Fodder species
Pasture establishment
Herbage productivity
Fertilization
Grazing or cutting strategies
Offer-residue rate
Conclusions
As the number of small ruminants increases in humid areas due to improved management and disease control, greater feed supplies will be required, calling for intensified fodder production systems. At present, forage is largely grazed directly or collected from natural vegetation, and more information is needed on the feeding value of various indigenous species, possibilities for increasing their production by cultivation and possible toxicity. Feeding with farm byproducts is limited in the humid tropics because the staple root crops do not produce large amounts of edible residues. However, forage crops could be planted in rotation with food crops, either partially or totally replacing the traditional bush fallow, with the dual purpose of restoring soil fertility and providing animal feed. Many kinds of shrubs and trees should be considered for this purpose, in addition to the conventional forage grasses and legumes, as these might be better suited to the environment and the overall farming system. In some instances, more extensive systems of commercial sheep and goat production may be justified, both in the derived savanna area and in the forest zone on commercial plantations, and these operations will also require improved fodder production.
The choice of suitable fodder species in the humid tropics is wide, including both native and introduced plants. In addition to the grass and legume species available, the browse species are particularly valuable for goat production, adding variety to the diet and providing substantial levels of nutrients, particularly during the dry season when the grasses are of poor quality. Studies in East Africa (Wilson, 1957) show that goats prefer browse, shifting from grass and herbs to browse during the dry season (Malechek and Leinweber, 1972; Coblentz, 1977) when browse may constitute up to 80% of total intake.
A number of authors, including Messager (1977), Crowder and Chheda (1977) and Ruthenberg (1974), have listed useful forage species for livestock production in the humid tropics, taking into account dry matter production, ease of establishment, length of growing season, quality of dry-season feed provided, and in some cases ease of eradication when fodder plants invade neighbouring food crops. Among the grasses, recommended species include Brachiaria ruziziensis, B. mutica, Cynodon plectostachyum, C. nlemfluensis, Panicum maximum and Pennisetum purpureum, as well as their hybrids with P. typhoides, Chloris gayanus and Cenchrus ciliaris. Recommended legumes include Pueraria phaseoloides, Centrosema pubescens and Stylosanthes guyanensis. Less information is available on shrubs, but Asare (1974), Oppong (1974) and Brinkman and Adu (1977) have identified Leucaena leucocephala, Cajanus cajan, Flemingia spp., Griffonia carpinifolia and Baphia nitida as promising species.
A longer list of the names and primary uses of grasses, herbs, legumes and shrubs found on experimental plots and research stations throughout West Africa is given in the Appendix. This list is far from comprehensive, however, particularly as regards the herbs and shrubs. Lagemann (1977) also lists plants found and widely used by farmers in eastern Nigeria.
The techniques for establishing pasture grasses and legumes in the tropics are well known and have been reviewed in detail by Messager (1977) and Crowder and Chheda (1977). The best time for planting or sowing is just before the rains, or alternatively during the short August dry season, and the application of a complete nitrogen, potassium, phosphorus (N-P-K) fertilizer is advised.
Brachiaria ruziziensis, Panicum maximum and Cenchrus ciliaris are normally established from seed, as are all the legume species and Leucaena, Tephrosia and Cajanus shrubs. Establishment from seed is the easiest method, especially if mechanized cultivation is possible, though very thorough seedbed preparation is essential as well as weed control in the early stages of growth. The greatest constraint on establishment from seed at present is the fact that seeds of pasture species are not widely available on the market. Stylosanthes guyanensis is the exception, however: the techniques for producing seed from this species are well known (Messager, 1976), it has been found to produce seed readily, and seed farms are functioning in some countries. A government seed production programme is being initiated in Nigeria.
Vegetative establishment requires more labour for planting but less for seedbed preparation. This method is also limited by the fact that plant materials do not withstand long periods in transport. Species such as Cynodon spp. and Brachiaria mutica are readily established from stolons, but they tend to invade adjacent crops and are difficult to eradicate. It is also difficult to assess the purity of the strain when collecting the stolons for planting. Pennisetum purpureum is normally established by shoot cuttings, while Panicum maximum and Tripsacum laxum may also be established from crown splits, though this is a more delicate operation.
As already mentioned, the major constraint on the establishment of improved pastures in the humid tropics is the lack of seed. This is a problem for research projects and an even greater constraint on efforts to introduce improved pasture species to local farmers. Research projects can rely to some extent on imported seed, but any large-scale extension programme must be based on adequate local seed production. The amount of seed required for establishment and the amount of seed produced are given in Table 7 for a number of promising pasture species.
Table 7. Sowing and seed production rates for tropical forage species
|
Species |
Sowing Bate (kg/ha) |
Seed Production Rate (kg/ha) |
|
Andropogon gayanus |
10-20 |
50-60 |
|
Brachiaria ruziziensis |
25 |
80-120 |
|
Cenchrus ciliaris |
5 |
65-80 |
|
Melinis minutiflora |
3-4 |
30-40 |
|
Panicum maximum |
20-30 |
80-100 |
|
Setaria anceps |
8-10 |
40-60 |
|
Pueraria phaseoloides |
6-10 |
60-80 |
|
Stylosanthes guyanensis |
6-10 |
60-80 |
Source: Compiled by the authors.
Data available on the dry matter yields of forage crops in the humid tropics derive exclusively from trials conducted at research stations and usually refer to one year only. Generalizations from these data should be cautious, particularly when applied to small-farm conditions.
Trials with giant star grass (Cynodon plectostachyum) in Nigeria yielded 16.7 t of dry matter per ha in the first year, while 10 cultivars of this species yielded an annual average of 10 t dry matter/ha over a two-year period (Ademosun and Chheda, 1974; Chheda and Akinola, 1971). In a two-year study in the same area, elephant grass (Pennisetum purpureum) and 26 hybrids of P. purpureum and P. typhoides yielded an annual average of 18 t dry matter/ha during the first year and 14.5 t in the second (Aken'ova et al., n.d.). These results were obtained with the annual application of 364 kg nitrogen fertilizer per ha. Further north, in Nigeria's subhumid zone, annual yields of 6 to 8 t dry matter/ha were obtained for Digitaria smutsii, 6 t for P. maximum and 8 t for a mixture of elephant grass (Pennisetum purpureum), P. maximum, star grass (Cynodon dactylon), Centrosema pubescens and Stylosanthes guyanensis (De Leeuw, personal communication; Crowder and Chheda, 1977). In Ivory Coast, Messager (1977) recorded average annual yields over three years of 13 t dry matter/ha for P. maximum, 10 t for Brachiaria mutica and 8 t for Stylosanthes guyanensis, while Andropogon gayanus yielded 4 t dry matter/ha and Cenchrus ciliaris 4.2 t. In Togo, Ruthenberg (1974) estimated annual production on a commercial farm at 16 t dry matter/ha for P. maximum, 5 t for star grass (Cynodon dactylon) and 7 and 5 t for two types of natural pasture.
In addition to annual yield levels, the production of fodder at various times of the year is an important consideration in planning an animal feeding programme. All grasses produce more abundantly during the wet season, but the variation in productivity according to season differs among plant species. Cynodon cultivars, for instance, produce a daily average of 40 kg dry matter/ha during the wet season and only 5 kg during the dry season, giving a ratio of wet season to dry season production of 8 to 1. This seasonal difference is accentuated in the case of the more productive varieties, whose wet to dry season production ratio is 9 to 1 (Chheda and Akinola, 1971).
In addition to differences between wet and dry season, there are different production patterns within the wet season. In one study, 10 varieties of Cynodon were cut twice during the growing season. In the first year of establishment, differences in overall yield were due almost entirely to differences in growth which took place between the first and second cut, but during the second year, the differences in yield were all obtained at the time of the first cut. These results suggest that after the first year of establishment the main differences in yield between improved and unimproved plant varieties occur during the early part of the growing season, when all production is at its peak and any excess will be difficult to utilize. In most cases, fertilization with nitrogen also produces excess forage at this time (Ademosun and Chheda, 1974).
The growing season for legumes is one to two months longer than for grasses, and, if kept as standing hay, legumes retain their feed value longer. Messager (1977) reported a crude protein content of 10% for a legume pasture at the end of the dry season, which was the lowest value recorded over a year. Among the grasses, Brachiaria spp. appears to remain green at the beginning of the dry season for a longer period than native species, such as A. gayanus, probably making better use of the sparse rains and residual soil moisture available at that time (Messager, 1977). As already mentioned, shrubs and trees are particularly valuable as a source of feed during the dry season. Species such as Griffonia spp. and Baphia spp. remain leafy throughout the dry season with a crude protein content of 15 to 20% (Fianu et al., 1972).
Substantial reductions in annual dry matter yield have also been observed for various fodder species between Successive years of establishment. A drop of about 50% in dry matter yield has been reported by Asare (1974) between the first and second years of establishment for a number of pasture species and Ademosun and Chheda (1974) reported a similar drop of 33% for Cynodon spp. Some examples of reductions in dry matter yields from the first to the third year of pasture establishment are given in Table 8.
Table 8. Reductions in annual pasture yields from first to third year of establishment
|
Species |
Annual Yields (tonnes dry matter/ha) | |
|
|
first year |
third year |
|
P. maximum |
19 |
9 |
|
B. mutica |
15 |
8 |
|
S. guyanensis |
10 |
5 |
Source: Messager (1977).
Fertilization, particularly with nitrogen, has a dramatic effect on tropical pastures, not so much in raising overall yield levels as in maintaining the level achieved during the first year of establishment over successive years (Messager, 1977). With fertilization, P. maximum and B. mutica yields have been boosted to annual levels of 24 t dry matter/ha with annual averages over three years as high as 32 t. Smaller applications of nitrogen (88 kg per ha) may increase production by 50 or even 100% (Crowder and Chheda, 1977). Without fertilization, drops in dry matter yield of 55% have been recorded for P. maximum and 40% for B. mutica during successive years of production, but these were reduced to 10% with the application of nitrogen. At Bouaké in Ivory Coast, Cathou found that the production of 20 t dry matter/ha of good quality forage could be maintained with the application of 300 kg nitrogen, 100 kg phosphorus and 400 kg potassium annually. Ademosun and Chheda (1974) recorded drops in S. guyanensis dry matter yields of 50% without fertilization, reduced to 36% with modest fertilizer application, while Messager (1977) found drops in Cynodon spp. yields of 33% with or without fertilization.
The effects of nitrogen application on the dry matter yields of various forage species are illustrated in Table 9. These figures show that on average an increased yield of 50 kg dry matter/ha can be expected from the application of 1 kg nitrogen. This ratio is consistent with general findings under tropical conditions. The extent to which it is desirable to use fertilizers depends on price relationships and alternative investment possibilities.
The application of fertilizer at the end of the growing season does not appear to increase the productivity of grasses substantially during the dry season. For example, Chheda and Akinola (1971) found that the application of 170 kg nitrogen per ha doubled normal pasture production during the dry season, but the amount produced, 1 dry matter/ha, was insignificant compared with the wet season yield of 13.2 t.
Table 9. The effects of nitrogen application on dry matter yields of selected forage species at various sites in West and Central Africa
|
Forage Species |
Fertilizer Application (kg nitrogen/ha) |
Yield (t dry matter/ha) |
Country |
|
Andropogon gayanus |
0 |
4 |
Ivory Coast |
|
|
150 |
7 |
Ivory Coast |
|
Brachiaria ruziziensis |
0 |
5 |
Congo |
|
|
300 |
15 |
Zaire |
|
|
600 |
49 |
Upper Volta |
|
Brachiaria brizantha |
0 |
5 |
Cameroon |
|
|
200 |
18 |
Central African Empire |
|
Brachiaria mutica |
400 |
27 |
Senegal |
|
|
700 |
33 |
Senegal |
|
Chloris gayana |
400 |
16 |
Upper Volta |
|
|
300 |
22 |
Central African Empire |
|
Panicum maximum |
0 |
7 |
Congo |
|
|
120 |
12 |
Ivory Coast |
|
|
300 |
25 |
Cameroon |
|
|
400 |
33 |
Senegal |
|
|
600 |
38 |
Upper Volta |
|
|
700 |
41 |
Senegal |
|
Hyparrhenia rufa |
0 |
4 |
Central African Empire |
|
|
300 |
14 |
Zaire |
|
Pennisetum merkere |
0 |
8 |
Zaire |
|
|
120 |
11 |
Senegal |
|
|
300 |
14 |
Ivory Coast |
|
|
400 |
33 |
Senegal |
|
|
700 |
38 |
Senegal |
|
Tripsacum laxum |
150 |
20 |
Central African Empire |
|
|
300 |
26 |
Ivory Coast |
Source: Compiled by J C Bille.
Most recommendations on intervals between grazing or cutting pasture grasses in the humid zone suggest six weeks as a compromise between the aims of maximizing forage production and quality and assuring the regrowth of the plants. Somewhat shorter intervals of four to five weeks are recommended for P. maximum, at least during the growing season, and somewhat longer intervals for Stylosanthes spp. (Asare, 1974; Messager, 1977).
Barring extremes, the height of cutting does not seem very important. The optimum cutting height for Stylosanthes spp. is 10 cm in the first year of establishment, rising gradually to 25 cm in the third year. For P. maximum, fairly low cuts at 10 cm twice during the growing season are preferable to more frequent cuts at 25 cm (Crowder and Chheda 1977; Messager, 1977; Guerin, 1977). As with Stylosanthes spp., Hyparrhenia rufa should be cut at gradually increasing heights.
The offer-residue rate refers to the proportion of the forage on offer taken by the grazing animals. It is an estimate calculated by subtracting the amount of dry matter left in the pasture after grazing from the amount available before grazing. It is not exactly the same as dry matter consumption because it includes other losses of dry matter, due to trampling and insects for instance, but it is a useful indication in situations where no precise measures of dry matter intake are possible.
The offer-residue rate varies according to pasture species, variety, season, age and management strategy. It is affected by the relative palatability of the plants and the intensity of grazing. During the early part of the growing season, a Cynodon-legume mixture gave an offer-residue rate of 60%, while elephant grass (Pennisetum purpureum) and P. maximum pastures gave 45%. From May to September, P. purpureum × P. typhoides hybrids gave an offer-residue rate of 45%, while a pure strain of P. purpureum gave 35%. Stylosanthes guyanensis tends to have a lower offer-residue rate than S. humilis because of its thicker stems (Crowder and Chheda, 1977; Brinkman and Adu, 1977). Crowder and Chheda also observed offer-residue rates of 60 and 45% during the early wet season compared with 25% for the same pastures during the dry season.
In general, heavy stocking during a short period will result in an offer-residue rate close to 60%, while continuous stocking at a lower level results in a rate of 34 to 40% independent of pasture species. Messager (1977) found that P. maximum gave an offer-residue rate of 78% when grazed at four-week intervals, while the rate dropped to 35% when the pasture was grazed at six-week intervals and to 18% with grazing at eight-week intervals. Chheda and Saleem (1973) calculated an offer-residue rate of 45% for Cynodon IB8 slashed regularly from May to January at 10 to 18 cm after grazing, while the rate for the same period dropped to 25% when the grass was slashed at 25 cm or not at all.
Overall, it would appear that average annual yields of 4 to 6 t dry matter/ha can be expected from tropical pastures over a two- to three-year period. Yields can be increased substantially by fertilizer application, particularly with nitrogen, but this may not necessarily be economically justified in many situations. Among the grasses, dry matter yield and protein content fall sharply during the dry season, while shrubs and certain legumes maintain their quality and productivity. The improved varieties developed so far only tend to increase production when it is least needed, i.e. during the early part of the growing season. Without fertilization, substantial drops in production are reported after the first year of establishment, so any evaluation of dry matter yields should cover at least two or three years.
Of the forage produced, only around 40% is actually available for grazing, and the actual amount consumed is 15 to 30% lower than this. This Implies an annual consumption rate of approximately 2 000 kg dry matter/ha which in turn implies a carrying capacity of about 300 kg liveweight per ha.
