Livestock Research for Rural Development 23 (5) 2011 Notes to Authors LRRD Newsletter

Citation of this paper

Feeding management for dairy cattle in smallholder farming systems of semi-arid tropical Kenya

D M G Njarui, M Gatheru, J M Wambua, S N Nguluu, D M Mwangi* and G A Keya*

KARI-Katumani Research Centre, PO Box 340 - 90100, Machakos, Kenya
* KARI Headquarters, PO Box 57811 - 00200, Nairobi, Kenya
donaldnjarui@yahoo.com

Abstract

Inadequate nutrition is a major constraint that impact negatively on the growth and viability of dairy cattle farming in Kenya. A study was carried out to assess feeding practices, feed availability and coping strategies among smallholder dairy farmers in semi-arid region of Eastern Kenya. A total of 136 dairy cattle farmers were interviewed between June - August 2008 using a structured questionnaire.

The study revealed that animals depended on natural pastures and home-grown fodders, mainly Napier grass and crop residues. Maize stover was the principal crop residue and was the main livestock feed during the peak of long dry season with >80% of farmers using it to feed their dairy cattle. Between 88 and 92% of farmers provided supplements inform of dairy meal concentrates to the lactating cows but the quantity was low and amount fixed (usually about 2 kg/day) throughout the lactation period and not commensurate with milk production. Majority (97.5%) of the dairy farmers interviewed practised some form of feed conservation. Grass hay was the most popular form of feed conserved but the quality was poor and quantity stored was inadequate to sustain the dairy herd kept during the dry season. Approximately 95% of dairy farmers stored crop residues for their livestock but the storage methods were inappropriate to maintain the quality. About 93% of farmers experienced seasonal fluctuation of feed availability with highest number (70-80%) reporting peak severe feed shortage in September and October. Feeding conserved fodders to dairy cattle was the most important strategy adopted by smallholder farmers to mitigate against feed scarcity followed by purchase of fodders from other farmers. To improve feed availability, farmers should maximize conservation of surplus feed resources experienced during wet seasons using simple and cost effective methods such as hay box for making hay and polythene tubes for silage.

Keywords: Coping strategy, crop residues, feed availability, feed resources, feeding practice


Introduction

Dairy farming is ranked the third most important agricultural enterprise in the semi-arid region of eastern Kenya (Itabari et al 2006). The growth of the dairy industry in the region is stimulated by high demand of milk and milk products due to the expanding population and improved incomes in the rural-urban trading centres (Njarui et al 2010). However, inadequate nutrition is a major constraint that impact negatively on the growth and viability of the dairy farming in the region (Njarui et al 2009). Inability to provide sufficient quantity and quality of feed to livestock is widespread in East Africa (Hall et al 2007) and has been reported in coastal lowlands of Kenya (Reynolds et al 1993). In Kenya, majority of smallholder farmers keep more animals than they can feed from their own land. Estimate by Reynolds et al (1996) in 1990s showed that smallholder dairy farmers produced about70% of the feed required from their own resources and this situation has remained relatively the same.

 Over 80% of farmers in the semi-arid region of eastern Kenya are smallholder (Njarui and Mureithi 2006) and practice mixed crop-livestock subsistence farming. Livestock are kept for manure, draught power in crop production, milk and security against crop failures. In Upper Midland agro-ecological zone 4 (UM4) and Lower Midland ecological zone 4 (LM4) where dairy farming is widely practiced farmers keep on average 1.2 and 2.6 dairy cattle per household, respectively (Njarui et al 2009) in addition to other livestock. In UM4, intensive dairy production system is practiced and animals are confined in sheds where feeds are delivered. In LM4, production is based on semi-intensive system where animals are confined but sometimes are allowed to graze freely on pastures. Most of the land is allocated for crop production leaving very limited land for pastures. On average, the land holdings is 2.12 ha per household in UM4 and 4.48 ha in LM4 with less than 10% under sown pastures (Njarui et al 2009). This area is insufficient to produce all the required feeds for the number of livestock kept. Farmers allocate the most fertile part of their land for food and cash crops production and the less fertile land for pasture production (Njarui and Mureithi 2006) with no or little inorganic fertilizer or manure application. The scarcity of feed is exacerbated by low (500 - 800 mm/year) and erratic rainfall, long dry season and frequent droughts. This results in cessation of growth of pastures and consequently limiting fodder production. The quality also declines significantly during the dry season and is insufficient to meet animal production potential (Thairu and Tessema 1987).

 Maintaining access to adequate quantity and quality of feed resource is crucial for milk production in dairy cattle. There is a need therefore to evaluate the feed management among the smallholder dairy farmers in order to improve feed availability for dairy cattle. Likewise, there is need to understand the coping strategies that farmers adopt in order to develop targeted interventions to bridge feed scarcity during the dry season and maintain animal productivity. The objective of the study was to assess feeding practices, seasonal feed availability and coping strategies for dairy cattle among smallholder farmers in semi-arid region of Eastern Kenya.


Methodology

Description of the study area

 The study was carried out between longitude 37o35’ - 37o62’E and latitude 1o32’ - 1o50’S in lower Kangundo and Mwala region in Machakos district, Kenya. Kangundo lies in UM4 while Mwala fall in LM4 (Jaetzold et al 2006). Both agro-ecological zones have high dairy cattle concentration and are major producer of milk in the semi-arid region of eastern Kenya. The number of dairy cattle within Kangundo administrative boundary is estimated at 3700 while in Mwala is 4100 (MoL&FD 2007). The two regions have diverse farming systems with UM4 representing a relatively high agricultural potential for crop production because it receives more rainfall than LM4. The altitude ranges from 1100 m above sea level on the eastern side (LM4) to 1550 m above sea level on the northern region in UM4. Figure 1 shows the sites where the survey was undertaken.

The annual rainfall ranges from 700 - 900 mm, with bimodal pattern, the long rains occurring from March to May and the short rains from October - December. Between the long and short rains there is usually a long dry season that lasts sometimes over four months (June - September). Pan- evaporation exceeds total rainfall in all the months except in November where rainfall exceeds total evaporation. Mean temperature ranges from 17 - 24oC. The soils are often shallow and contain low organic matter and high sand content (Kusewa and Guiragossian 1989). The predominant soils are the acrisols and vertisols.

Figure 1. Map of the study area
Sampling method and data collection

The target population for the study was defined as consisting of smallholder farmers having at least one grade dairy cow. A list of all farmers with dairy cattle was compiled using key informants from the region.

Several methods of sample size determination exist. These are census for small populations, use of sample size of a similar study, use of published tables and use of formulas to calculate a sample size. Since the population of interest (number of farmers owning at least one grade dairy cow) was considered to be large and no similar study had been carried out in the region previously, the use of formula was adopted. The sample size in each agro-ecological zone was then calculated using the following equation:

Where;

- n is the sample size required,

- Zα/2 is the critical value, the positive z value that is at the vertical boundary for the area of α/2 in the right tail of the standard normal distribution (Zα/2=1.96 for 95% confidence),

- E is the desired level of precision (sampling error) and,

- σ is the standard deviation of an attribute in the population (Israel, 1992).

In this study, the attribute used was the mean daily amount of milk delivered by each farmer to co-operative societies or milk shops in the study zones. Consequently, the mean, standard error of the mean and standard deviation were calculated using monthly milk delivery records for three years (2005 - 2007). The values obtained for the mean, standard error of the mean and standard deviation were 6.2, 0.21, 1.05 and 4.58, 0.40, 1.29 for LM4 and UM4, respectively. Using the above formula and substituting E with the standard error of the mean, a sample size of 96 and 40 farmers was obtained for LM4 and UM4, respectively. The households sampled were selected using simple random sampling from a list of dairy farmers and interviewed using a pre-tested structured questionnaire.

Data on type of feed resources, feeding practice, feed conservation, seasonal feed availability and coping strategies during the period of feed scarcity were collected using a questionnaires. Farmers were asked to give their experience on this using recall system. The study was carried out between the months of June and August 2008. The household head or the most senior member of the household was interviewed only once. Data collection was carried out by direct questioning, informal discussion and observation.

 Data analysis

 The data were entered in a spreadsheet and analysed using the Statistical Package for Social Sciences (SPSS) version 12 for Windows (SPSS 2002).


Results and Discussion

Feed resources and feeding management
Basal diet

The dairy cattle feeds were based on natural pastures, home-grown fodder, mainly Napier grass and crop residues. Feed utilization within a year was similar in both ecological zones. The natural pastures and Napier grass were mainly fed from January to July and December with 75-90% of farmers using them (Figure 2). However, it was observed that Napier grass was sometimes not fed to animals when natural pastures and other cultivated pastures were available but instead was left to grow and fed during the dry season. Usually, at this stage of growth, the quality is poor with few proportion of green leaves compared with the dry leaves. Napier grass contains moderate crude protein (CP) content (6-12%) during the wet season, but declines to less than 5% during the dry season (Njoka-Njiru et al 2006). Usually, Napier grass was cut and delivered to animals under zero-grazing system. About 79% of farmers in UM4 practice intensive dairy production system while 64% in LM4 practice semi-intensive system (Njarui et al 2009). The natural pastures consisted of annual and perennial grasses and acacia shrubs.

Dry season feeding was a problem and this is not an exception in the region, but is also common in other parts of Kenya. Feed scarcity has also been reported in coastal lowland of Kenya (Reynolds et al 1993; Abdulrazak 1995). Maize stover was the principal crop residue and was the main livestock feed during the peak of long dry season (August-September) with over 80% of farmers using it to feed to their dairy cattle. Although stovers are important sources of roughage they are of low nutritive values and do not provide adequate nutrients required for animal production. Consequently, animals fed on maize stover would require supplements of higher nutritive value. Njarui and Mureithi (2006) found out that dry maize stover have low CP (2.5% of dry matter) and is highly fibrous with neutral detergent fibres exceeding 70% of dry matter and making them unsuitable for livestock. Also, the poor storage methods practiced by farmers pre-dispose the crop residues to rains and sunlight resulting in further deterioration of quality.

The quantity of crop residues available to livestock fluctuates between seasons. Due to erratic and poor distribution of rainfall, there is crop failure in two out of five seasons (Stewart and Faught 1984), thus the quantity of residue in some seasons is usually low. On average, farmers plant 1.44 ha of maize and harvest about 988 kg/ha of grain per season (Ndegwa and Kooijman 1999). In an average season the ratio of grain to stover is 1:1 (Thairu and Tessema 1987), and therefore from every farm household, there is potential equivalent amount of 1.98 t/ha of stovers available per year for livestock. Apart from pigeon peas, other grain legume residues (beans, cowpeas, green grams etc) are also found in various quantities but make little contribution to overall feed production because the area cultivated is small. Pigeon pea residues were used as supplements during the dry season reaching peak utilisation in September (75% of farmers used it).

Figure 2. Feed utilization for dairy cattle by type and month.

Close to 85% of farmers in UM4 had constructed feeding troughs compared with 66.3% in LM4. In production system, where animals were partly grazed with some stall feeding, without a feeding trough, usually the feed were presented to animals without chopping. This can result to trampling and inefficient utilization. Overall, across the zones, approximately 51.3% of the farmers offered feeds to their dairy cattle twice per day. Another 16.3 and 13.8% of farmers reported that they offered feeds once and three times per day, respectively. Nevertheless, the quantity given at a time was low and the quality was poor and consequently, in 65% of the farms, animals did not receive adequate feed required for production. Reynolds et al (1996) reported that only about 66% of dairy farmers maintain adequate amount of fodder in the trough thus preventing ad-libitum access to roughage for lactating cows.

 Feed supplements and minerals.

 Dairy meal was the principal commercial supplement offered to dairy cattle. Milling by-products such as bran and wheat pollard and herbaceous or browse legumes were not used for feeding. Dairy meal was available at a cost from livestock feed stores located within several rural-urban centres in the region and was normally offered to lactating cows during milking time. Between 88 and 92% of farmers offered feed supplements to their lactating cows with the highest proportion in UM4. Majority of farmers interviewed offered supplements to their cows twice a day (62%) usually in the morning and afternoon during milking time (Table 1). About 11% of farmers offered feed supplements once per day while 5.9% offered three times a day. Approximately 6.6% of farmers offered supplement to their cows occasionally. However, it was established that the cows that were offered supplements three times a day were not based on milk yield but the number of times they were milked per day.

 The proportion of farmers who did not offer supplements (10.3%) cited drastic price increase of commercial feeds as main limiting factor. During the survey period, the cost of dairy meal had increased by over 100%, from Ksh. 800 (US$ 10.67) to slightly over Ksh. 1600 (US$ 21.33) for a 70 kg bag of dairy meal due to increased cost of energy. However, this was not matched with increased price of milk thus becoming un-economical to offer concentrates. Some of the farmers who did not offer commercial supplements reported that they occasionally used sweet potatoes vines or legume residues from cowpea and pigeon pea during milking. Nonetheless, feeding of sweet potatoes vines was mainly restricted during the rainy season when they were abundant.

Table 1. Commercial feed supplementation regimes to dairy cattle

Frequency of feeding concentrate

% of farmers

UM4

LM4

Total

Never

7.5

11.5

10.3

Once a day

12.5

10.4

11.0

Twice a day

65.0

59.4

62.0

Three times a day

10

4.2

5.9

Occasionally

5.0

7.3

6.6

Others

0

7.3

5.1

Sample size

40

96

136

The quantity of supplements offered was generally low and the amount was fixed (usually about 2 kg/day) throughout the lactation period and was not adjusted to specific nutrient required based on milk production. Similar findings were also reported by Omore et al (1996) in the traditionally dairy region of Central highlands of Kenya. As a result, the farmers did not realize the full potential in milk production from their cows. Mineral supplements inform of blocks and molasses were also available in all farms where dairy supplements were provided to livestock. While mineral blocks were available to the animal most of the time, molasses was usually provided during the dry season where it was mixed with dry fodders to improve palatability. Between 76 - 91% of farmers offered molasses at any month of the year.

 Feed conservation

 Feed conservation either as hay or silage is important as it extends feed availability and quality for livestock during the dry season. Virtually all farmers (97.5%) interviewed practiced some form of feed conservation with 94.9% conserving mainly grass hay and only 5.1% indicated that they made silage for their cows but not regularly. However, only 6.9% baled hay albeit loosely tied while 80.4% did not tie the grass with strings at all. Approximately 2.0% of the farmers left grass in field as standing hay. Standing hay is low in quality and sometimes termites destroy a large proportion, especially during the dry season. The quantity conserved was little and insufficient to sustain their herd during period of feed scarcity. It was observed that the timing in cutting of grass for baling was not at the recommended stage of growth (beyond 50% flowering) and this resulted in hay of poor quality. Thairu and Tessema (1987) reported that the CP of grasses drops to 2 - 4% in the dry season after post flowering stage.

Across the two zones studied, majority of farmers cited inadequate amount of grass as the main reason for not making hay, with 46% of farmers in UM4 and 33% in LM4 (Figure 3). The second most important constraint in LM4 was lack of baling skills (cited by 25% of farmers) while in UM4 farmers did not see the need to bale hay (cited by 27% of farmers). On the other hand, the two main reasons farmers cited for not making silage across both zones were, lack of technical skills (59.1%), high investment requirement (13.4%). The conventional methods for making silage and hay generally require mechanization and are expensive for the low income smallholder farmers who own a few animals. However, there are simple and cheaper methods that are now available but have not been promoted widely. These include use of polythene tube to make silage and hand box for hay. These do not normally require mechanization and are carried out manually.

Figure 3. Reasons for not baling hay in Upper Midlands 4 (UM4) and Lower Midlands 4 (LM4) agro-ecological zones.

Storage of crop residues

 Storage of crop residues is important since it promotes efficient utilization by livestock as opposed to grazing in situ, which leads to wastage. Maize stover was the principal crop residue followed by pigeon pea residues which are obtained after threshing the grain. Bean and cowpea residues are less important as the quantity harvested is low. Majority of the farmers (94.9%) interviewed stored crop residues for their livestock. Even where storage was practiced a large proportion was left in the farm for the animals to graze in situ hence resulting in inefficient utilization. Where the crop residues are stored, during feeding, it is usually thrown in cattle boma (enclosure) and this result in trampling and wastage.

 Close to 5% of the farmers do not store crop residues primarily because they lacked storage space and did not appreciate the need to do so. The major storage structures were hay barn and granary. Other storage facilities were gunny bags and trees. Approximately 35.1% of farmers used hay barns to store crop residues while about 3% of farmers used granaries. About 48.2% stored grain legume residues in gunny bags before subsequently storing them in hay barns or granaries. Normally, granaries are used for storing food crops. Only a few farmers (1.8%) hang the crop residues, mainly maize stover on tree branches. In some farms, the hay barns were not covered on top and the stored feeds was subjected to rapid deterioration from weather and decline in quality rapidly. In most farms where the barns were covered, they were not properly constructed and did not protect all the stored feeds adequately from rain and direct sunlight. As a result, this also leads to decline in nutritive quality over time.

 Seasonal variation of feed resources

 Due to over-reliance on rains for feed production and limited amount of feed conserved in most households there was un-even availability of feeds. Out of the total farmers interviewed, 92.9% experienced seasonality in fluctuation of feed availability while 7.1% did not. Feed availability followed closely the rainfall pattern with relatively adequate feed being reported during the green seasons (June-August) after the long rains from March to May (Figure 4). However, only a few farmers reported excess feed with the highest number recorded in January (26% of farmers), falling to nil in September and October. Highest number of farmers (70 - 72%) reported peak severe shortage of feed during the long dry season mainly in September and October. This implies that during these months, animals do not receive adequate feed to meet their full production potential, thus resulting in low productivity. It is a concern that over 50% of farmers reported moderate feed shortage during the long rainy season (March- May) and part of short rains (December) thus raising doubt on sustainability of the dairy farming in semi-arid Kenya.

 Feed shortage was determined by land size, livestock numbers and management. Farmers with small hectarage of land tended to experience feed shortage in most of the months even when rainfall was adequate as most of the land was devoted to crop production. Njarui et al (2009) found that less than 10% of the land is devoted to cultivated pastures. Low availability of feed has direct effect on performance of dairy cattle with milk production declining during the dry season when feed is scarce.

Figure 4. Seasonal variation of feeds availability within a year.
Coping strategy during the dry season

 During period of feed scarcity, dairy farmers adopt a range of coping strategies to off-set the gap and improve feed access to their cattle. The ranking of these strategies is shown in Tables 2 and 3 for agro-ecological zones; UM4 and LM4, respectively. In UM4, majority of farmers (57.9%) used conserved fodder crops to feed their dairy cattle as most important mitigation strategy against feed scarcity (Table 2). About 58.1 % of smallholder dairy farmers ranked purchase of fodder from other farmers as the second most important coping strategy. The third most important strategy was feeding forages not normally used. In LM4, the first two strategies adopted by dairy farmers were similar to those of UM4 (Table 3). Renting grazing land was the third most important strategy and was ranked third by 28.9% of the farmers.

Table 2. Coping strategies adopted by dairy farmers during season of feed scarcity in Upper Midland agro-ecological zone 4

Strategies

% of farmers for the ranks

Overall rank*

Total farmers responded

1

2

3

Feed conserved fodders

57.9

23.7

18.4

1

38

Feed less to all animals

14.3

57.1

28.6

4

7

Feed less to some animals

0

11.1

88.9

7

9

Rent grazing land

20.0

20.0

60.0

6

5

Reduce herd size

0.0

0.0

0.0

8

0

Purchase fodder from other farmers

32.3

58.1

9.7

2

31

Purchase fodder from markets

0

33.3

66.7

5

6

Feed forages not normally used

22.2

11.1

66.7

3

18

*1= most important


Table 3. Coping strategies adopted by dairy farmers during season of feed scarcity in Lower Midland agro-ecological zone 4.

Strategies

% of farmers for the ranks

Overall rank*

Total farmers responded

1

2

3

Feed conserved fodders

60.0

16.9

23.1

1

65

Feed less to all animals

39.1

30.4

30.4

5

23

Feed less to some animals

39.1

26.1

34.8

6

23

Rent grazing land

15.8

55.3

28.9

3

33

Reduce herd size

14.3

42.9

42.9

8

7

Purchase fodder from other farmers

30.9

47.1

22.1

2

68

Purchase fodder from markets

3.6

32.1

64.3

4

28

Feed forages not normally used

16.7

11.1

72.2

7

18

*1= most important

Conserved feed was most preferred because majority of farmers stored feeds for their livestock and thus it was readily available. Farmers without dairy cattle or with excess fodder were important source of purchased feed resources in both agro ecological zones during the period of feed scarcity. However, as more farmers adopt dairy farming in the region, there will be limited fodder available for sale and this could jeopardize the sustainability of dairy industry in the semi-arid zone of Kenya. Nevertheless, this can be avoided by improving efficiency in conservation and diversifying in other methods of storage.

In UM4, the forages not normally used consisted of banana pseudo stem and weeds from cropland. Acacia pods were sometimes fed to lactating cows to improve milk production. Details on the quantity and the contribution of weeds, banana pseudo-stems and acacia pods to animal diet were not established in this study. In LM4, farmers with relatively larger farm sizes were able to rent grazing land at a fee during period of scarcity. The grass was either grazed or cut and delivered to the animals.


Conclusions


Recommendations


Acknowledgements

The authors are grateful to smallholder dairy farmers who participated in the study. Special thanks are due to the staff of Ministry of Livestock Development for the larger Machakos district and the local provincial administrative officers for their support. Our gratitude is extended to the Director, Kenya Agricultural Research Institute (KARI) for supporting this study. This study was funded by the European Union and KARI under the Kenya Arid and Semi-Arid Land Programme.


References

 Abdulrazak S A 1995 The effects of supplementing roughage diets with leguminous tree forages on intake, digestion and performance of crossbred cattle in coastal lowland Kenya. PhD Thesis, University of Aberdeen, UK. 217 pp.

 Hall A, Sulaiman R V and Bezkorowajny J P 2007 Reframing technical change: Livestock fodder scarcity revisited as innovation capacity scarcity – A conceptual Framework. Nairobi, Kenya: ILRI and UNU/MERIT. http://mahider.ilri.org/handle /10568/277

 Israel G D 1992 The Evidence of extension program impact. http://edis.ifas.ufl.edu/pd 005.

 Itabari J K, Bett C, Bauni S M, Githunguri C M, Kariuki C W and Ng’ang’a J K 2006 Research priorities for KARI-Katumani mandate area. KARI-Katumani Research Centre. 61 p.

 Jaetzold R, Schmidt H, Hornetz B. and Shisanya C. 2006. Farm Management Handbook of Kenya. Vol. 2. Natural Conditions and Farm Management Information, 2nd Edition, Part C, Eastern Kenya , Subpart C1, Eastern Province.

 Kusewa P and Guiragossion V 1989 Research priorities for enhancing crop productivity in marginal areas of Kenya. In: Agricultural research in Kenya, achievements, challenges and prospects. Proceedings of the 1st KARI scientific conference held at Panafric Hotel Kenya, 14 - 16 August 1989. pp. 21-36.

 MoL&FD 2007 Ministry of Livestock and Fisheries Development. Livestock Production Annual Report 2006. Machakos district.

 Ndegwa R and Kooijman M 1999 Review Workshop. District Farming Systems Team, Machakos, 29th June - 1st July 1999. NDFRC - Katumani. September 1999. 71 p.

Njarui D M G, Gatheru M, Wambua J M, Nguluu S N, Mwangi D M and Keya G A 2009 Dairy Cattle Value Chain Assessment: Characterization of Milk Production in Semi-Arid Kenya. Kenya Arid and Semi-Arid Land Programme (KASAL). (In press).

 Njarui D M G, Gatheru M, Wambua J M, Nguluu S N, Mwangi D M and Keya G A 2010 Challenges in milk processing and marketing among dairies in the semi-arid tropical Kenya. Livestock Research for Rural development 22 (2). Article #35. http://www.lrrd.org/lrrd22/2/njar22035.htm

 Njarui D M G and Mureithi J G 2006 Enhancing maize and fodder production by use of legumes in semi-arid region of eastern Kenya. In: Mureithi J G, Gachene C K K and Wamuongo J W (eds) Enhancing agricultural productivity in East Africa. Development and up-scaling of green manure legume technologies in Kenya. pp. 203-234. ISBN 9966-879-71-4.

 Njoka-Njiru E N, Njarui M G, Abdulrazak S A and Mureithi J G 2006 Effect of intercropping herbaceous legumes with Napier grass on dry matter yield and nutritive value of the feedstuffs in semi-arid region of eastern Kenya. Agricultura tropica et subtropica, 39: 255-267.

 Omore A O, Mcdermott J J and Gitau G 1996 Factors influencing production of smallholder dairy farms in central Kenya. In: Proceedings of the 5th Scientific Conference of the Kenya Agricultural Research Institute (KARI), 14 - 16th October 1996. KARI, Nairobi, Kenya. pp. 370 - 379.

 Reynolds L, Mureithi J G, Mullins G and Thorpe W 1993 Feed resources availability and milk production on smallholder dairy farms in the sub-humid coastal region of Kenya. In: Gill L, Owen E, Pollott G E and Lawrence T L (eds) Animal production in developing countries. Proceedings of a meeting held at Ashtondm UK, 2 - 4 Septemberr 1991, BSAP (British Society of animal Production) Occasional publication Paper No 16. Wye, UK pp 158-159.

 Reynolds L, Metz T and Kiptarus J 1996 Smallholder dairy production in Kenya. In: Hursey BS (ed) World Animal Review. 87 (2), 66 -72. http://www.fao.org/docrep/ W2650T/w2650t07.htm

Thairu D M and Tessema S 1987 Research on animal feed resources: medium areas of Kenya. In: Kategile J A, Said A N and Dzowela BH (eds) Animal feed resources for small scale livestock producers. Proceedings of the second PANESA workshop held at Kabete, Nairobi, Kenya, 11-15 November 1995.  Nairobi, Kenya. pp. 125 -148.

SPSS 2002 Statistical Procedures for Social Sciences. SPSS B1 survey tips. SPSS Inc. Chicago, USA.

Stewart J I and Faught W A 1984 Response farming of maize and beans at Katumani. East African agricultural and Forestry Journal, 44: 29-51.



Received 10 September 2010; Accepted 1 March 2011; Published 1 May 2011

Go to top