) available to dairy animals in Central Kenya (Table 1). The fodder is in the form of dry maize stover, thinnings and green stover. Methu et al. (1996) estimated that farmers grow an average of 0.36 ha of maize per season, from which they harvest 0.9 t of maize stover. Therefore, with two growing seasons/year, approximately 1.8 t of maize stover is harvested. In Central Kenya, farmers keep on average two animals (Methu et al. 1996; Staal et al. 1998). Therefore, the maize stover produced can be fed for between three and six months but intake and utilisation is limited by the inherent characteristics of maize stover. Currently, maize stover provides approximately 17% of the DM available to cows (Table 1).
Maize thinnings and green maize stover form about 6% of the DM available to cows. While green maize stover (harvested after the maize cob reaches physiological maturity) is feed when available, farmers go out of their way to plant seeds closely with the aim of thinning the extra plants for livestock feed. On-farm work on smallholder farms has shown that this practice could increase the DM yield by between 1.1 and 2.4 t DM/ha per season (Lukuyu, personal communication). As the crude protein (CP) content of the thinnings is higher than that of dry maize stover (Onim et al. 1991), high-density planting increases the quantity and quality of forage available.
Good quality Napier grass can support the production of between 7 and 10 kg of milk/animal per day (Anindo and Potter 1986; NDDP 1990) but actual production on farms is only about 5 kg/cow per day (Gitau et al. 1994). The poor performance is attributed to an inadequate year round supply of feed. Napier grass is grown with little or no chemical or organic fertiliser and the DM yield is low. The digestibility and the nitrogen (N) content of Napier grass declines rapidly as the grass matures, especially during the dry season, curtailing milk production. The utilisation of maize stover, which is the main roughage during the dry season (Said and Wanyoike 1987; Mwangi 1994), is constrained by the low CP content (Nicholson 1984; Little and Said 1987; Methu 1998). This system seems to offer itself for the integration of both herbaceous and shrubby legumes. It is apparent, therefore, that milk production in Central Kenya is limited by both the quantity and quality of feed available. Therefore, any strategies aiming at increasing milk production must address both the quantity and quality issues. In this regard herbaceous and multipurpose/shrubby legumes can contribute and are discussed below.
The benefits of integrating legumes into fodder systems have been demonstrated. A review by Saka et al. (1994) looks at the benefits in different farming systems in sub-Saharan Africa (SSA). This case study will, therefore, not attempt to review the benefits of forage legumes in animal production systems but will highlight some recent work in Kenya with forage legumes and their contribution to DM yield of fodder crops and animal performance. Experience with the participatory introduction of herbaceous and shrubby legumes into the farming systems will be discussed in detail.
The DM yield achieved will depend on the production system and the legume species used. In cases where nitrogen does not limit grass growth, tropical grasses will always out-yield legumes grown in pure stands and the yield gap may be as high as 10 t. Therefore, in a situation where population pressure is high, as in Central Kenya, pure legume plots are not envisaged.
Many workers have reported higher grass DM yields when grass is grown in a mixture with legumes (Ibrahim 1994; Mureithi et al. 1995; Shehu and Akinola 1995). Napier grass/legume work in Central Kenya did not give a higher grass DM yield (Mwangi 1999); however, when forage legumes were integrated into the forage system, the total (grass + legume) DM yield was higher than the sole Napier grass yield by between 20 and 38% (Table 2). Mureithi (1992) reported similar findings when Napier grass was grown together with Clitoria ternatea at the Kenyan coast. The higher DM yield was therefore the additive effect of the legume DM rather than its effect on grass performance. This would imply that the Napier grass/legume mixture was possibly utilising resources (soil, space etc.) more efficiently, resulting in a higher forage DM yield.
Table 2. The effect of growing Napier grass together with Desmodium intortum or Macrotyloma axillare on heifer carrying capacity of land and live weight gain by dairy heifers.
Napier grass alone |
Napier grass/ |
Napier grass/ | |
DM yield (kg/ha per year)a |
20,040 |
27,780 |
24,480 |
Intake (kg DM/day)b |
6 |
7 |
7 |
Live weight gain (LWG) (actual) (kg/day)c |
0.39 |
0.42 |
0.42 |
LWG (estimated) (kg/day)d |
0.50 |
0.56 |
0.51 |
Carrying capacity (heifers/ha per year)e |
9.1 |
9.8 |
8.6 |
Live weight gain (kg/ha per year)f |
1295 |
1690 |
1530 |
Advantage (kg live weight/ha per year)g |
– |
395 |
235 |
a. Yield in work at Muguga.
b, c and e. Calculations
based on a Napier grass/D. intortum mixture (85:15%) (Kariuki et al.
1998b).
d. Estimated live weight gain (LWG) made using the allocation of
nitrogen in organic resources for animals and crops (ANORAC) model (Thorne and
Cadisch 1998).
g. Advantage of mixture over Napier grass alone.
The potential effect of integrating forage legumes into a Napier grass mixture on animal performance is shown in Table 2. Kariuki (1998a) supplemented dairy heifers on Napier grass with Desmodium intortum and reported a higher live weight gain than with the Napier grass alone treatment. As the dairy enterprise in Central Kenya is characterised by low live weight gain in young stock, these results indicate the potential of legumes in the system. Calculations based on these results (Kariuki et al. 1998a) and on agronomic data collected in Muguga (Mwangi 1999) indicate that integrating legumes into the Napier grass fodder system, would increase the carrying capacity from 8.2 to 9.8 heifers/ha per year and total live weight gain from 1280 to 1690 kg/ha per year (Table 2).
The above-cited cases demonstrate the potential role that forage legumes (herbaceous and shrubby) can play in the livestock system, not only in Kenya but also in SSA. Unfortunately, this potential has not been translated into tangible benefits in smallholder farms mainly because of poor adoption of forage legume technologies by smallholder farmers.
Several attempts have been made to introduce herbaceous legumes on smallholder farms in Central Kenya. Desmodium intortum and D. uncinatum were introduced to smallholder farms in Central Kenya by the National Dairy Development Project (NDDP) a decade ago. The project recommendation was that the legumes should be grown and harvested together with Napier grass with an aim of improving the N supply to dairy cattle. More recently, the Kenya Agricultural Research Institute (KARI) introduced both herbaceous and shrubby legumes into the same area. The legumes were to be intercropped with food crops and planted grasses (Wandera 1995). The Legume Research Network Project (LRNP) also introduced herbaceous legumes mainly as a green manure crop in areas of Embu in Central Kenya.
Despite these and many other attempts to introduce shrubby and herbaceous legumes on smallholder farms, adoption has been low (Paterson et al. 1996a). In 1994, the NDDP reported that out of 222 farms with a total of 536 ha surveyed in Eastern Kenya, only 42 farms were growing herbaceous legumes on approximately 7.2 ha of land. However, the report did not indicate the proportion of the legume in the DM and, therefore, the importance of the legume in the system could not be determined.
This case study will, therefore, deal with the recent activities to introduce D. intortum cv. Greenleaf through the National Agricultural Research Project phase II (NARP II) and Calliandra callothyrsus through the System-wide Livestock Project (SLP). Factors that affect the adoption of the legumes and methods used in attempts to overcome constraints are discussed.
As indicated earlier, Napier grass is the main planted forage in Central Kenya. Therefore, all attempts to integrate D. intortum into the farming system have focused on its role as a companion crop to Napier grass. In this work a participatory approach was used. A survey was conducted in Kandara division of Maragua District, one of the areas where the NDDP introduced D. intortum in Central Kenya with the objective of identifying constraints to adoption and documenting farmers’ experiences with the legume. A total of 33 farmers and a total of 13 key informants (farmers involved with NDDP work) were interviewed. After the survey, on-farm studies involving 15 smallholder farmers in the area were established. The on-farm studies were looking at the effect of growing Napier grass with or without legumes on DM yield. During these studies (a period of approximately 18 months) dialogue with the farmers was maintained. The experiences highlighted below were mainly from this period of constant interaction with the farmers.
The farmers identified the major constraints to adoption (Mwangi 1999) as:
The main source of legume seeds so far in Kenya has been the small quantities supplied by researchers conducting on-farm experiments. The introduction of forage legumes has not been matched by the supply of seeds by commercial seed companies, as is the case for other crops like maize. When available, most of the seeds are imported and are expensive. D. intortum seed will cost approximately 2000 Kenya shillings (KSh) per kg (US$ 1 = KSh 80 at March 2001). Farmers grow on average 0.2 ha of Napier grass (Staal et al. 1998) and; therefore, to grow D. intortum together with Napier grass they would require 500 g of seed at a seed rate of 2 kg/ha. This amount would cost KSh 1000 (US$ 12.50). In an area where the monthly income is estimated at KSh 6664 (US$ 83.3) (Staal et al. 1998) the seed cost would take approximately 15% of the total monthly farm income. This high cost of seed and the fact that the seeds are not readily available make the otherwise good technology unattractive to farmers.
The legume seeds are small; therefore, sowing requires extra care and extra labour is required at a time (planting season) when the demand for labour for planting food crops is high. The tiny seedlings that emerge make weeding difficult. Many farmers that we worked with indicated that this weeding problem would be a major issue if they had to adopt the legume. In several instances the farmers unintentionally uprooted the legumes together with weeds. The incidences of uprooting the legume were higher where labour was hired (personal observation).
The conclusions from the study were that if the issues of availability, cost and difficulty of handling the seed and seedling were not addressed then the technology was unlikely to be adopted. An alternative method of establishing the legume through stem cuttings (vines) was envisaged. Establishment from stem cuttings was successful and the survival rate was over 90% (Mwangi 1999). The farmers could easily relate to the planting of stem cuttings, as this was similar to planting sweet potato vines, which they do all the time.
When the farmers learnt that they could establish D. intortum from stem cuttings but that the number of stems that the research team could provide was limited they started small nurseries, mainly near shallow wells on the farm or under banana plants where water and shade was available. These nurseries have now become sources of planting material, thus ending the dependency on the research team. Farmers used the materials from the nurseries to experiment with the legume. When the experiments started the only niche for the legume that was discussed with the farmers was the Napier grass stands. Farmers later planted the legume under coffee, banana and avocado trees and on soil conservation structures. This planting and experimentation by farmers would not have been possible if the legume had to be established from seed. Planting stem cuttings made weeding of the stands easier. Therefore, solving the propagation issue and putting the solution into farmers’ hands increased the potential for adoption of the D. intortum technology. It must be emphasised that difficulties with propagation are a critical constraint to fodder adoption. The wide spread adoption of Napier grass is mainly attributed to the ease with which it can be propagated (Mwangi et al. 1995).
Anecdotal information indicates that farmers not involved in the study have received planting material from farmers with nurseries and in turn have started their own nurseries. The effectiveness of the nurseries as a source of D. intortum planting material will be assessed through a planned adoption study.
As stated previously, the main constraint to animal performance in Central Kenya is the inadequate year round supply of good quality forage. Therefore, any forage introduced to the area must be fast growing and high yielding. This explains the adoption of Napier grass by the majority of smallholder dairy farmers (Mwangi 1994; Mwangi et al. 1995; Staal et al. 1998), which is likely to be used as a model to assess other forage crops.
The initial growth of D. intortum is slow especially when established from seed and cannot be compared with that of Napier grass even when stem cuttings are used. When Napier grass is established from rooted splits, the farmers can take the first cut three to five months after planting depending on rainfall but D. intortum takes between eight and nine months before a cut can be taken. During this period, Napier grass will have produced 7.7 t of DM/ha compared with only 3.7 t from the legume (Mwangi 1999).
As the legume was relatively new to the farmers, the initial low growth rate made them decide that the potential was low and that the legume was a waste of time and resources. The main question asked by the farmers at this stage was ‘will the legume ever get to a stage where we can harvest and have substantial amounts to feed the cow’? At this point the cause would have been lost if farmers were not given something to hold on to. Therefore, a decision was made to take the farmers to the research centre where they could see an already established Napier grass/D. intortum mixture. The farmers initially could not relate the established legume to the struggling seedlings on their farms and the visit had a major impact on farmer perception of the technology. As trips to the research centre might be costly and do not always reflect conditions on smallholder farms, the research team has emphasised farmer-to-farmer visits to enhance adoption of the legume technology. During these visits the farmers see the potential of the legume and always carry planting material back to their farms. Plans are underway to conduct an adoption study to determine how effective this approach has been, but anecdotal information indicates that it has been successful.
If farmers have to adopt a technology, they must be able to clearly see the benefits. Sometimes beneficial technologies are not adopted because the benefits cannot be clearly demonstrated or are long term. The major benefits of forage legumes include higher DM yields (Keya et al. 1971; Keya and Kalangi 1973; Reategui et al. 1995; Shehu and Akinola 1995; Mwangi 1999), biological nitrogen fixation (BNF) (Cadisch et al. 1989; Thomas and Sumberg 1995; Mwangi 1999), improved soil fertility and better animal performance due to the improved N supply in the diet (Kariuki et al. 1998a, b, c). Some of these benefits are difficult to demonstrate on-farm and others like soil fertility improvements are long term.
When the research team introduced D. intortum to farmers in Central Kenya, higher DM yields, higher N content yields and improved BNF were the benefits emphasised. These were the same benefits observed in on-station work (Mwangi 1999). A benefit that was observed on-station, but not emphasised by the research team when it discussed the benefits of the legume with farmers, was weed suppression by the legume when grown together with Napier grass. This turned out to be the most important benefit to the farmers, as they could easily observe and quantify it.
Napier grass is usually cut after 4–6 weeks of regrowth. After each cut the plot should be weeded. Therefore, with a minimum of four cuts per year the plot has to be weeded four times. It takes eight man-days to weed one ha of Napier grass. At the current rate of Ksh 150 (US$ 1 = KSh 78.9) per day, it would cost at least KSh 4800 (US$ 60) per year to weed one hectare of Napier grass. The saving incurred due to weed suppression by the legume is attractive to farmers and is the main benefit they see in the legume technology. Therefore, although the farmers have not realised the effect of the technology on livestock performance, which the research team emphasised, they are ready to adopt the technology because of other benefits, such as weed control.
Poor persistency of D. intortum when grown together with Napier grass was cited by the farmers as a major constraint in this technology. One farm had a substantial amount of D. intortum growing in a pure stand, but apart from this there was no trace of the legume in the Napier grass plots. It was suspected that Napier grass, which is a very competitive grass, had ‘edged out’ the legume through competition. This apparent lack of persistency could have been caused by one of several factors: (i) The management (i.e. spacing, harvesting frequency, manure application etc.) of Napier grass was not adjusted to accommodate the legume in the intercrop (ii) The legume was planted in the same row with Napier grass, tending to maximise between species competition (iii) The legume seed was drilled into an established stand of Napier grass, therefore, giving the young seedlings little chance to survive.
Since the 1980s, C. calothyrsus has been seen as a potentially important N rich supplement for increased milk production. The focus has been on integrating calliandra into the existing cropping system on smallholder farms. In this regard, several niches have been identified into which calliandra can be cultivated (Paterson et al. 1996b) including:
As with the herbaceous legumes, the benefits of growing C. calothyrsus in different niches have been clearly demonstrated in research experiments. Napier grass DM yield in the rainy season was not affected by intercropping it with C. calothyrsus (NARP 1993) but dry season DM yield was reported to increase (Nyaata 1998). Feeding trials have considered C. calothyrsus as a supplement to the basal diet of Napier grass and as a substitute to dairy meal (Paterson et al. 1996a). Supplementing milking animals with one kg of fresh C. calothyrsus increased milk yield from 10 to 10.75 kg/cow per day (Paterson et al. 1996b). Despite these benefits, adoption of C. calothyrsus by smallholder farmers has been poor. Farm survey reports attribute poor adoption of C. calothyrsus to lack of seed/seedlings but personal observation and discussion with farmers indicate that objectives of introducing the trees were not discussed sufficiently with farmers.
Although fodder trees are currently gaining popularity in smallholder farming systems, most of the seeds come from international research centres, such as International Centre for Research in Agroforestry (ICRAF), non-governmental organisations (NGOs) and localised harvests in Western Kenya and Embu (Franzel et al. 1999; Wambugu 2000). This short supply of seeds has constrained the adoption of calliandra by smallholders (Gerrits 2000; Personal observation). Farmers rarely collect or use seeds from their own farms or from their neighbours, as they still expect the tree seedlings or seeds from projects, NGOs and international centres (Franzel et al. 1999). Currently, efforts are underway to overcome this constraint by training and encouraging farmers and farmer groups to produce and use their own seeds (Gerrits 2000; Wambugu, unpublished information). Unlike the case of D. intortum, vegetative propagation of calliandra is not an option. Therefore, the way forward is for the farmers, either in groups or as individuals, to produce seeds by having seed orchards or by not coppicing a number of trees so that they can produce seeds.
A farmer needs 500 calliandra trees to feed a cow throughout the year at a rate of two kg DM/day (Paterson et al. 1996b, c), whereas often farmers have less than a hundred trees. In the Embu area, which is considered to have the highest adoption in the country, only one farmer out of 45 sampled, claimed to have enough calliandra trees to feed his cows through out the year. It is currently not uncommon to find a farmer with less than 10 calliandra trees. Therefore, the amount available to feed to livestock is small.
The major introduction of multipurpose trees was spearheaded by the NDDP in many of the dairy areas in the country. It appears that the number of trees required to effectively feed a cow and the management of the trees to maximise biomass production was not discussed (personal observation). On farms where the NDDP operated, it is common to find two or three tall trees with no efforts having been made to increase the number planted or to manage the trees to form a hedge. In the case of dairy farmers in Central Kenya, there is need to expand the number of calliandra trees to reach the recommended 500 trees/cow or to recommend appropriate feeding practices to farmers which make best use of the limited quantity available.
After evaluating forage legumes and validating the results on-farm, the main challenge is the scaling up and scaling out of the technology. If a technology is to have impact in an area, then many more farmers than those involved in the pilot/testing group must adopt the technology. In the past, introduction of both herbaceous and shrubby legumes has been through individual farmers. This approach has its limitations, especially where nurseries have to be established.
In the scaling up and scaling out stage, the group approach was adopted. The group approach means that the farmers can share the cost in terms of labour or inputs required for the nursery. The presence of many organised farmer groups (i.e. self-help groups, farmer co-operatives etc.) provides a favourable environment for awareness creation, training and distribution of planting materials. Participatory approaches are being used in training and working with farmers and partners from public extension services, community-based organisations and NGOs. This is aimed at developing and enhancing strategic partnerships with farmers, researchers, extension agents (both public and private) and farmers. In this approach, SLP is working with 150 farmer groups, 4 local NGOs, 1 international NGO and 10 community-based organisations (Wambugu 2000) to disseminate C. Calothyrsus.
D. intortum integration with Napier grass was a success story in a small area and with a few farmers. The challenge now is that of scaling up. The Smallholder Dairy Project (SDP) is a collaborative effort between the Ministry of Agriculture and Rural Development (MoARD), KARI and the International Livestock Research Institute (ILRI) with funds from the Department for International Development (DfID). The project goal is to improve access by poor dairy farmers to goods, services and favourable market conditions and of other farmers to knowledge services. The D. intortum technology has the potential to contribute to the project goal by improving the feed supply to dairy cattle and, therefore, SDP is involved in the scaling up and scaling out process.
Geographic information system (GIS) tools were used to determine areas in the country where the legume would grow based on biophysical conditions. Furthermore, areas where adoption of the technology was most likely were identified using market accessibility, production systems and levels of intensification. After the areas were identified, the group approach was adopted. In areas where SLP was working to promote C. calothyrsus a decision was made to link into the same farmer groups. This reduced costs in terms of the time and resources required to mobilise farmer groups and to characterise them. Moreover, it also gave the farmers a choice in terms of the forage legume they could plant.
One of the lessons learnt from this work is that availability and cost can curtail the adoption of an otherwise excellent technology. In both cases presented here, availability of planting material and the ease of propagation have been shown to be critical issues that must be addressed if the technologies are to be adopted by farmers. Giving farmers free seedlings or seeds from international research centres, research projects or NGOs reduce the farmers’ abilities to use other resources available at farm level, as they will expect more free planting material. This was and might still be the case with C. calothyrsus in Central Kenya. On the other hand, it will be futile to introduce the technology without making some planting material available. Therefore, the aim should be to move the farmers from the point where they are dependent on international centres and NGOs to a point where they can handle the technology with little or no intervention from outside their systems. Consequently, it would be wise to address the problems of germplasm availability and propagation early in the phase of forage technology development, otherwise adoption will be poor.
Adoption of any technology is enhanced when farmers can easily see the benefits. Therefore, short-term benefits should be emphasised although the long-term ones should also be mentioned. Visits to research centres or to other farmers will help farmers to visualise the potential the technology has. In situations where ‘mother–baby’ trials can be set up, they should be encouraged as the mother trials help the farmer to see the potential of the technology in their own environment rather than in research centres where the situation might be very different. In mother–baby trials an on-station type trial (mother trial) is established in the target area. Biological data are collected from this trial but farmers are encouraged to observe what is happening and to pick a few of the species/cultivars of forage planted in order to experiment with them further on their farms in the baby trials.
Working with farmer groups might enhance adoption of technologies, as the farmers share experiences, labour and other inputs required for the technology to succeed. It may enhance wider adoption even if the most proactive farmers are upset since they feel they are making effort for the weaker farmers who gain benefits they feel should remain with them. It is necessary to tell the farmers the minimum they would have to do to see any effects/benefits of the technology.
Germplasm evaluation should be carried out in the system where the forage will be grown. Currently, forage legumes are evaluated in small plots and in pure stands while maybe the eventual aim is to grow them together with companion grass. In the current method of evaluation, DM yield is the main factor considered while the ability to fit into a system (i.e. intercropping) might be a better issue to consider.
The participatory approach is expensive and time consuming but the benefits might outweigh the cost.
Anindo D.O. and Potter H.L. 1986. Milk production from Napier grass (Pennisetum purpureum) in a zero-grazing feeding system. East African Agricultural and Forestry Journal 52:106–111.
Anon. 1985. Ministry of Livestock Development. Annual Report. Ministry of Livestock Development, Nairobi, Kenya.
Anon. 1987. Fertiliser use recommendation project: Description of the first priority sites in the various districts (Murang’a District). Report 21. Ministry of Agriculture. National Agricultural Laboratories, Kenya.
Bayer W. 1990. Napier grass—a promising fodder for smallholder livestock production in the tropics. Plant Research and Development 31:103–111.
Cadisch G., Sylvester-Bradley R. and Nosberger J. 1989. 15N-based estimation of nitrogen fixation by eight tropical forage-legumes at two levels of P:K supply. Field Crops Research 22:181–194.
CBS (Central Bureau of Statistics). 1994. Kenya population census 1989. Volume 1. Office of the Vice-President and Ministry of Planning and National Development, Nairobi, Kenya.
DANIDA (Danish International Development Agency)/MoLD (Ministry of Livestock Development). 1991. Kenya dairy master plan. Study report. DANIDA and MoLD, Nairobi, Kenya.
Franzel S., Arimi H., Murithi F. and Karanja J. 1999. Calliandra calothyrsus: Assessing the early stages of adoption of a fodder tree in the highlands of Central Kenya. AFRENA Report 127. ICRAF (International Centre for Research in Agro-forestry), Nairobi, Kenya. 19 pp.
Gerrits A. 2000. Dissemination of fodder materials in the Central Highlands of Kenya. In: Calliandra calothyrsus and other fodder trees: Moving forward in research and development. Oxford Forestry Institute, Oxford, UK and ICRAF (International Centre for Research in Agro-forestry), Nairobi, Kenya. pp. 12–14.
Gitau G.K., O’Callaghan C.J., McDermott J., Omore A., Odima P.A., Mulei C.M. and Kilungo J.K. 1994. Description of smallholder dairy farms in Kiambu district, Kenya. Preventative Veterinary Medicine 21:155–166.
Ibrahim M.A. 1994. Compatibility, persistence and productivity of grass-legume mixtures for sustainable animal production in the Atlantic zone of Costa Rica. PhD thesis, Wageningen Agricultural University, Wageningen, The Netherlands.
Kariuki J.N., Boer H., Tamminga S., Gitau G.K., Gachuiri C.K. and Muia J.M. 1998a. Rumen degradation and intestinal digestion of protein in Napier grass and other Kenyan forages. Animal Feed Science and Technology (in press).
Kariuki J.N., Gitau G.K., Gachuiri C.K., Tamminga S. and Muia J.M.K. 1998b. Effect of supplementing Napier grass with desmodium and lucerne on intake and weight gains in dairy heifers. Livestock Production Science 60(1):81–88.
Kariuki J.N., Tamminga S., Gachuiri C.K., Gitau G.K. and Muia J.M.K. 1998c. Intake, DM degradation and rumen fermentation as affected by varying levels of desmodium and sweet potato vines in Napier grass fed to cattle. Animal Feed Science and Technology (in press).
Keya N.C.O. and Kalangi D.W. 1973. The seeding and superphosphate rates for the establishment of Desmodium uncinatum (Jacq) DC by oversowing in uncultivated grasslands of Western Kenya. Tropical Grasslands 7:319–325.
Keya N.C.O., Olsen F.J. and Holliday R. 1971. The role of superphosphate in the establishment of oversown tropical legumes in natural grasslands of Western Kenya. Tropical Grasslands 5:109–116.
Little D.A. and Said A.N. (eds). 1987. Utilization of agricultural by-products as livestock feeds in Africa. Proceedings of a workshop held at Ryall’s Hotel, Blantyre, Malawi, September 1986. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia. 179 pp.
Mbogoh S.G. 1984. Dairy development and internal marketing in sub-Saharan Africa: Some preliminary indicators of policy impacts. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia.
Methu J.N. 1998. Strategies for utilisation of maize stover and thinnings as dry season feed for dairy cows in Kenya. PhD thesis, University of Reading, Reading, UK.
Methu J.N., Owen E., Abate A., Mwangi D.M. and Tanner J. 1996. Smallholder dairying in Central Kenya Highlands: Practices in the utilization of maize stover as a feed resource. In: Fungoh P.O. and Mbadi G.C.O. (eds), Focus on agricultural research for sustainable development in a changing economic environment. KARI (Kenya Agricultural Research Institute), Nairobi, Kenya. pp. 243–251.
Mureithi J.G. 1992. Alley cropping with leucaena for food and fodder production in smallholder farms in lowland Coastal Kenya. PhD thesis, University of Reading, Reading, UK.
Mureithi J.G., Tayler R.S. and Thorpe W. 1995. Productivity of alley farming with leucaena (Leucaena leucecaphala Lam. de Wit) and Napier grass (Pennisetum purpureum K. Schum) in coastal lowland Kenya. Agroforestry Systems 31:59–78.
Mwangi D.M. 1994. Survey on feeds, feeding and livestock management in Central Kenya. In: Proceedings of the 2nd KARI scientific conference. KARI (Kenya Agricultural Research Institute), Nairobi, Kenya.
Mwangi D.M. 1999. Integration of herbaceous legumes into Napier grass fodder systems in Central Kenya: constraints and potential. PhD thesis, University of London, London, UK.
Mwangi D.M., Thorpe W., Methu J.N., Chui J. N., Musembi F. K., Lukuyu M., Chege L., Mutuota J. and Odongo N.E. 1995. Factors affecting the adoption of planted forages in the Nairobi milk shed: Report on participatory rural appraisals conducted in Kiambu, Kenya. Kenya Agricultural Research Institute (KARI), National Agricultural Research Centre, Muguga, Kenya.
NARP (National Agro-forestry Research Project). 1993. National Agro-forestry Research Project. Kenya Agricultural Research Institute (KARI) Regional Research Centre—Embu. AFRENA. Report 69. Nairobi, Kenya.
NDDP (National Dairy Development Project). 1990. Dairy Report: National Dairy Development Project. NDDP, Naivasha, Kenya.
Nicholson J.W.G. 1984. Digestibility, nutritive value and feed intake. In: Sundtol F. and Owen E. (eds), Straw and fibrous by-products as feed. Elsevier, Amsterdam, The Netherlands. pp. 340–372.
Nyaata O.Z. 1998. Management of Calliandra calothyrsus in association with Pennisetum purpureum for increased dry season forage production on smallholder dairy farms in Central Kenya. PhD thesis, University of Reading, Reading, UK.
Omore A., Muriuki H., Kenyanjui M., Owango M. and Staal S. 1999. The Kenya dairy subsector. A rapid appraisal. MoA (Ministry of Agriculture)/KARI (Kenya Agricultural Research Institute)/ILRI (International Livestock Research Institute), Nairobi, Kenya.
Onim J.F.M., Otieno K. and Getz W. 1991. Development of Maseno double cobber maize cultivar for high forage and grain yield. In: Proceedings of the ninth Small Ruminant CRSP scientific workshop held at Ketri-Mugaga, 27–28 February 1991. KARI SR-CRSP (Kenya Agricultural Research Institute, Small Ruminant–Collaborative Research Support Programme), Nairobi, Kenya. pp. 108–116.
Paterson R.T., Kiruiro E. and Arimi H.K. 1996a. The use of Calliandra calothyrsus for milk production. NARP (National Agro-forestry Research Project), Embu, Kenya.
Paterson R.T., Roothaert R. and Kariuki I.W. 1996b. Fodder trees in agro-forestry: The Kenya approach.
Paterson R.T., Roothaert R., Nyaata O.Z., Akyeampong E. and Hove L. 1996c. Experience with Calliandra calothyrsus as a feed for livestock in Africa. In: Evans D.O. (ed), The genus Calliandra. Proceedings of an international workshop. Winrock International Institute for Animal Agriculture, Bogor, Indonesia.
Potter H.L. 1987. Inventory of feed resources for the smallholder farmer in Kenya. In: Proceedings of the 2nd PANESA Workshop. IDRC (International Development Research Centre), Nairobi, Kenya. pp. 2–22.
Reategui K., Vera R.R., Loker W.L. and Vasquez M. 1995. On farm grass-legume pasture performance in the Peruvian rainforest. Experimental Agriculture 31:227–239.
Said A.N. and Wanyoike M.M. 1987. The prospect of utilizing urea-treated maize stover by smallholders in Kenya. In: Little D.A. and Said A.N. (eds), Utilization of agricultural by-products as livestock feeds in Africa. Proceedings of a workshop held at Ryall’s Hotel, Blantyre, Malawi, September 1986. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia. pp. 15–26.
Saka A.R., Haque I., Said A.N., Lupwayi N.Z. and El-Wakeel A. 1994. Forage legumes in crop–livestock systems of sub-Saharan Africa. Environmental Sciences Working Document 24. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia. 82 pp.
Shehu Y. and Akinola J.O. 1995. The productivity of pure and mixed grass-legume pastures in the northern Guinea savanna zone of Nigeria. Tropical Grasslands 29:115–121.
Staal S., Chege L., Kenyanjui M., Kimari A., Lukuyu B., Njubi D., Owango M., Tanner J., Thorpe W. and Wambugu M. 1998. A cross-sectional survey of Kiambu District for the identification of target groups of smallholder dairy producers. KARI/ILRI collaborative project research report, Nairobi, Kenya.
Stotz D. 1983. Production techniques and economics of smallholder livestock production systems in Kenya. Farm management handbook of Kenya. Ministry of Livestock Development, Nairobi, Kenya.
Thomas D. and Sumberg J.E. 1995. A review of the evaluation and use of tropical forage legumes in sub-Saharan Africa. Agriculture, Ecosystems and Environment 54:151–163.
Thorne P.J. and Cadisch G. 1998. ANORAC. Allocation of nitrogen in organic resources for animals and crops. NRI (Natural Resources Institute), Chatham, UK.
Wambugu C. 2000. Dissemination of fodder materials in the Central Highlands of Kenya. In: Calliandra calothyrsus and other fodder trees: Moving forward in research and development. Oxford Forestry Institute, Oxford, UK, and ICRAF (International Centre for Research in Agroforestry) , Nairobi, Kenya. 9 Pp.
Wandera J.L. 1995. Pasture/Fodder Research Program. National Agricultural Research Centre, KARI (Kenya Agricultural Research Institute), Kitale, Kenya.
