H.N. Le Houérou
Principal Scientist, ILCA, Addis Ababa
4. Reaction of browse plants to various management treatments
5. Genetic investigations and improvements
6. The search for less costly establishment methods
7. Research on defensive live-hedges
Knowledge on browse species and their utilization by animals in Africa is certainly by no means negligible, especially in non-tropical Africa, but the knowledge is sporadic, partial, scattered, often qualitative and distributed, if not actually buried, in a mass of literature dealing with grazing problems in general.
It was therefore logical and desirable to bring this aspect of African grazing to light — since it is often neglected, although important — and hence the idea of gathering together in a single volume all the data available so as to draw up a balance sheet of knowledge and gaps in knowledge on this subject.
The mass of material gathered together for this symposium clearly illustrates two points:
a) There exists a plentiful body of data on the browse species used by domestic and wild animals, their ecology, their geographical distribution, their relative palatability for different animals and their chemical composition and theoretical feed value. Although the knowledge is far from exhaustive, especially for East and southern Africa, it may provisionally be considered satisfactory.
b) There exist far fewer data on the biology and productivity of browse species, either individually or in populations, the dynamics and demographic characteristics of populations, their real nutritive value, their utilization by animals and their reactions to the various management methods and treatments to which they are subjected. In these fields we are dealing more with the confession of a `state of ignorance' rather than the profession of a `state of knowledge'.
Certain biological aspects are understood for a few species. These include, naturally enough, Faidherbia albida and Acacia Senegal in the inter-tropical zone and to a lesser degree species such as Acacia tortilis and Balanites aegyptiaca. In the Mediterranean zone the cultivated species are of course the best known, but they are limited to a small number: a few Atriplex species, some Cacti (primarily O. ficus-indica), a few exotic phyllodineous Acacias, the mulberry, the olive tree, the locust tree and American vines.
Studies on the biology of browse species, such as those undertaken by Poupon on Acacia senegal and Commiphora africana, those by Cissé on Combretum aculeatum, Feretia apodanthera, Cadaba farinosa and Acacia seyal, those of Grouzis and Sicot on the phenology of Acacia tortilis, A. nilotica, Balanites aegyptiaca, Combretum aculeatum, Guiera senegalensis, Boscia senegalensis and Ziziphus mauritania, those of Pellew on Acacia tortilis, A. hockii and A. xantopholea, those of Dayton on Combretum spp. are the kind of work which should be encouraged, since they are likely to be important in terms of application as regards the management of populations and the domestication of these species. In the dry tropics the study of the Capparidaceae deserves high priority owing to their high palatability and considerable feed value. These include Maerua, Bocia, Cadaba, Capparis and Crataeva etc, while Salvadora persica constitutes a salt cure all by itself.
It should be noted that the phenomenon of pre-rainy season leaf growth, which is so astonishing in the Sahel and Sudanian zones and in the Miombo and which has fascinated generations of botanists and physiologists, would appear to have been rationally explained (see the contribution by Walker), but this will have to be verified on Sahelian and Sudanian species. The new leaves would appear to emerge uniquely on account of the translocation of water and nutritive elements within the plant. Transpiration does not appear to take place owing to the presence of a thick waxy cuticle on the young leaves until the moment at which the rains arrive.
The counter-phased phenology of Faidherbia remains the subject of conjecture and speculation, defying the expertise of botanists. A careful study of the phenology of colonies, specially in areas with a bimodal rainfall regime, would undoubtedly throw some light on this dark subject. According to the qualitative observations made in Mali (Hiernaux and Cissé, pers. comm.) it appears that a genetic factor is involved: the same individuals appear to present the same phenological characteristics from one year to the next. Determination of the part played by environment and heredity in the displacement of the phenological stages is a subject of very great interest, likely to lead to highly important applications in the management of colonies and the establishment of plantations. Investigations at farm level would probably enable working hypotheses based on empirical knowledge to be constructed.
Productivity per individual, per species and per colony still presents many problems. It is true that the determination of allometric relationships has enabled progress to be made, but the method also has its limits. The regression curves cannot necessarily be extrapolated from one colony to another, nor from one region to another for a given species (see the contribution by Bille). The uncertainty grows even greater when multi species colonies are involved (which is the usual case).
As far as consumable biomass is concerned, it is necessary to fall back on simplifying theories as to accessibility as a function of the animal involved. It is thus hopeless to expect a degree of accuracy in the measurements comparable to the precision which can be achieved for grass cover.
Moreover, it is appropriate to ask, as Harrington and Wilson have done, whether such accuracy is in the end necessary or even desirable.
Estimates of the productivity of populations over wide areas have rarely been done through measurements, although this is the final objective which will enable budgets to be assessed on a regional basis using phytosociological inventories and maps. The attempts by Hiernaux (see contribution to this symposium) show the way forward in this field. Almost all other measurements have been carried out on a pin-point basis (Poupon, Bille, Cissé, Menaut, Malaisse, Pellew, Dayton, etc.) and can be extrapolated only with difficulty owing to the absence of detailed regional inventories, including mapping and population studies of browse plant associations.
The nutritive value of browse is a subject full of uncertainty, contradictions and controversies. It cannot but be so for as long as the many detailed studies on indigestibility, digestibility and animal performances which are needed have not been carried out.
Nutritive value varies considerably from one browse species to another, from one plant part to another and from one phenological stage to another for a given plant part. In vivo and in vitro digestibility of organic matter for example, can vary from over 70% to under 30% according to the species, the plant and the phenological stage. It is unlikely that under these conditions it will be possible to discover an overall applicable regression curve which takes this factor into account on the basis of data obtained by chemical analysis, even using the most recently developed methods such as the determination of 1) the levels of cell-walls (or fibre which is insoluble in neutral solvents - neutral detergent fibre); 2) the cellular content and 3) cellulose and lingo-cellulose which is insoluble in acids (acid detergent fibre).
These methods give results which have an excellent correlation with the digestibility coefficients of classical forages. But it is unlikely that the same will be true for many browse plants in which the high tannin content in particular may well subtract some of the nutrients from the digestive process, especially as regards the action of microflora (anti-nutrients or digestibility inhibitors).
Research work should therefore be directed in two complementary directions (see the contributions by Harrington and Wilson, and by Dicko-Touré):
b) Determination of voluntary intake on pasture and any variations in this during the annual cycle and periods of feed scarcity (bite counting, utilization of fistulated animals, calculations of times spent on browsing, etc.)
It would obviously be very interesting to resume trials of this kind under various ecological conditions so as to gain an objective picture of the role of browse plants in various types of pasture and environment.
Most browse species require intervention by man to effect optimal utilization by animals. Trees grow naturally out of reach of most animals (except for the giraffe, which can reach branches at a height of 5m). Some shrubs have an intricate habit which makes a fair amount of the available forage inaccessible to animals. In both cases the trees and shrubs have to be subject to some kind of cutting (see Piot, Lawton, Boudet and Toutain, Cissé). Numerous methods exist, such as cutting back, lopping, pruning, and trimming, varying to some degree or other according to the species and the management method.
It would therefore be useful to test the reaction of the important and cultivated species to the various methods at different intervals of utilization and for the period at which intervention should be practised so as to maintain optimum yield conditions. The work of Cissé in this respect is a model which should be followed and copied.
In the Mediterranean zone it is known that phyllodineous Acacias and Atriplex have to be cut back periodically. But as far as I know nobody is in a position to say what the ideal interval for cutting back as a function of environment is, nor what the maximum intensity of utilisation compatible with sustained yield should be.
As Piot has done, it may be reasonably assumed that thinning out in many cases constitutes an ideal practice, but this is not necessarily true everywhere and for all species. Polling and clumping, for example, seem particularly suitable for mulberries.
For some species the question of whether to try for leaf and branch production, fruit production or mixed leaf, fruit and wood production arises. This is the case for Faidherbia albida, Acacia tortilis, Acacia nilotica subsp. arabica and Prosopis cineraria. To my knowledge, no experiments of this sort have yet been undertaken, despite the obvious value which this would represent. Empirical tests would apparently be of little help, at least in the case of Faidherbia, since the three types of utilization are found at farm level.
Browse plants generally form genetically heterogenous populations in which variability has rarely been studied. The work of Liacos, Moulopoulos and Papanastasis on Quercus Coccifera in Greece provides a shining exception. The work of Sarson and Hamrouni on Atriplex halimus in Tunisia could also be cited, since it ended in the selection of an ecotype with high palatability. Similar work has been undertaken on A. nummularia in South Africa (De Kock). The work of Australian scientists (Hutton, Jones) on Leucaena undertaken as part of the search for productive varieties with low mimosine levels, as well as that of Felker et al on American Prosopis varieties, clearly have very great interest.
Cloning is a technique which greatly facilitates the work of selecting ligneous varieties (Franclet and Le Houérou, Van Epps and McKell). However, almost nothing is known on the heritability of characteristics isolated by cloning, when clones are subsequently sexually multiplied.
I would be inclined to agree with McKell on this point and say that seeking high palatability types is not necessarily and always desirable. There is sometimes an advantage in keeping types with moderate palatability whenever control of herd management is inadequate. There may also be some advantage in preserving genetic diversity, for example as regards early or late seasonal leaf growth, the setting of fruit, the habit (fastigiated, recumbent, prostrate, intricate etc.).
The classical way to establish plantations consists of planting out young plants which have been raised in the nursery. This method has the advantage of providing a high success rate, especially in the arid and semi-arid zones.
However, the cost of this method is high (around US$ 0.3 - 0.5 per tree or shrub), which often makes the operation prohibitive for the small holder or herdsman.
It is therefore sensible to look for less expensive methods, such as direct drilling combined with soil preparation techniques which will provide the young plants with adequate supplies of water and protect them from competition from native species. Some successful results have been achieved (Malcolm et al, Benjamin, Savory et al), but with a small number of species.
In the Mediterranean zone quickset hedges are an effective and inexpensive means for protecting plantations and combating erosion by partitioning of slopes. The main varieties used are spiny cacti, Acacia horrida, Aberia cafra, Agave americana and dead hedges made of Ziziphus lotus (the word "zeriba", meaning an enclosure of spiny wood would appear to be derived from the arab word "seder", the name for the jujube tree). In the tropics dead hedges using acacia (A. seyal, Faidherbia, A. nilotica, Dichrostachys, etc.) are used to protect cultivated fields, and for enclosures around dwellings, etc. Because of termites they rarely last for more than two seasons and their constant renewal is the cause of the onslaught on tree populations.
Research on the establishment of live hedges, especially various varieties of Acacia and Capparis, could certainly make some indirect contribution to the protection of woodlands and of cultivated land against erosion. It is true that quickest hedges have a number of drawbacks since they provide homes for undesirable species such as rodents, granivorous birds etc. But here again a balanced situation usually develops and it can be said that, at least as far as the Mediterranean zone is concerned, the advantages far outweigh the disadvantages.
Some species of Euphorbia are traditionally used as fences in tropical Africa: E. balsamifera in West Africa, E. grandis, E. candelabrum, E. tirucalli in East Africa.
Four basic themes seem to have top and equal priority, since the pursuit or abandonment of research on such and such species or such and such a group or community would appear to depend on the results obtained in the study of these topics. The four topics are interdependent, since each taken in isolation has very little significance in terms of utilization by animals. The themes are:
In point of fact it is only when a species is ecologically adaptable, productive, sought after by the animals, of good feed value and able to propagate that it is relevant to continue more complex and longer term studies on its biology, physiology and genetic properties, or still further studies on its population dynamics, response to management and the regeneration of populations.
In natural stands the relationship between the ligneous community and the grass cover deserves more detailed study. In some cases especially when the pasture is in good condition, the browse population tends to curb grass production or at least to differentiate it. In other situations where the pasture is degraded to a greater or lesser degree, the opposite may be observed: the ligneous stratum increases the productivity and prolongs the growth of the grass stratum. These studies would obviously have immediate application to management principles, strategies and methods for rangeland.
The study of population dynamics in ligneous stands would also provide good feedback on the state of equilibrium and ecological health of rangelands, especially savanna rangeland. It could provide a means of early detection of trends and for this reason it is highly relevant in monitoring operations.
The search for less expensive methods for establishing plantations remains a valid priority, provided that the four basic conditions referred to above are fulfilled. Studies on the reaction of plants to various management methods, in particular for the species utilized in agroforestry, is a prerequisite if their utilization is to be optimized.
The study of animal behaviour should enable the structure and composition of herds to be better adapted so as to encourage optimum utilization of browse. In many cases mixed herds seem likely to take better advantage of resources, taking into account any possible intervention by wild herbivores.
Scanning populations in the search for species or ecotypes with clearly defined desirable characteristics should remain the basis of genetic improvement. In the tropics, particular attention should be paid to Capparidaceae and the possibility of domesticating these species and using them on plantations.
The development of semi-intensive mixed systems using grass and ligneous species, both feasible and suitable for extension in the smallholder system of the humid and sub-humid inter-tropical zone, such as the combination Leucaena and Panicum maximum in the Guinean zone of Ivory Coast, should be encouraged.
Finally, the participation of socio-economists in research teams is needed, so as to orient these teams towards acceptable solutions suitable for extension and, wherever possible, to contribute to the acceptability or extension of new technology.
