VIII. Integration of browse in pastoral or agropastoral systems

The integration of browse plants within pastoral and agropastoral systems in Africa*

G.C. Boudet And B. Toutain

Agrostologues, Institut d' Elevage et de Médecine Vétérinaire des Pays Tropicau
                Maisons-Alfort, France

1. Tropical browse plants: truth and fiction

2. Agroforestry in Africa

3. The future of browse plants in tropical African pastoral systems

3.1 In semi-arid tropical Africa

3.2 In tropical sub humid Africa

3.3 In tropical humid Africa

Conclusions

References

1. Tropical browse plants: truth and fiction

The goats which climb the argan tree (Argania spinosa) in southern Morocco or browse acacias in Somalia by balancing on a plank show, as nothing else can, the true situation with regard to the utilization of browse plants by the goat. Similarly, the zebu cattle clustered at the foot of a calcedrat tree (African or Senegal mahogany) being pruned by a Fulani herdsman demonstrate the extent to which cattle also enjoy some browse products.

Experiments carried out in 1976 by Blancou, Calvet et al, in Dakar (Table 1), using the fistula method, showed that a zebu may consume ligneous products (leaves, fruit or creepers) in amounts ranging from 2 to 60% of its daily intake between March and June. The animal makes thorough use of the opportunity to improve on its ordinary routine diet, but the proportion of browse increases as the dry season progresses (6% in March–April, 20% in April–May, 45% in May–June).

Table 1. Proportions (in %) of different feeds in overall intake.

Source: Blancou J., et al (1977): Composition of natural grazing consumed by cattle in a tropical environment: note on a new study technique.

These results have apparently been refuted by the work of Nebout et al, carried out in 1978 in northern Upper Volta (Table 2). Goats ingest only 180 g of browse, corresponding to one third of the intake in weight and 40% of the nitrogen requirement. Zebu cattle only consume 135 g, a minute proportion of daily intake and nitrogen requirements. However, an estimate by the authors reaches the conclusion that 50% of browse resources are probably consumed during the year by the livestock present in the area.

Table 2. Estimate of dry-and wet-season rations: Estimate of dry-season ration
Goats:

Cattle:

Estimate of wet-season ration:
Goats:

Cattle:

Source: Nebout, J.P., Toutain D. (1978) Study on browse trees in the Sahel zone. (Oudalau, Upper Volta), CTFT/IEMVT.

The work of Rose-Innes and Mabey, written many years ago now (1964), provides a sensible explanation for these apparently contradictory results (Table 3 and Figure 1). During the dry season of 1961-62, Rose-Innes fed browse leaves ad libitum to batches consisting of five African shorthorns in the Accra plain of Ghana.

Table 3. Daily grass and browse intake (lb.) in trials x and y (five beasts)

Figure 1. Studies on browse plants in Ghana. (Source: Rose-Innes and Mabey).

The animals consumed 40% of their daily ration in the form of leaves (Griffonia simplicifolia), with an intake of 3.3 kg of DM/100 kg live weight. However, the animals had 100 artificially grown bushes per hectare available to them, whereas on natural grazing land there are only 7 bushes to the hectare, of which only 0.5 ha consist of species which are easily accessible to livestock. The consumption of browse is normally highly inadequate.

From these different observations it may be concluded that browse plants can satisfy the needs of cattle or goats, on condition that the animals can gain easy access to them and that a sufficient quantity is available to them directly or with the aid of man.

2. Agroforestry in Africa

Agroforestry is a long-standing tradition amongst certain groups in Africa. There is even talk of an entire culture sustained by the Palmyra tree (Borassus aethiopium), the leaves of which are used to make matting, while the fruits are eaten and the stems are chopped up for timber which does not rot and is seldom attacked by termites. The baobab (Adansonia digitata) is protected or even planted by villagers, who harvest the bark for making ropes, the leaves for use in sauces and for-making "spinach", and the fruit for the meal it provides. In the Sahel, the gum Arabic tree (Acacia senegal) is protected by the farmers of the Zinder-Gouré area in Niger, and especially in Sudan (the El Obeid area in Kordofan) where 4 or 5 years of millet, groundnuts and sesame cropping are followed by so years of fallow, using shrubs with Acacia senegal and harvesting the gum from the fifth year onwards.

In the Sudanian zone, the West African néré or African locust bean (Parkia biglobosa to the west and P. clappertoniana to the east) is left standing in fields because it provides soumbala, which is extracted from the pods and is a much appreciated flavouring in sauces. The shea butter tree (Butyrospermum paradoxum) is also left standing by the farmers in the Sudanian zone on account of the "butter" which can be extracted from the seeds.

In the Serere area, and in the homelands of the Mandingo of Casamance in Senegal, of the Bambaras of Segu in Mali, the Bissas of southern Upper Volta, Hausas of Niger and northern Nigeria, as well as of the Djebel Marra in Sudan, the fields are dotted with Acacia albida, probably of anthropic origin.

The Hausa sultans of Zinder in Niger actually set themselves up as protectors of the "Gao", decreeing draconian preservationist measures (see Griffard, 1964). Anyone who cut down a tree without permission had his head cut off, and anyone who mutilated a tree without good reason would have his arm amputated. Acacia albida cultures involving mixed farming are often contrasted with the strict arable farming cultures using the shea butter tree, in the Mossi country. Acacia albida is a curious tree which loses its leaves at the beginning of the rains, a factor which is favourable for cereal cropping; it comes into leaf and flowers at the end of the rainy season, providing browse and shade during the dry season, both in West and in southern Africa (Botswana and Zambia), where it is locally found on alluvial soils along watercourses. In East Africa it may even foliate and flower twice, according to the two rainy seasons, and its phenology may become even more complicated in areas of transitional rainfall regime. Amongst the examples cited, only the West African Acacia albida culture is clearly agropastoral in character.

3. The future of browse plants in tropical African pastoral systems

In order to avoid repetition and open up constructive debate, only a few ideas will be outlined, leaving room for discussion.

3.1 In semi-arid tropical Africa

The strictly pastoral areas of the Sahel zone are those with an annual rainfall between 100 and 400 mm. Within the pastoral economy browse plants are virtually entirely consumed by domestic herbivores (camels, goats, sheep and cattle). In their communication to this symposium, Clanet and Gillet describe a traditional pastoral system making optimum use of the browse resources provided by Commiphora africana stands in Niger and Chad. Browsing the new leaves at the end of the dry season in areas where groundwater is easily accessible avoids the need to migrate large distances southwards in search of the first rains. Throughout the Sahel browse products are particularly valued: green or dry leaves, whether on the tree or fallen to the ground, flowers, infrutescences and green or dry fruits. Some species are so much sought after that they adopt a special "browsed" habit, with stump like branches and twisted spherical shapes, or else like a long-handled ceiling brush, or even with a prostrate habit flush with the ground or creeping along it. Particularly affected in this way are Balanites aegyptiaca, Cadaba glandulose, Feretia apodanthera and Maeru crassifolia. The herdsmen often beat the trees to bring down the fruit, but sometimes they prune them, either to make sure that ,the leaves are consumed or else to place only a few fruits within reach of their livestock.

However, the regulations are strict and the forestry decree passed by the former AOF on 4 July 1935 (Article 22) prohibits lopping in the Sahel zone. This measure was taken up by most of the Sahel countries and Article 39 of the Malian forestry code, passed on 17 February 1968, states that in the Sahel zone, the limits of which are fixed in Article 52, "the cutting, mutilating or felling of trees and shrubs for the purpose of feeding animals are strictly forbidden".

It may also be noted (Article 1) that the forested areas of Mali consist of any land which has escaped bush clearance, this having occurred on perimeters where the peasants have cut down part or all of the trees and shrubs in order to grow crops, with the result that in the Sahel all pastoral land is legally subject to forestry regulations and the hand of man is prevented from intervening to make browse available to livestock. Such a regulation is clearly not without sense in an area where desertification is a risk. It is indeed justified by the ecological conditions, but on the other hand it is virtually inapplicable in these low-populated areas. However, in order to remain within the law it is necessary to take these regulations into account in pastoral improvement projects in the Sahel, in which trimming has to be prohibited.

This preservationist measure can only have a favourable effect on the maintenance of browse plants in an ecological zone in which their survival is somewhat precarious. One possible result is that agroforestry improvement projects in the Sahel can be limited and restricted to forestry schemes alone (forestry plantations for protecting water points or for stabilizing shifting sand). These various kinds of plantation can be integrated into "green belt" projects, as has been done in recent years to combat desertification in these areas.

At the same time grazing restrictions lasting a few months (6 months to 1 year) can allow the consideration of stands of browse planted for regeneration purposes, and indeed their rapid propagation is often astonishing, as also is the recovery of the crown in the species which are consumed the most.

To sum up, a Sahel strategy for managing browse plants should consist of preservationist measures (prohibition of trimming and periodic restrictions on grazing), rather than on planting campaigns with highly unreliable results.

3.2 In tropical sub humid Africa

With annual rainfall between 400 and 800 mm distributed over 3 to 5 months, the sub humid Sahelo-Sudanian area of Africa is the home of mixed farmers, living alongside sedentary or transhumant herdsmen. In this ecological zone an agropastoral system using Acacia albida has built up, with 10 to 50 trees per hectare.

Fruit production can exceed 100 kg per tree, reaching 600 kg/ha for a population of 50 adult trees. The leaves are also used by all domestic herbivores, and if the herdsman cuts only a few branches, the crown of the tree recovers rapidly and branches can be cut anew the following year. On the other hand, if the crown is totally cut off, the chances of survival for the tree are reduced and in any case it will not be available for trimming for a long time.

The products of this tree are particularly rich in nitrogen: 17% for young leaves, 14% for older ones, 19% for flowers, 11 to 15% for whole pods, 27% for seeds, but only 6% for pulped pods. However, ripe seeds are indigestible and it is probably necessary to mill the fruits to make the nitrogen contents digestible for livestock.

Is this traditional kind of agroforestry still relevant? Although Charreau and Vidal (1965) were enthusiastic about the fertilization provided for cereals by this tree, they concluded that planting and protecting the tree could only be recommended at a rate of 20 stocks per hectare under a traditional semi-sedentary system in which manual labour plays a leading part.

Given the necessary future development of agriculture, whereby manual labour will give way first to animal and then to mechanized traction, it is difficult to see how fields of crops dotted with natural obstacles will fit in. Since wholesale deforestation remains contraindicated, the present authors advocate maintaining woodland belts as windbreaks between plots of cultivated land.

At the same time, Klein has studied the eradication of Acacia albida from cultivated land and advocated treating the stumps, cut at or preferably 20 cm below ground level, using 0.15 litre per stump of a solution of 2-4-D - 2-4-5-T in water (5 litres of chemical per 75 litres of water).

According to these authors, the traditional agropastoral system will thus have to evolve into a more clearly defined system combining open fields with wooded belts. This kind of scheme resembles the agroforestry system presented by J. S. Douglas for the Limpopo valley in southern Africa, with 300 mm of rain; the algorrobo tree (Prosopis) was planted in 1957 in contoured strips along hillsides, while the flat areas were left as grassland. Using these browse trees presupposes beating and often partially trimming them in order to make the feed available to livestock. In addition, better utilization of the pods by animals would imply previously milling them.

Similarly, Lawton (1968) advised harvesting part of the leafy branches produced in coppice regrowth, starting with the young leaves produced in the middle of the dry season. The browse would be milled and distributed to animals the following year, between the end of the rains and the foliation period for browse plants during the following dry season. His work related to the copper belt in Zambia, where the woodland savanna contains a mass of edible browse plants and rainfall is under 1000 m. Species such as Afromosia angolensis, Abizzia adianthifolia, Anisophyllea boehmii, Baphia bequaertii, Brachystegia floribunda and B. speciformis, Julbernadia paniculata. Syzygium guineense were involved. Lawton advocates enriching the natural grazing by sowing or planting these species in parallel strips, some of the leaves being browsed directly and others being harvested and stored for the beginning of the next dry season.

In an agropastoral system based on browse it is necessary to achieve a satisfactory planting method for ligneous species, i. e. distribution adapted to the requirements of cultivation methods and to the prehensive capacity of the animals, as well as in sufficient quantities to meet the needs of livestock. This presupposes protecting the young plants in species which are to be consumed, and destroying undesirable species while multiplying those which are less well represented and have a low propagation capability.

In 1964, for example, Giffard suggested April sowing of Acacia albida seeds, previously soaked in boiling water, in polyethylene wrappers, followed by planting out in August. By that time they would have a stem a few centimetres long, but the root would already measure 25 cm. The success rate expected from this method would be comparable to the natural reseeding process carried out by animals when they eat the pods and excrete the seeds. However, the young plants are usually shorn at root collar level by animals during the dry season, so that the new shoots subsequently emerging are no wider than a finger, while the underlying root may reach the diameter of a stick or post. Several years of protection would thus appear necessary before a sapling emerges strongly enough and a young trunk is established.

The difficulties of multiplying ligneous species have frequently been indicated, especially where there is a browsing risk. In Cameroon, Piot had to protect live hedges for 3 years with barbed wire. Most foresters prefer managing a village copse of trees clustered in one place, using wire fencing to close it off, rather than trying to plant wooded belts interspersed with agricultural fields.

Attempts at propagating ligneous plants are often disappointing, and, at Niono, the ILCA research team (Hiernaux et al) obtained the following results:

b) no success with apex cuttings;

c) a satisfactory success rate with rootstock cuttings from: Acacia albida 63%, Balanites aegvptiaca 62%, Grewia bicolor 60%.

Propagation from scalded seeds or rootstock might thus be considered for these browse species, but there remains the problem of protecting them from browsing for the first few years when they are planted as browse hedges or anti-erosion belts on agricultural land.

Are these kinds of agroforestry schemes really feasible in a mixed farming context and at an acceptable cost, enabling cereals, milk, meat, firewood and timber all to be produced at the same time? Is it possible, as Giffard has suggested, to reafforest a fifth of cultivated land with strips 50 m wide and set perpendicular to the direction from which the harmattan blows, bordering cultivated plots 200 m across?

Since there is no spontaneous regeneration it will doubtless be necessary to plant solid stretches of forest which will be fenced off for several years. Once the trunks have formed it will be possible to turn these reafforested areas over to grazing, and at a later stage the animals themselves will provide the dissemination method. However, an arrangement whereby the area is closed for browsing for several years should always be planned so as to allow the browse plants to grow and the wooded belts to materialize.

At a later stage selective deforestation, cutting species which are not consumed and recovering firewood and charcoal, together with trimming and clump cutting for edible species, will enable woodland areas to be managed in accordance with optimum utilization by livestock.

3.3 In tropical humid Africa

Where rainfall is over 800 mm, as in the Zambian copper belt studies by Lawton, there is a forest climax and browsing leads to disequilibrium between the grass and ligneous components of the plant cover, to the benefit of ligneous plants.

Introducing livestock to a savanna in the humid tropics thus normally leads to bush encroachment, resulting in increased tsetse infestation and the risk of trypanosomiasis. The herdsmen frequently burn part of the savanna early in the season in order to convert extra grazing areas, a process which also encourages bush encroachment. Only fires at a later stage in the season, which are difficult to control, provide a means of checking the invasion of ligneous plants. Otherwise the disequilibrium increases, the bushes grow into trees and the shade they bring eliminates grass species. Of the various ligneous species present many are consumed by livestock, and in the Adamaoua area of Cameroon Piot advocated managing rangeland by selective bush clearance and trimming in order to improve dry-season feed supplies of cattle at pasture. It was in the same kind of environment that Rose-Innes carried out his experiments on browse consumption.

In his communication to this symposium, Audru suggests managing bushy pastures as a "forage forest", clearing bushes selectively and if possible introducing other edible species such as Leucaena leucocephala. A pastoral system in humid tropical Africa would thus be possible using:

b) Dry-season grazing prepared by early burning off, on which ligneous plants which are not consumed are cut and browse plants lopped when they grow too high. These pastures could be improved by adding other palatable species, such as Leucaena leucocephala for its leaves, or Morus alba and Pithecellobium samam for their fruit.

The absence of selective rhizobia could, however, impede the development of Leucaena, and these introduced species might well have to be protected against browsing for an initial period.

Moreover, many browse species are invaders, such as Daniellia oliveri, a species which throws out suckers, or Psidium guajava (guava), which forms thickets on some pastures in Madagascar and Brazil following the spreading of seeds by livestock. Sometimes the disequilibrium between grasses and ligneous plants can lead to the invasion of species which are the forerunners of rain forest (Harungana madagascariensis in Ivory Coast, Cameroon and Central Africa). The shade provided then becomes so dense that all the grassland savanna species disappear and are often replaced by further ligneous forest species (e. g. Fagara spp.). The forest climax has thus become established and complete bush clearance for food crops is then the only feasible solution.

This kind of bush clearance, felling ligneous plants and lighting fires, is practised in dense forest for rain fed rice cultivation, and it leads to a steady increase in derived savanna in the area, as the forest is gradually extinguished. In Brazil leys are sown as a catch crop amongst the rice, so that the forest extinction stage is circumvented. The browse species introduced may be Hyparrhenia rufa, Brachiaria decumbens, Brachiaria humidicola and Panicum maximum. These species use the soil well and every year the pasture can be cleared off and regenerated by burning, for 15 years on end. In tropical humid Africa Panicum maximum is the best known grass variety and Audru sensibly suggests combining it with browse hedges consisting of Leucaena, Morus and Pithecellobium. The animals would be able to use these for shelter as well as for leaves and fruit to supplement the nitrogen in their intake.

Over and above these strictly pastoral schemes in humid tropical Africa, the possibilities for combining crop and livestock husbandry in this ecological zone should not be overlooked, especially since anti-erosion strategies are imperative to maintain the fertility of cultivated land. Anti-erosion belts, possibly using embankments, have to be covered in grass, and the edible grasses which can be used are many: Panicum maximum in clumps, Brachiaria spp. as solid cover and Cynodon plectostachyus (star grass) as a runner, while Pennisetum purpureum can be cut.

These anti-erosion belts could feasibly be sown with a mixture of grass species, including some with an upright and others with a prostrate habit, forming turf. They could also be fringed with species of a more suffrutescent than ligneous nature, such as Cajanus cajan (pigeon pea). This pluriannual species grows very quickly within the period of a crop cycle and is resistant to browsing when the village herds enter the fields to consume crop residues and weeds.

Conclusions

From dry to humid tropical Africa the contribution of browse plants to animal feeds is a real one, especially in terms of nitrogen. Increasing the share of browse plants in animal intake implies a considerable effort on the part of herdsmen. They would have to:

b) maintain the necessary balance between the species present (by selective bush clearance);

c) make the browse available to livestock (by beating down the fruit in the Sahel, by trimming and by cutting back).

The improvement of forage supplies by increasing the contribution of ligneous plants implies gaining control of tsetse fly without excessive bush clearance, and it is a policy which should be especially advocated for unfertile land traditionally given over to extensive livestock production. The various domestic herbivores can be combined in these areas, so that they themselves contribute to the balance necessary between the grass and ligneous covers. The possible participation of wild animals might also be considered (giraffes and Derby elands in eastern Africa). The level of investment should, however, be related to the expected productivity, which is approximately 10 kg of live-weight per hectare per year in the Sahel, and 20 kg in the sub humid areas.

For highly productive, fertile land browse plants need not be ruled out altogether, since they can be used for improving agricultural land, as browse hedges and anti-erosion stands. They should not, however, lead us to forget that in high-yield mixed farming systems the only crops likely to satisfy the needs of high-producing animals in terms of both quality and quantity are the herbage type feed crops (grasses and legumes) either as pure crops or in combination.

References

Blancou, J., et al (1977). 'Composition du pâturage naturel consommé par les bovins en milieu tropical: note sur une technique d'étude nouvelle'. Colloque de Bouakesur l' Elevage Bovin en Zone Tropicale Humide.

Boudet, G. (1970).'Management of savannah woodland range in West Africa'. Australia, 11th Intl grassland congress.

Charreau, C., Vidal P. (1965). 'Influence de l'Acacia albida Del. sur le sol, nutrition minérale et rendements de mils Pennisetum au Sénégal'. Agro. trop. 20 (6.7): 600–26.

Douglas, J.S. (1973). 'L'agri-sylviculture: pour accroître la production alimentaire de la nature'. Unesco, Impact Science et Societe. 23 (2): 127–44.

Giffard, P.L. (1964). 'Les possibilités de reboisement en Acacia albida au Sénégal'. Bois et Forets des Tropiques. 95: 21–33.

Hiernaux, P. et al (1978). 'Rapport annuel d'activités de la section d'Ecologie'. Bamako, CIPEA-Mali, 79: 200.

Klein, R. (1961). 'Le défrichement et le débroussaillage des terrains de culture en zone Sahélo-soudanienne d'Afrique; observations faits au CRA de Bambey sur les procédés chimiques'. Agro. trop., 16 (3): 259–65.

Lawton, R.M. (1968)). 'The value of browse in the dry tropics (Valeur des jeunes pousses dans les régions tropicales séches)'. E. Afr Agric. and Forest. J. (Nairobi). 33 (3): 227–30.

Le Houérou, H.N. (1978). 'Resources and potential of the native flora for fodder and sown pasture production in the arid and semi-arid zones of North Africa'. Int. arid. land conference on plant resources. Lubbock, Texas Tech. University.

Le Houérou, H.N. (1978). 'The role of shrubs and trees in the management of natural grazing lands (with particular reference to protein production)'. Jakarta, 8th World forestry congress.

Le Houérou, H.N. (1979). 'Agro-forestry techniques for the conservation and improvement of soil fertility in arid and semi-arid zones'. Symposium on integration of research for the valorization of natural biological resources in the arid and semi-arid zone of Latin America. Rome, Inst. Italolatino-americano.

Le Houérou, H.N. (1979). 'Le rôle des arbres et arbustes dans les pâturages sahéliens'. Dakar, CRDI, Groupe de travail sur le role des arbres au Sahel.

Pelissier, P. (1953). 'Le rôle d'Acacia albida en Pays Serer. Les paysans Séréres. Essais sur la formation d'un terroir au Sénéga1'. Cahiers d'Outre-Mer, 6: 106–27.

Pelissier, P. (1966). 'Les paysans du Sénégal. Les civilisations agraires du Cayor à la Casamance. Saint-Yrieix.

Piot, J. (1969). 'Végétaux ligneux et pâturages des savanes de l'Adamaoua au Cameroun'. Rev. Elev. Méd. Vét. Pays Trop. 22 (4): 541–59.

Piot, J. (1970). 'Pâturage aérien au Cameroun. Utilisation des ligneux par les bovins'. Rev. Elev. Méd. vét. Pays Trop. 23 (4): 503–17.

Rose-Innes, R. (1965)).'The concept of the "Woody pastures" in low-altitude tropical tree savanna environments'. Proc. 9th Int. Grassland Congress, Bresil, Sao-Paulo.

Rose-Innes, R., Mabey G.L. (1964).'Studies on browse plants in Ghana. III. Browse/grass ingestion ratios. a) Determination of the free-choice Griffonia/grass ingestion ratio for West African Shorthorn cattle on the Accra plains using the "simulated shrub" technique'. Empire Journal of Experimental Agriculture. 32 (127): 180–190.

Wickens, G.W. (1969). 'A study of Acacia albida Del. (Mimosoideae)' Kew bull. 23 (2): 181–202.