Feed production and utilisation by dual-purpose goats in smallholder production systems of western Kenya

K Otieno ¹, J F M Onim ² and P. P. Semenye ²

¹ Small Ruminant Collaborative Research Support Program (SR-CRSP)/Kenya
Agricultural Research Institute (KARI)
PO Box 252, Maseno, Kenya

² Winrock International Institute for Agricultural Development/SR-CRSP
PO Box 252, Maseno, Kenya

ABSTRACT

The yields and quality of pastures, fodder grasses and food-crop-related sources of feeds in western Kenya are reviewed.

Yields of up to 5.8, 40, 1 and 14.1 t DM/ha for pastures and fodder grasses (per year) and maize thinnings and sweet potato vines (per season), respectively, are reported. Deficiencies in phosphorus and sodium in pastures in the region are reported. The potentials and limitations of the feed resources in the region with respect to the feed needs of dual-purpose goats are discussed.

RESUME

Production et utilisation des aliments du bétail par la chèvre à double fin en petit élevage dans l'ouest du Kenya

La production et la qualité des herbages des parcours, des graminées fourragères et des sous-produits des cultures vivrières rencontrés dans l'ouest du Kenya ont été analysées dans cette étude.

Des rendements pouvant aller jusqu'à 5,8; 40;1; et 14,1 t/ha ont été enregistrés respectivement pour les herbages des parcours et les graminées fourragères (par an), les produits du démariage du mais et les tiges de patate douce (par saison). Les carences des parcours de cette région en phosphore et en sodium ont été décrites et les potentialités ainsi que les limites des ressources alimentaires disponibles sur place ont été examinées.

INTRODUCTION

Over the past 10 years the Small Ruminant Collaborative Research Support Program (SR-CRSP) has been undertaking multidisciplinary farming systems research in the intensive farming regions of western Kenya, with a view to developing an entire production technology package for dual-purpose (meat and milk) goats. The research has been concentrated mainly in selected sites (clusters) within Kakamega and Siaya districts.

Kakamega district, at an altitude of 1400-1850 m, is considered to be a zone of high agricultural potential (Pratt et al, 1966); it has high annual average rainfall of 1200 2100 mm, distributed bimodally, and lush vegetation. According to Jaetzold and Schmidt (1982), however, 85% of the district is covered with infertile Acrisols except for a small fertile area of Nitosols around Kakamega town.

Siaya district, on the other hand, is considered to be a zone of medium agricultural potential (Pratt et al, 1966). Rainfall is bimodal but lower than in Kakamega district and second season rains are less reliable. Annual average rainfall is about 1140 rum, ranging from 800 mm at the shores of Lake Victoria to 2000 mm near the border with Kakamega. The altitude ranges from 1140 to 1500 m. Soils here are generally of low fertility; they range from black cotton soils along the lake shores to red friable loams in the north-east of the district. According to Jaetzold and Schmidt (1982) the dominant soils in the district are mainly soils developed on basic igneous rocks and occur on lower level uplands. They include Luvisols, Nitosols and Ferralsols, the later being limited in fertility and water-holding capacity.

One of the major concerns of the SR-CRSP has been to study the crop and forage production systems currently used by smallholders in the region and to develop and test interventions to improve feed production without adversely affecting food-crop production. This paper highlights some of the research findings. It also examines the practical implications of these results with respect to the utilisation of these feeds by the dual-purpose goats being introduced in the region by SR-CRSP.

MATERIALS AND METHODS

Study sites (clusters) were chosen in Kakamega and Siaya districts to represent two agro-ecological zones. The sample sites used were drawn from a much larger national sample frame which had been designed earlier by Kenya's Central Bureau of Statistics of the then Ministry of Finance and Economic Planning for use in the first phase of its Integrated Rural Surveys.

After the selection of the study sites and identification of households within each site, a baseline survey was undertaken to provide information on such aspects as household composition, land-use patterns, livestock numbers and types and other capital resources. Monitoring surveys then followed routinely at 28-day intervals to assess the seasonal changes in farming activities, quantity and quality of feeds, and condition of the animals (Sands, 1983). Samples of the major feeds consumed by livestock were taken at intervals over a year and their quality determined by laboratory analyses. After the surveys, several experiments were carried out, both on-station and on-farm, to develop feed production technologies which can be integrated into the existing cropping systems of the region (SR-CRSP, 1981-90a, b). Technology packages (Techpaks) have been developed and are being tested on the farms. The adoption patterns of the individual components of the Techpaks are being monitored.

RESULTS AND DISCUSSION

The major sources of feed for livestock in the smallholder farming systems of the two districts in western Kenya are mainly unimproved, uncultivated communal grazing areas, roadsides, homesteads, school compounds, planted fodder grasses and food-crop-related sources.

Natural grazing

The natural pastures in the region consist mostly of grasses of the genera Cynodon, Hyparrhenia, Digitaria, Pennisetum, Loudetia, Sporobolus and Cymbopogon.

There are scanty data on the potential productivity of these pastures. Goldson (1977), however, reported a mean annual yield for 18 pastures in six small farms in the region to be 4747 kg DM/ha. Basing their estimates on Goldson's results for botanical composition and annual rainfall, Sands et al (1982), working in the same region, gave specific DM yield figures for Kakamega and Siaya farms as 5757 and 3752 kg/ha per year, respectively. Of these annual totals, the long rains contribute 3838 and 2502 kg/ha per season for Kakamega and Siaya while the short rains contribute 1919 and 1250 kg/ha per season, respectively. Table 1 shows some estimates of on-farm grazing DM yields by farm size classes based on the seasonally available grazing areas. It can be seen from the foregoing that the greatest problem is lack of precise information on the productivity of the available grazing areas which still contribute over 50% of the feed resources of livestock in the region. This information would be useful in designing feeding programmes for the livestock in a situation where the opportunity cost of growing feeds would be too high and labour is constraining.

The species composition of the pastures varies with district, microclimate and season. Table 2 gives a summary of the quality parameters for some of the dominant species across the two districts. The crude-protein (CP) values for C. dactylon, P. purpureum and roadside mix compare favourably with values reported for grasses by other workers, especially for early stages of growth (French, 1943; Thairu and Tessema, 1987). The CP value for C. afronardus is, however, more characteristic of late stage of growth or dry season. Most of the areas in Siaya district are, unfortunately, dominated by this grass which is also deficient in calcium, phosphorus, magnesium and zinc. It is also low in digestibility and digestible-energy content as well as in dry-matter intake, and these conditions persist throughout the year (Sands et al, 1982).

Table 1. Annual production of fodder and digestible energy from grazing areas

Farm size (ha)	Short rains (1980)		Long rains (1981)
	Dry-matter yield (kg/farm)	Digestible-energy yield (MCal/farm)	Dry-matter yield (kg/farm)	Digestible-energy yield (MCal/farm)
< 0.5	295	761	516	1338
0.5-1.0	576	1489	716	1828
1.0-1.5	1257	3245	1202	3002
1.5-2.0	1336	3501	1786	4610
> 2.0	2384	6150	1989	5043

Source: Sands et al (1982)

The macro- and microelement levels of pastures are equally important and their importance in livestock production has been adequately documented by Underwood (1966) for temperate regions. The mineral levels in pastures in western Kenya have been reported by Sands et al (1982), Musalia et al (1989) and Mbwiria et al (1984). Table 3 gives a summary of the results. It is apparent from the data reported by Sands et al (1982) and Musalia et al (1989) that the sampling method used has a real effect on the results obtained in mineral analysis for samples taken from the same study area. Whereas Sands et al (1982) collected samples of individual grass species at intervals over a year and pooled these per species for analysis, Musalia et al (1989) collected representative samples of forage materials consumed by goats on a monthly basis for a period of five months covering both the wet and dry seasons. The latter workers reported generally higher levels of minerals. For example, for the macroelements, it is only phosphorus and potassium (in C. dactylon and C. afronardus) levels which were higher in the data of Sands et al (1982) than in those of Musalia et al (1989). For the microelements, it is only copper levels in the data of Sands et al (1982) that were close to those of Musalia et al (1989), the rest being very low. Again Musalia et al (1989) indicated higher mineral levels for lower altitudes than for higher altitudes. Since Musalia et al (1989) were mimicking grazing goats when taking their samples their results are probably closer to the practical situation. This is because a grazing animal is usually able to select a diet of higher quality than that of the average pasture.

It is, nevertheless, clear from both studies that Mg is adequate in the area irrespective of sampling method. The same is true of Fe which is exceptionally high and Na which is low. It is apparent from this that it is difficult to rely on experimental data for establishing mineral supplementation regimes for grazing livestock and more so for those that browse as well.

Table 2. Mean quality parameters for some of the dominant grass species in western Kenya

Quality parameters	C. dactylon		C. afronardus		P. purpureum		Roadside mix
	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Crude protein	21.25	1.00	6.56	2.17	10.06	2.96	18.73	5.52
Neutral detergent fibre (NDF)	64.56	4.03	74.36	1.35	65.92	3.72	69.39	6.86
Acid detergent fibre (ADF)	31 55	1.24	44.90	2.34	38.25	3.54	33.59	4.08
Hemicellulose	33.21	2.87	29.46	1.50	27.67	1.68	35.80	3.72
Cellulose	27.33	1.38	38.13	1.70	34.28	2.84	29.25	3.35
Lignin (L)	4.02	0.38	6.78	0.86	3.97	1.09	4.34	0.89
L/ADF	0.13	0.01	0.15	0.01	0.10	0.02	0.13	0.02
Silica	2.32	0.56	3.31	0.45	4.40	0.61	2.67	0.52
In vitro cell-wall digestibility	69.79	4.42	45.46	11.48	64.81	5.47	69.42	5.36
In vitro organic-matter digestibility	79.02	2.80	55.87	9.57	72.66	4.85	76.23	5.17
Predicted cell-wall digestibility	59 97	3.83	54.49	2.83	67.94	5.49	60.02	4.18
Predicted organic-matter digestibility	78 69	3 00	69 26	2.61	86.05	6.30	75.89	5.88
Predicted digestible energy (MCal/kg DM)	3.14	0.11	2.12	0.39	2.90	0.19	3.06	0.21
Predicted dry-matter intake (g/kg BW0.75)	61.43	5.92	43.30	3.57	59.50	5.73	52.35	11.20

Predictions based on summative equations of Goering and Van Soest (1970)

Source: Sands et al (1982)

It is known that concentrations of mineral elements in forages are generally dependent upon the interactions of edaphic, biotic and climatic factors (McDowell, 1976). Specific regional differences in soil characteristics such as differences in geological formations, drainage and pH, account for most of the naturally occurring mineral deficiencies in livestock (Hartmans, 1970; Mitchell et al, 1957; Latteur, 1962; Pfander, 1971;). Some differences are species-specific, some plants being able to accumulate higher levels of particular mineral elements than others growing on the same soil (Beeson, 1961). Stage of maturity of pastures also affects levels of mineral contents with elements such as P. K, Mg, Na, Cl, Cu. Co, Zn and Mo declining as the plant matures (Underwood, 1966). Pasture management and yield also affect plant mineral composition through, for instance, grazing pressures which will determine the predominant species of forage in an area and consequently affect mineral composition.

Table 3. Macro- and microelement levels in pastures in western Kenya

Sample analysed		g/kg DM					mg/kg DM
		Ca	P	Mg	K	Na	Fe	Zn	Cu	Mn
C. afronardus		1.56	2.2	0.9	16.3	0.012	75	19	2	12
P. purpureum		2.12	3.5	1.4	54.8	0.038	51	18	5	80
Roadside mix a		4.72	3.3	1.9	3.8	0.122	202	49	9	77
Tethering sites b
	Grass	3.60	1.6	1.8	5.1	0.20	222	28	7	118
	Browse	6.30	1.3	2.1	3.4	0.19	317	33	13	111
28 genera of grasses		3.70	2.8	3.1	28.0	0.60	480	34.9	29.4	165
Expected range		0.40	0.2	0.3	-	0.01	-	1.0	1.1	9
		to	to	to		to		to	to	to
		60	7.1	10		21.2		120	100	2400
Critical level (goats)		3.60	2.7	0.9	-	0.9	15	45	7	45

^a Roadside mix is mainly Digitaria scalarum and Cynodon dactylon
^b Data for tethering sites are means of four study sites (Maseno, Masumbi, Hamisi and Kaimosi)

Sources: Sands et al (1982); Musalia et al (1989); Kayongo-Male and Thomas (1975); Minson (1977; 1982)

Mineral requirements, on the other hand, are highly dependent on the level of productivity (ARC, 1965; NRC, 1970) and breeds (Phillips, 1956; Correa, 1957; Wiener and Field, 1969). For example, under improved management practices where levels of milk production and growth rates are highly improved, mineral nutrition becomes a crucial part of the management strategy. It is, however, difficult to pinpoint specific mineral requirements since exact needs depend on the chemical form in which the mineral is available and interrelationships which exist between them, such as occur with Ca and P. Although estimates of requirements are useful, they can only provide guidelines and cannot be a replacement for the critical diagnostic test - a positive response to correction of a suspected deficiency.

In summary, pastures in western Kenya are generally low in P and Na and these deficiencies are more elaborate in some grass species than in others. Phosphorus levels in Cymbopogon, for example, are inadequate to support milk production and acceptable fertility levels. The grass species is also low in Mg, Zn, and Na. The excessive levels of Fe can have adverse effects by increasing the likelihood of Cu deficiency developing. It has been shown that a relatively minor increase in dietary iron to almost 1 g/kg for cattle rapidly reduced liver plasma copper to concentrations indicative of deficiency (Humphries et al, 1981).

Fodder grasses

Fodder grasses are mainly important as cut-and-carry sources of feed. The most widely grown fodder grass in Kenya is Napier grass (Pennisetum purpureum) and its derivatives such as bane grass (P. purpureum x P. typhoides) which is the one mostly recommended for dairy enterprises. Productivity of bane grass varies from one area to another and depends on whether or not fertiliser or manure is applied. Yields of up to 10 t DM/ha after eight months of growth have been reported (KARI, 1985). This represents on average a daily growth rate of about 180 kg/ha.

In western Kenya, Mathuva et al (1985) reported a DM yield of 3.4 t/ha for a first cutting height of 1.3 m with a crude-protein level of 13.3% when fertiliser (NPK 20-20-60) was applied at the rate of 100 kg N/ha. The cumulative yield for a year (three cuts) was 40 t DM/ha.

The quality of Napier grass at utilisation stage is often high. Sidahmed et al (1984) reported an average in vitro apparent dry-matter digestibility of 76.2% for western Kenya. Research on the use of the Napiers in small ruminant production systems is still scarce. However, studies by van Eys et al (1986) have shown that goats fed only Napier grass of 6-8 weeks regrowth with a 12% CP content had an average daily weight loss of 1 g/day compared to a weight gain of 21 g/day for goats fed Napier supplemented with Gliricidia maculata, Leucaena leucocephala or Sesbania grandiflora. Thus, although CP requirements of ruminants for moderate levels of production are met at dietary DM concentrations of 11-12% (ARC, 1980), this study showed that productivity of goats fed Napier grass containing otherwise adequate levels of CP may still be limited by the inefficient utilisation of nitrogen. This is again demonstrated in a study by Yates and Panggabean (1988), in which goats were offered either a basal diet of Napier grass ad libitum with no supplement or Napier grass supplemented with either an energy- and protein-containing concentrate or fresh chopped Leucaena leaves and small twigs, both supplements being offered at levels of 25, 50 or 75% or ad libitum. The objective of this study was to look at the effects of increasing amounts of supplements on feed intake and performance. Those animals which were on 100% Napier grass lost weight at the rate of 19 g/day whereas there were positive responses in liveweight change with increased intakes of concentrates and Leucaena, with maximum liveweight gains of 76 and 43 g/day for ad libitum concentrate and Leucaena, respectively. It was shown that to maintain liveweights, approximately 180 and 220 g/day of concentrate and Leucaena, respectively, were required in addition to the basal diet. The efficiency of protein utilisation can thus be increased by supplementation with tree legumes.

Food-crop-related sources

Fresh feeds

The major food crops in western Kenya include maize (Zea mays), beans (Phaseolus vulgaris), sweet potatoes (Ipomaea batatas), sorghum (Sorghum vulgare) and cassava (Manihot utillisima). The distribution of these crops in the region follows the agro-ecological zones, with cassava and sorghum being mostly predominant in the lower altitude zones of Siaya district.

The interactions between the food crops and animal production systems are stronger in Kakamega than in Siaya district. This is so because, with a population density estimated at over 800 persons/km², Kakamega district has very little land left for grazing. Maize, which is the major cereal crop in the district, is also a major source of feed for livestock. It can generate feeds through thinning and leaf-stripping just before the crop is ready for harvesting of ears. Usually a farmer would plant more than one seed per hill as a security against germination losses. The extra seedlings are eventually thinned out at weeding and are a useful source of feed, particularly for smallstock. Depending on the original plant population, between 350 and 1000 kg DM/ha can be generated with a mean crude-protein content and in vitro dry-matter digestibility of 21 and 59%, respectively. Additional feed can be generated in this way just after silking stage from those plants that will have arborted.

Leaf-stripping can begin 90 days after planting with the removal of one leaf per plant per week, starting with the bottom leaves. Four leaves, including the flag leaf and the leaves subtending the cobs, should be left on each plant. This procedure can give about 800 kg DM/ha if done for up to nine weeks. Table 4 shows some estimates of quality reported for fresh and dry (hay) leaf-strippings. It has been shown that fresh maize leaves contain sufficient protein, macroelements and energy to support lamb gains at about 100 g/day with an intake of 770 g/day and food conversion efficiency of 8-10 (Kayongo-Male and Abate, 1982).

Sorghum can also provide thinnings and strippings, although early stripping should be avoided as it can have a negative effect on grain yield. The advantage with sorghum is that after harvesting the heads the crop can be ratooned and the regrowth used as feed. However, all sorghums (grain or forage types) contain at least trace amounts of cyanogenic glucoside. When an animal feeds on the plant the glucoside is broken down by a glucosidase enzyme and hydrogen cyanide is released; the enzyme is present both in the plant tissue and in the rumen liquor. The hydrogen cyanide released is rapidly absorbed into the bloodstream where it combines with haemoglobin to form cyanohaemoglobin which does not take up oxygen. The poisoned animal then dies from respiratory paralysis. To reduce the chances of poisoning, it is better to avoid grazing wilted or very young plants, and grazing very hungry animals.

Table 4. Quality estimated of fresh and dry (hay) maize leaf strippings

Quality parameters	Content (% DM)
	Fresh leaves	Dry hay
Crude protein	12.8	9.5
Acid detergent fibre	41.8	47.5
Lignin	6.5	6.3
In vitro dry-matter digestibility	55.7	50.5
Calcium	0.60	-
Phosphorus	0.25	-

Source: Adapted from Kayongo-Male and Abate (1982)

The other major food crop in the region with potential for feed is the sweet potato which is traditionally grown to provide tubers for human consumption. Several cultivars occur in the region, often planted together in mixtures, and they differ in their potential for tuber and vine production. Karachi (1982) evaluated 31 local cultivars for their tuber yields and protein contents of both tubers and vines after nine months of growth He observed significant differences in tuber protein content, mean tuber numbers per square metre, mean tuber weight and tuber yields. There were also significant differences in dry-matter content of the vines. On the basis of these observed differences he classified the cultivars into vine, tuber, and multipurpose types (Table 5). Similar studies have been done at Maseno using cultivars collected within a 15 km radius around the Maseno research station (Onim et al, 1985). Large differences in tuber and vine yields were again observed, the best tuber-yielding cultivar giving 20.9 t/ha of fresh tuber (equivalent to 6.81 t DM/ha) with a crude-protein content of 8% after three months of growth. The poorest tuber yielder gave the highest vine yield of 14.1 t DM/ha in eight months, indicating a negative relationship (r = -0.36) between vine DM yield and tuber fresh-weight yield. The mean vine yield of 8.5 t DM/ha in eight months is much higher than the range of 4.0-5.6 t DM/ha per harvest reported by Ruiz et al (1980). The performance of sweet potatoes in western Kenya is thus very good and the observed large and significant variations (P<0.01) among the forage parameters and tuber yields indicate a potential for improvement through selection. In terms of forage production a high vine:tuber ratio would be the ideal choice. However, under smallholder conditions where tubers are equally in high demand, a compromise between tuber and vine yield is a better option and this is where the multipurpose types become most appropriate.

Nutritional studies at Maseno using crosses of Toggenburg on East African goats have shown differences in dry-matter intakes of sweet potato vines according to the physiological status of the animals, with the lowest intakes being recorded in dry does (2.5-2.8% body weight) followed by kids of six months of age (3.9-4.8% body weight) and then lactating does (6.1-6.8% body weight) (Brown and Nderito, 1982). These differences would be expected, although the high value recorded for the lactating does is more characteristic of temperate dairy breeds. Values of up to 5-7% of body weight have been cited as suitable for dairy goats in a temperate environment (Mackenzie, 1967). Devendra (1967) reported a value of 1.6% as the maintenance level for Kambing Katjang goats (meat) in Malaysia. When the same goats were placed on three different planes of nutrition above maintenance, the value rose to 2.2, 2.6 and 2.7% for each level, respectively.

Table 5. Classification of sweet potatoes on the basis of tuber and vine yield

Purpose	Cultivar	Tuber:vine ratio	Vine:tuber ratio	Total dry-matter yield (t/ha)
Tuber	Muibei	0.5	1.9	17.7
	3011	0.5	2.0	17.4
	3009	0.8	2.0	21.3
	Giganda (E)	1.0	1.0	16.2
Multipurpose	Nyaliech	0.3	3.5	28.6
	Opiemo	0.4	2.5	29.6
	Nyakonde	0.6	1.7	28.9
	Mania	0.6	1.7	21.0
Vines	Namunjuna	0.2	6.4	21.0
	Calorine lea	0.2	4.6	21.9
	Musinyamu	0.3	3.7	24.0
	Toilo	0.3	3.5	20.8

Source: Karachi (1982)

Dry-matter intake (DMI) as a percentage of body weight thus rises from that of meat goats through the dual-purpose types to the highest levels being achieved in the dairy types. Indeed, for meat and fibre production, goats seldom exceed DMI above 3% of body weight. Again, although the DMI for dairy breeds are higher they vary depending on breed-environment interaction. Typical values are about 3.3% for indigenous goats in the tropics, 3.6% for exotic breeds introduced in the tropics and 5.0% for goats in temperate regions. The relatively lower intakes for tropical dairy breeds and also for exotic breeds imported into the topics are attributable to a combination of body size and high ambient temperatures which tend to depress appetite.

The very high DMI values for lactating does reported under experimental conditions in Maseno are surprising for a feed with high water content (86%) given that a high level of water in feed has been reported to be one of the main causes of low DMI in the region (Semenye et al, 1989). Indeed in all the intake studies undertaken in Maseno so far the DMI for vines have been the highest (Brown and Nderito, 1982; Said et al, 1985). The one advantage with the high levels of water in vines is that a goat on the diet does not require free water unless it is a lactating doe producing over 2.5 kg of milk per day (Brown et al, 1983). However, when offered to a lactating as a sole diet it can only support suboptimal levels of production. It is more appropriate for growing goat kids that can be weaned on it much earlier (three weeks postparturm) after achieving almost three times their birth weight. Early weaning has the advantage of releasing milk for the farm family.

The quality of the vines is high with a CP of over 20% and a digestibility of about 70%, which compares well with dry-matter digestibility values reported by other workers (Ffoulkes et al, 1978; Ruiz et al, 1980).

Crop residues

Maize stover is by far the most abundant cereal crop residue in Kenya with an estimated annual availability of up to 5 million tonnes, accounting for over 70% of the total annual production of the lignified arable farm byproducts (Said et al, 1982). Its use in livestock feeding is, however, mostly limited to small-scale farms. In western Kenya, Sands et al (1982) reported that stover yields were about 150% of the grain yield; this estimate was based on field measurements in which grain accounted for 37-49% of maize plant DM at harvest. Table 6 shows estimates of these yields according to farm classes. In more recent studies, Onim et al (1986) estimated stover yields for different maize cultivars, both local and commercial, under experimental conditions in Maseno. Under these conditions, in which the local cultivars had been selected specifically for double cobs and all trial plots received fertiliser, stover and grain yields were much higher than those obtained from farmers' fields and in this case the ratio of crop residue to grain of 2:1 proposed by Owen (1976) would not be applicable (Table 7).

Table 6. Estimated stover yield in western Kenya according to farm size classes

Farm size	Maize stover yield (kg/farm)
(ha)	Short rains (1980)	Long rains (1981)
< 0.5	99	348
0.5-1.0	168	736
1.0-1.5	149	1043
1.5-2.0	273	1648
> 2.0	279	1893
Siaya	162	584
Kakamega	192	1221

Source: Sands et al (1982)

Table 7. Long-rains grain and stover yields of different maize cultivars planted under experimental conditions at Maseno

Maize cultivar	Grain yield (kg/ha)	Stover DM yield (kg/ha)
Kaimosi Double Cobber	6 494	7 269
Hamisi Double Cobber	7 136	6 778
Nyahera Double Cobber	5 691	7 177
Masumbi Double Cobber	5 803	5 765
H614	7 395	10 233
H511	4 926	6 224
H622	5 901	6 603
Mean	5 861	7 150
Mean grain to stover ratio	1:1.2

The value of maize stovers as livestock feed is, as with other crop residues, limited by their low digestibilities and crude-protein and mineral contents. Their nutritive value can be improved either by treatment methods that increase the availability of nutrients by breaking down the lignocellulose constituents, or through supplementation methods that add deficient nutrients or correct nutrient imbalances. It has been shown that treatment of roughages with anhydrous or aqueous ammonia or with urea can increase digestibility, intake and liveweight gain in ruminants (Jackson, 1978; Sundstol et al, 1978).

However, the effect of ammoniation is influenced to a certain extent by type of roughage and quality before treatment (Sundstol et al, 1978). Generally, materials of lower digestibility respond better to treatment than those of higher initial value. In some cases ammoniation has been reported to decrease N digestibility of roughages (for example, Oji et al, 1977). Recent work in Kenya, Tanzania and Bangladesh has shown that Magadi soda (a natural alkali salt deposit) is effective in improving digestibility and intake of straws (Musimba, 1980; Kategile et al, 1981; Saadullah et al, 1981).

In studies with male castrated sheep, Butterworth and Mosi (1986) reported that estimated intakes of metabolisable energy (ME) for crop residues fed alone were not sufficient for maintenance of a 20 kg male castrated sheep. However, when levels of Trifolium tembense hay supplement were increased successively, there was an increase in average ME intake sufficient to support modest levels of production. There were also significant increases in nitrogen retention, particularly for maize stover, oat straw and wheat straw diets.

Thus legumes appear to be a viable option and one that would be more cost effective under smallholder conditions. Moreover, the conventional energy and protein feeds such as grains and oilseed cakes are neither readily available nor affordable to the small-scale farmer.

Multipurpose tree legumes

Tree legumes are currently gaining prominence in sub-Saharan Africa due to their many uses and the fact that they can easily be integrated into existing cropping systems. With the increasing demand for agricultural land in western Kenya, many of the indigenous tree legumes are being phased out as more land is cleared for cropping. Some of the introduced legume species such as Leucaena are slow to establish and vulnerable to wild ruminant browsers at the seedling stage. There are therefore just a few trees scattered within the homesteads and these play a negligible role as sources of feed. One of the indigenous species which grows readily in the area is Sesbania sesban. It is seen in many farms left standing between crops; farmers believe, and rightly so, that it improves the yield of the companion crops. This is understandable taking into consideration the fact that it is a profusely nodulating legume. However, little information is currently available on the response of Sesbania to cutting management for forage generation and work is currently being undertaken in SR-CRSP to establish this.

Major limitations to the utilisation of available feeds by dual-purpose goats

The areas available for natural grazing are rapidly diminishing due to the increasing human population and the attendant increase in demand for cropping land. Even in Siaya district, where communal grazing areas have been relatively abundant, the effect of population increase is beginning to show. Moreover, the land demarcation policy has meant that those areas that were initially used communally are now under strict individual ownership. The few areas that are still left available are dominated by poor grasses of the genus Cymbopogon. These grasses cannot maintain adequate levels of fertility and productivity. The levels of dry-matter intake recorded in the district with grazing goats is very low (1.6% of body weight) with consequent deficits in energy intakes and thus poor growth (Karimi et al, 1985). In Kakamega district the available grazing areas are seriously overgrazed with the consequence that the actual amount of herbage on offer at any particular time is lower than the estimated potential productivity. This has a severe effect on dry-matter intake.

The Napier grasses, particularly bane grass, form an important source of cut-and carry forage in the region. However, the plots under these are often very small and most of them are on embankments of erosion-control cut-off drains and terraces. Moreover, they are rarely given manure or fertiliser and being very demanding crops they cannot do well under poor soil conditions. The small plots are often over-harvested and in many cases they are established under eucalyptus trees where the shade affects productivity. It is difficult to convince farmers to clear the trees since these are important sources of income. Introduction of shade-tolerant grasses may be a solution. There is also an inefficient utilisation of the Napier grass due to wastage at feeding time. Many farmers simply throw the harvested grass on the ground and goats, being fastidious feeders, will not eat any contaminated forage. SR-CRSP scientists are making a concerted effort to encourage farmers to hang the feeds, and this is currently picking up well.

Food crops still remain a very important source of feeds in mixed farming systems in the region. However, these can only provide maintenance diets in the form of leaf strippings, thinnings, stovers and sweet potato vines. It is therefore necessary to integrate tree legumes into the cropping systems to provide a source of supplement to the crop-related feeds. There is an urgent need to develop cropping systems that would incorporate both food crops and tree legumes in a self-sustaining manner and with optimal productivity under smallholder conditions. The SR-CRSP is already experimenting with a model of such a farm.

REFERENCES

ARC (Agricultural Research Council). 1965. The nutrient requirements of farm livestock 2. Ruminants. ARC, London, UK.

ARC (Agricultural Research Council). 1980. The nutrient requirement of farm livestock 2. Ruminants. ARC, London, UK.

Beeson K C. 1961. Occurrence and significance of selenium in plants. In: Agriculture Handbook 200. US Department of Agriculture, Washington, DC, USA. pp. 34-40.

Brown D and Nderito M. 1982. Nutritional strategies for dual-purpose goats. In: Proceedings of the Small Ruminant-CRSP Kenya Workshop, 15 March 1982, Nairobi, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 84-90.

Brown D L, Chavulimu E and Fitzhugh H A. 1983. Responses of lactating goats to free water restriction and a diet based on sweet potato vines. In: Proceedings of the 2nd Small Ruminant-CRSP Kenya Workshop, 22 February 1983, Naivasha, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 29-36.

Butterworth M H and Mosi A K 1986. The voluntary intake and digestibility of combinations of cereal crop residues and legume hay for sheep. ILCA Bulletin 24: 14-17. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia

Correa R. 1957. [Deficiency of Co in cattle I. Clinical study and experimental demonstration of the disease in Brazil]. Arquivos do Instituto Biologicos 24(15):199-227.

Devendra C. 1967. Studies on the nutrition of indigenous goats of Malaya. 2. The maintenance requirement of pen-fed goats. Malaysian Agricultural Journal 46:80-97.

van Eys J E, Mathius I W. Pangsapan P and Johnson W L. 1986. Foliage of the tree legumes Gliricidia, Leucaena and Sesbania as supplement to Napier grass diets for growing goats. Journal of Agricultural Science, Cambridge 107:227-233

Ffoulkes D, Hovell F D DeB and Preston T R. 1978. Sweet potato forage as cattle feed: Voluntary intake and digestibility of mixtures of sweet potato forage and sugarcane. Tropical Animal Production 3: 140-144.

French M H. 1943. The composition and nutritive values of Tanganyika feedstuffs. East African Agricultural and Forestry Journal 8:126.

Goering H R and Van Soest P J. 1970. Forage fiber analysis. Agriculture Handbook 379. US Department of Agriculture, Washington, DC, USA. 20 pp.

Goldson J R. 1977. Pasture Research Project Technical Report 21. Ministry of Agriculture, National Agricultural Research Station, Kitale, Kenya.

Hartmans J. 1970. The detection of copper deficiency and other trace element deficiencies under field conditions. In: Mills C F (ed), Trace element metabolism in animals. Proceedings of the WAAP/IBP International Symposium, Aberdeen, Scotland. Livingstone, Edinburgh and London, UK pp. 441-445.

Humphries W R. Young B W. Phillip O M and Bremner I. 1981. The effects of iron and molybdenum on copper metabolism. Proceedings of the Nutrition Society 40:68A.

Jackson M G. 1978. Treating straw for animal feeding. FAO Animal Production and Health Paper 10. FAO (Food and Agriculture Organization of the United Nations), Rome, Italy.

Jaetzold R and Schmidt H. 1982. Farm management handbook of Kenya. Vol II. Natural conditions and farm management information, Part A, A West Kenya, Nyanza and Western Province. Kenya Ministry of Agriculture and the German Agricultural Team (OAT) of the German Agency for Technical Cooperation (GTZ), Nairobi, Kenya. 397 pp.

Karachi M K 1982. The performance of sweet potato (Ipomoea batatas L. Lamb) in western Kenya. II. Yield of 31 cultivars. East African Agricultural and Forestry Journal 47:60-67.

KARI (Kenya Agricultural Research Institute). 1985. Animal nutrition. Technical Note 1. KARI, Animal Production Research Department, Muguga, Kenya. 26 pp.

Karimi S K, Semenye P P. Sidahmed A E, Fitzhugh H A and Onim J F M. 1985. Feed composition and selection by dual purpose goats and local goats from Masumbi cluster in Siaya district, western Kenya. In: Proceedings of the 4th Small Ruminant-CRSP Kenya Workshop, Kakamega, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 45-48.

Kategile J A, Said A N and Sundstol F (eds). 1981. Utilization of low quality roughages in Africa. Proceedings of a workshop held at Arusha, Tanzania, 18-22 January 1981. AUN Agricultural Development Report 1. AUN (Agricultural University of Norway), Aas, Norway. 220 pp.

Kayongo-Male H and Abate A N. 1982. Role of systematic maize (corn) defoliation. In: Proceedings of the Small Ruminant-CRSP Kenya Workshop, 15 March 1982, Nairobi, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. p. 9.

Kayongo-Male H and Thomas J W. 1975. Mineral composition of some tropical grasses and their relationship to the organic constituents and estimates of digestibility. East African Agricultural and Forestry Journal 40:428-438.

Latteur J P. 1962. Cobalt deficiencies and sub-deficiencies in ruminants. Centre d'information du cobalt, Brussels, Belgium.

Mackenzie D. 1967. Goat husbandry. 2nd edition. Faber and Faber, London, UK

Mathuva M, Onim J F M, Fitzhugh H A and Otieno K. 1985. Yield and quality evaluation of potential forage grasses in western Kenya. In: Proceedings of the 4th Small Ruminant-CRSP Kenya Workshop, Kakamega Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 37-44.

Mbwiria S K, Dickinson J O and Bell J F. 1984. Blood selenium concentration of sheep and goats from selected areas of Kenya. In: Proceedings of the 3rd Small Ruminant-CRSP Kenya Workshop, 5-6 March 1984, Kabete, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 181-187.

McDowell L R. 1976. Mineral deficiencies and toxicities and their effect on beef production in developing countries. In: Smith A J (ed), Beef cattle production in developing countries. Centre for Tropical Veterinary Medicine, University of Edinburgh, Edinburgh, UK. pp. 216-241.

Minson D J. 1977. The chemical composition and nutritive value of tropical legumes. In: Skerman P J (ed), Tropical forage legumes. FAO (Food and Agriculture Organization of the United Nations), Rome, Italy. pp. 186-194.

Minson D J. 1982. The chemical composition and nutritive value of tropical grasses. In: Skerman P J (ed), Tropical grasses. FAO (Food and Agriculture Organization of the United Nations), Rome, Italy. pp. 167-182.

Mitchell R L, Reith J W S and Johnson I M. 1957. Trace-element uptake in relation to soil content. Journal of the Science of Food and Agriculture, Supplement 8:51-59.

Musalia L M, Semenye P P and Fitzhugh H A 1989. Mineral status of dual-purpose goats and forage in western Kenya. Small Ruminant Research 2: 1-9.

Musimba N K R. 1980. Digestibility and nutritive value of maize stover, rice straw, wheat straw and oat straw treated with sodium hydroxide and magadi soda. MSc Thesis. University of Nairobi, Nairobi, Kenya. 86 pp.

NRC (National Research Council). 1970. Nutrient requirements of domestic animals. 4. Nutrient requirements of beef cattle. Publication 1754. NRC, Washington, DC, USA.

Oji U I, Mowat D N and Winch J E. 1977. Alkali treatments of corn stover to increase nutritive value. Journal of Animal Science 44:798-802.

Onim J F M, Mathuva M, Otieno K, Hart R and Fitzhugh H A. 1985. Potential of food-feed crops in western Kenya. In: Proceedings of the 4th Small Ruminant-CRSP Kenya Workshop, Kakamega, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 18-24.

Onim J F M, Mathuva M, Otieno K and Fitzhugh H A. 1986. Comparing some maize populations to standard varieties for their feed and grain yields in western Kenya. In: Proceedings of the 5th Small Ruminant-CRSP Workshop, Kabete, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 9-20.

Owen E. 1976. Farm wastes. North Holland Publishing Company, Amsterdam, The Netherlands. 300 pp.

Pfander W H. 1971. Animal nutrition in the tropics. Problems and solutions. Journal of Animal Science 33:843-849.

Phillips R W (ed). 1956. Recent developments affecting livestock production in the Americas. FAO Agriculture Development Paper No 55. FAO (Food and Agriculture Organization of the United Nations), Rome, Italy. pp. 83-98.

Pratt D, Greenway P and Gwyne M. 1966. A classification of East African rangeland with an appendix on terminology. Journal of Applied Ecology 3:369-382.

Ruiz M E, Pezo D and Martinez L. 1980. The use of sweet potato (Ipomoea batatas L. Lamb) in animal feeding. I. Agronomic aspects. Tropical Animal Production 5:144-151.

Saadullah M, Haque M and Dolberg F. 1981. Practical methods for chemical treatment of rice straw for ruminant feeding in Bangladesh. In: Kategile J A, Said A N and Sundstol F (eds), Utilization of low quality roughages in Africa. Proceedings of a workshop held at Arusha, Tanzania, 18-22 January 1981. AUN Agricultural Development Report 1. AUN (Agricultural University of Norway), Aas, Norway. pp. 151-155.

Said A N. Sundstol F. Tubuei S K, Musimba N K R and Ndegwa F C. 1982. Use of by-products for ruminant feeding in Kenya. In: Proceedings of a workshop on by-product utilization for animal production, Nairobi, 26-30 September 1982. IDRC-206e. IDRC (International Development Research Centre), Ottawa, Canada. pp. 60-70.

Said A N. Sidahmed A E, Yilala K, Onim J F M, Amwai B and Mokaya E. 1985. Preliminary investigation on the effects of feeding tropical forages on intake, digestibility and retention of nitrogen by dual-purpose pregnant does. In: Proceedings of the 4th Small Ruminant-CRSP Kenya Workshop, Kakamega, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 270-281.

Sands M W. 1983. Role of livestock on smallholder farms in western Kenya: Prospects for a dual-purpose goat. PhD Thesis. Cornell University, Ithaca, NY, USA. 218 pp.

Sands M, Fitzhugh H A, Kekovole J and Gachuki P. 1982. Results of Small Farm Systems Survey with implication to the potential for dual purpose goats in small farms in Kenya. In: Proceedings of the Small Ruminant-CRSP Kenya Workshop, 15 March 1982, Nairobi, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 12-34.

Semenye P P. Musalia L and Fitzhugh H A. 1989. Towards devising a feeding strategy for DPGs on small farms in western Kenya. In: Proceedings of the 7th Small Ruminant-CRSP Kenya Workshop, Nairobi, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 57-72.

Sidahmed A E, Reed J. Butterworth M, Otieno K and Fitzhugh H A. 1984. Chemical composition and digestibility values for some important forage resources consumed by dual-purpose goats in western Kenya. In: Proceedings of the 3rd Small Ruminant-CRSP Kenya Workshop, 5-6 March 1984, Kabete, Kenya. SR-CRSP (Small Ruminant Collaborative Research Support Program), Nairobi and Maseno, Kenya. pp. 53-57.

SR-CRSP (Small Ruminant Collaborative Research Support Program). 1981-90a. Proceedings of the Small Ruminant-CRSP Kenya Workshops. SR-CRSP, Nairobi and Maseno, Kenya.

SR-CRSP (Small Ruminant Collaborative Research Support Program). 1981-90b. Annual Reports of the Small Ruminant-CRSP. University of California, Davis, California, USA.

Sundstol F. Coxworth E and Mowat D N. 1978. Improving the nutritive value of straw and other low quality roughages by treatment with ammonia. World Animal Review 26:13.

Thairu D M and Tessema S. 1987. Research on animal feed resources: Medium-potential areas of Kenya. In: Kategile J A, Said AN and Dzowela B H (eds), Animal feed resources for small-scale livestock producers. Proceedings of the second PANESA workshop, held in Nairobi, Kenya, 11-15 November 1985. IDRC-MR165e. IDRC (International Development Research Centre), Nairobi, Kenya. pp. 125-148.

Underwood E J. 1966. The mineral nutrition of livestock. The Central Press, Aberdeen, UK

Wiener G and Field A C. 1969. Copper concentrations in the liver and blood of sheep of different breeds in relation to swayback history. Journal of Comparative Pathology 79:7-14.

Yates N G and Panggabean T. 1988. The performance of goats offered elephant grass (Pennisetum purpureum) with varied amounts of Leucaena or concentrate. Tropical Grasslands, 22(3): 126-131.