Management and utilization methods for ligneous forages: Natural stands and artificial plantations**

Jacques Piot

Director, CTFI, Ouagadougou, Upper Volta.

1. Introduction

1.1 Their perennial nature

1.2 Their long-growth phase

1.3 Their relatively delicate character when young

1.4 Their height

2. Traditional management and utilization

3. Possible methods for the management and utilization of ligneous plants

3.1 Management of natural formations

3.2 Management of artificial stands

4. Research topics

References

1. Introduction

The considerable importance of browse in animal feeds is highlighted by the fact that it has brought together this symposium in Addis Ababa, an observation which also shows that there must be many aspects of this question which remain to be understood, studied or discovered.

The topic of the present paper is one of these aspects. Many observations have been made on this subject; there has also been criticism; and interpretation to a greater or lesser extent; and yet it has been studied very little in any depth. This is because the management of ligneous plant is obviously linked to the specific and unique characteristics of these species in relation to other classical forage plants. These characteristics are as follows:

1.1 Their perennial nature

All ligneous and most subligneous species are pluriannual, and as a result provide a constant biological resource. This situation is in contrast, at least insofar as Sahel zones with annual plants are concerned, to one in which the grass cover varies in different places according to the pattern of the rainy season.

Similarly, in the more Sudanian zones, where the grass cover primarily consists of perennial and mostly short-stemmed grasses, the ligneous strata evolve more slowly that the grass cover owing to external influences such as burning of various kinds, intensive grazing or the lack of it, whether or not resting periods are allowed, etc.

1.2 Their long-growth phase

The phenology and development of ligneous species do not depend to any marked degree on the rainy season. The obvious typical example is Faidherbia albida, in which foliation actually occurs during the dry season, but prolonged green stages, in which leafing may be observed well before the return of the rains and foliage lasting on the tree long after the end of the rainy season, are very frequent.

Moreover, the ability of ligneous plants to utilize the deeper soil layers enables them to replenish their upper parts when these are destroyed by fire or lopping etc., even at the height of the dry season. Quite apart from this phenological aspect, which is peculiar to ligneous plants, it should be emphasized than some species are also able to provide a feed source throughout the year.

Whether in the form of leaves, fruit or even flowers — although the latter are not very rich in feed value — these plants enable a supplementary feed source to be found during difficult periods, the only real problem being that the quantities available are unfortunately always fairly small.

1.3 Their relatively delicate character when young

The fact that ligneous plants are delicate when young is linked with their perennial nature, which frequently necessitates a biological cycle lasting several years, which is not conducive to regeneration by natural sexual processes. While the plants are young and small they are subject to constant browsing and are far more sensitive to fire than grasses, which can actually benefit from the effects of fire on their seeds from the point of view of natural propagation. The seeds of ligneous species, on the other hand, probably do not get a chance to grow without harassment until their first rainy season, during which the presence of grass may — but even this is not certain — prevent them from being browsed by herbivores.

Even under optimum conditions the young plants will reach 20 or 30 cm high and still have a poorly lignified and delicate stem, although the taproot part of the root system will already be deeply entrenched in the soil. For example, on a sand dune a root of Faidherbia albida less than 1 year old was uncovered at 5 m depth (Maradi 1957/58, Niger). Under these conditions the upper parts of the plant may disappear without the root system suffering to any great extent. Moreover, it will sprout new shoots in a few weeks.

Hence it can be seen that, if fire is the only factor responsible for the destruction of the stem, the damage is done only once a year and as the years go by the plant strengthens its root system, which one day becomes strong enough to support the growth of a taller, stronger stem with better resistance to fire, or which might escape burning during a year's respite, when for one reason or another the fire does not pass that way.

On the other hand, if browsing occurs, with or without burning, and eliminates the upper parts of the plant, through constant repetition the plant will become exhausted and the seeds will disappear. Trials with Faidherbia albida var. Zinder between 1956 and 1961 showed indisputably that on 1 ha plots subjected only to fire the plant survived, whereas on those grazed by herbivores, particularly goats, the entire plant died.

1.4 Their height

The aspect of plant height is of course related to the height of the animals and, with the exception of giraffes and elephants, with which we are less concerned, without human intervention only part of the food available can be reached by the animals. This situation provides a sure guarantee as regards the preservation of ligneous species, but generally speaking a larger proportion of the available browse could be used without risk to the trees. Unfortunately it is true that indiscriminate intervention here on the part of man, which should be a beneficial thing, often does a great deal of harm owing to excess.

2. Traditional management and utilization

This section will allow us to make some observations intended to qualify what we "know" about the way in which ligneous plants are used. The classical method is still that of the Sahel herdsman who used his machete or hatchet to lop trees and shrubs at head height, preferably leaving the branches attached to the trunk, giving an umbrella-like habit wholly conducive to `roasting' and the death of the tree if fire occurs, which is still fortunately far from being the general case.

In fact, however, although this scenario is a very frequent one, it does vary considerably according to the pastoral area concerned. For example, it may well be met within the Saharo-Sahelian zone where nomadism is practiced and rainfall is 100 to 200 mm, but in the Sahel proper, where crops are grown, it has already become somewhat blurred. The necessary protection of crops against marauding livestock is afforded by 'zeribas', hedges made out of thorny species, so that a large proportion of lopping is carried out for a dual purpose. In these cases umbrella-shaped trees with the branches still attached are less frequently found.

Still further to the south, but still in the northern hemisphere, i.e. in the Sudanian or Sudano-Guinean zone, manual distribution of browse feed may be very rare or even non-existent, as in the Adamaoua massif in Cameroon. The fact that there are fewer herdsmen in these areas is doubtless one of the reasons for this situation, in which animals have access only to what they can reach.

All the same, declining herd supervision appears to be a general phenomenon affecting not only pseudo-sedentary farmers and transhumants, but also to an increasing extent the nomads proper. It is a situation which has some impact on the management of the herd and on its environment in general, as well as on that of ligneous plants in particular, for which there is less and less time available. In this situation either the entire tree is cut down, which remains the exception, or else the animals have access only to what they can reach. There are two general patterns, according to the region concerned:

In the Sudanian zone, where leaf decay is limited, or at least very spread out in time, supplies directly available to the animals at the height of the dry season consist of green matter, but are less than half of the total amount on the tree, as shown by Table 1, which refers to the N'Gaoundere region in Cameroon.

Table 1. Estimated forage production of trees during dry season

In the Sahel in the broad sense, all or virtually all the foliage on the tree will be consumed, but here again most of the leaves will not be consumed until they have fallen naturally to the ground, where the animals find them. It is probable that by this there will have been a considerable drop in feed value, but this is not absolutely clear, as is shown by Table 2.

Table 2. Some feed characteristics of fresh and dried leaves

However, it should be noted, at least as regards the Gardenia in this example what has happened is not really part of the plant cycle so much as a kind of forced haymaking, a phenomenon which is by no means uncommon for ligneous plants wherever burning off occurs as a matter of course every year. Nevertheless, even without trimming or pruning most of the plant material available in the Sahel area is used by animals all year round: there is evidence to show that browsing on shrubs and trees occurs even at the height of the rainy season. This is obviously true for goats, which are browsers, but also for cattle, sheep and other herbivores. Browse consumption thus does not occur only during the scarcity period, but rather distribution by herdsmen has been observed even during the height of the wet season, as was the case for Acacia seyal in Oudalan, Upper Volta.

However, this is only true for animals accustomed to consuming ligneous plants, and as early as 1970 I had written as follows about the Sudano-Guinea highland regions (Adamaoua): "In some animals an aversion to the consumption of foliage may be observed; they resign themselves to it only when there is really nothing else to eat, and therefore sometimes too late. On account of their nutritive value it is important that the leaves of successive new shoots should be consumed while young: it is therefore a good idea to force animals which are not accustomed to doing so to consume foliage before they have become too thin, and hence as early as possible in the dry season. This can be done by distributing young branches in the cowshed or by temporarily overstocking an enclosure with plenty of bushes". This in itself already provides us with one of the components in the management of ligneous plants, the utilization of which by animals, although obviously varying, should be as permanent as possible for another quite different reason.

Anyone who has fattened domestic animals or even merely supplemented their feed supply will know that the response to a new feeding programme is not immediate, but delayed by about 2 weeks. I myself found, while supplementing with Daniellia olivieri on natural savanna in the Adamaoua, that it was ineffective if supplies were provided over too short-lived a period. The need for a certain period in which the equilibrium of the rumen flora must adapt makes it highly desirable to maintain permanent contact between ligneous plants and animals, as we have seen occurs quite naturally in the Sahel.

This last observation is made in anticipation of cases where the ligneous component is freely distributed to livestock from a certain moment in the year onwards, since in point of fact — and I repeat this — the animals by themselves would naturally consume very little in the way of ligneous plants throughout the year. When we talk of livestock production we generally think of cattle, whereas in fact small ruminants, and especially goats, are very important as regards the consumption of ligneous browse. Thus in Oudalan, in Upper Volta a typical part of the Sahel, there are 13,000 goats as against 8,000 cattle, to which should be added 3,000 sheep and 900 donkeys, horses and camels, occupying a study area of some 40,000 ha, representing, I might add in passing, a considerable stocking density considering the area. But the point which should be emphasized is that goats alone consume 181 of ligneous DM between July and October, as against 86 t for cattle, while the figures for the months between November and January are 214 t and 98 t respectively.

Moreover, on an individual basis goats consume 183 g of foliage per day by direct browsing during the dry season, as against only 136 g for cattle, which only goes to show that cattle do not necessarily have a higher intake capacity than small stock.

As regards thorny species in particular, the large muzzle of cattle prevents them from consuming foliage as rapidly as a goat, which has a more elongated snout and lips which are more mobile. Hence the problem for cattle is to know whether they would in fact have time to consume their full ration in the form of ligneous forage, given 4 to 8 hours of grazing daily (Nebout and Granier).

Personally I do not believe that cattle are able by themselves to take in any very substantial part of their feed ration by browsing directly on trees. On the other hand, they are better than goats at consuming many of the deciduous parts of the ligneous plants which fall to the ground and are sometimes blown by the wind into hollows, tufts of grass or bushes. These deciduous parts represent nearly four times the quantity consumed during the first 3 months of the dry season (see above — November, December and January). These parts consist of anything that was previously out of reach of the livestock, but also of whatever was not palatable at the green stage but may be eaten with pleasure when dry (Combretum micranthum, Calotropis procera, although goats consume the latter at the green stage, etc.). This observation, which has also been verified for grass species (e.g. Loudetia spp. which are left at the green stage but are found to be palatable as dry feed) should also be included as a component of the management of ligneous plants. This traditional type of utilization, which as we have seen would be better termed natural, involves little intervention on the part of man, at least as regards a purely pastoral perspective.

On the other hand, the herdsman who uses his stick to shake down pods or other fruit from the tree cannot be accused of not acting entirely on behalf of his livestock, and it is my belief that this kind of activity is perhaps rather difficult to improve, given the present extensive nature of the rangeland, for the following reason:

Fruit is classically regarded as having a high nutritive value, especially the pods of legumes. However, this quality, which emerges as a result of chemical analysis, has for the most part been noted in seeds which have been broken down by mechanical grinding, whereas this is very far from being the case when they are munched by cattle, when the majority of seeds ingested in good condition pass straight through the digestive system without breaking down.

This fact has become well known since seeds recovered from faeces have been particularly subject to research owing to the fact that they are more likely to germinate than normal seeds, in which a marked period of dormancy has usually been induced.

The seeds digested are thus primarily those which are malformed, a few seeds which the molars have been able to grind down during rumination, but more especially those which have been eaten away by parasites, which are very frequent in some years.

Thus Poupon, at Fété Olé in the Ferlo area of Senegal, noted the development of parasites in Acacia senegal during the fruit bearing period of 1971–72, as represented in Table 3 below:

Table 3. Seed infestation

Thus nature preserves a certain quantity of seeds, and ensures that dissemination takes place, although in some years the level of infested seeds may reach almost 100%.

Nevertheless, in restricted grazing areas the recovery of the ligneous stratum is primarily based on seeds. Granier, in Niger, gives figures of 84% for seedlings, as against 16% for sprouts or suckers, and I have found similar values in the Oudalan area of Upper Volta, around the edges of land subject to flooding during the rainy season.

One should therefore be careful before advocating the grinding of seeds distributed to animals, as is for example done in the Maghreb with carob beans (Ceratonia siliqua). As regards fruiting, we should point out that the `rejuvenation' of ligneous plants by cutting or lopping severely disrupts the process of bearing fruit, although if often gives rise to a staggered production of green matter during the dry season, and sometimes throughout the entire dry season in favoured localities such as depressions and the edges of ponds, etc.

I have observed Guiera plants whose new shoots were devoured by cattle, sheep and goats right the way through the dry season. Nevertheless in cases such as these the problem of the productivity of plant material still arises. It is certainly better to preserve the production potential of 50 to 150 kg of pods per adult Faidherbia albida, than to lose the potential in return for foliage obtained by lopping which will severely threaten the future production potential of the tree. The same situation applies for Balanites aegyptiaca, which can produce 45 kg of fruit per adult tree (Granier, 10 measurements).

Of course it should be pointed out that fruit production varies considerably between one year and the next. It is nevertheless true that this element of the feed supply provided by ligneous plants is fundamental, especially since it is always available during the dry season, which is not the case for foliage. Finally, to conclude this section on traditional utilization, we should return to the question of the effect of burning on the production of ligneous species (cf Gardenia above).

In the Sudanian zone in the broad sense burning, whether it occurs late or early, will almost always result, within a maximum of 2 to 3 weeks, in the growth of new foliage. This re-growth will be virtually normal for the pyrophytic species of the savanna. If the entire stock of dried grass in the neighbourhood has not been destroyed, the feed balance of livestock will benefit substantially especially since leaves which have wilted under the influence of the flames will fall to the ground and become directly accessible to animals. This type of management method has been tried out experimentally and is now being applied in practice in the Adamaoua, for example. It is also a frequent method (whether natural or not) in cynegetic areas.

On the other hand in the Sahel zone, still in the broadest sense, only fires occurring early in the season will have the same effect, and the cost is the destruction of grass which is a basic necessity for feeding during the dry season. Luckily, fires such as these are becoming increasingly rare at the beginning of the dry season, and are also less and less likely to occur as the season progresses and grass becomes scarcer and scarcer, thus soon becoming inadequate to allow fire to spread over large areas. After this first part of my paper, which has primarily taken the form of a few thoughts, the second part will enable me to put forward a few management proposals.

3. Possible methods for the management and utilization of ligneous plants

Given the very varied climatic conditions of pastoral areas concerned with the utilization of ligneous plants, no one can claim to develop a blanket model. Moreover, the phenological behaviour of species varies increasingly as the environmental conditions become harsher. Thus, whether the year is a good one or a bad one, the species of the southern Sudanian zone will bud, flower and bear fruit at more or less the expected time. For some species, however, the quantity of fruit will vary considerably. These species classically present considerable variations from one year to the next and include Syygium guineense, Annona sengalensies, Ximenia americana, etc.

By contrast, in the Sahelian zone some phases of the cycle may disappear altogether under certain climatic conditions (Baobab trees in Oudalan have been seen without flowers or fruit, Commiphora africana also seen without flowers and fruit and Grewia bicolor without fruit, both the latter species at Fété Olé). Pests, and especially locusts, can also considerably reduce output in some cases. However, for ligneous plants it is fortunate that there is a certain delay between difficult rainfall conditions and adverse effects on the green phase, at least. This is because the re-growth of vegetation largely depends on the reserves accumulated during the preceding growth season. A problematic rainy season will obviously shorten the green phase, but its real impact will fall rather on the phenology of the following year. Poupon (1979) indicates clearly that `after an exceptionally dry year, ligneous plants take some time to recover their productive capacity'.

Despite these observations, the ligneous plants of the arid zone may be considered as less subject to variation than the grasses, and hence easier to manage.

3.1 Management of natural formations

The first problem to arise is that of the perennial nature of the ligneous part of formations used by pastoralists, and the question of its relation to any possible level of bush encroachment on the one hand, or the desirable level on the other. From a more classical point of view and especially during the recent dry years, thought has been turned towards the reconstitution of the ligneous cover, a problem which really arises in terms of regeneration, a subject we shall tackle at a later stage.

However, it is also necessary to devote some thought to rangelands, not all that rare, where the opposite situation exists and it is necessary to combat invasion by ligneous species, especially some species which are uninteresting from the pastoral viewpoint (such as Harungana madagascariensis, for example, or Lantana spp).

As regards this issue, I have demonstrated in Cameroon that burning off practices cannot be relied on to maintain the balance if the stocking level is fairly high.

Similarly, chemical arboricides are very difficult to use owing to the considerable variations in phenological behaviour of the many species which have to be eliminated, a factor which makes it impossible to apply treatment at the optimum moment for more than two or three species at the same time. Thus elimination by manual or mechanical means is the only remaining method.

First of all let us take manual methods, which can be selective and consist of the herdsman's hatchet, which day by day has a possibly vast influence on the environment. The increasing rareness of herdsmen, already indicated above, does not give much grounds for relying on this method, although it is only one which really allows the preferential treatment of better species.

Secondly there are mechanical tools of very varied kinds, consisting of chain saws, Landais rollers etc, of which the classical type is the rotary cutter, but these methods already represent a certain degree of intensification in pastoral methods. The most relevant type of case at the moment is, as we have already said, where there is a requirement for the protection, encouragement and re-establishment of trees, as is the case throughout the Sahel.

3.1.1 Protection

Under the heading of protection come all the measures helping to protect trees and shrubs from destruction, and also from a drop in productivity caused by excessive utilization of the ligneous plants available. The range thus includes the classical weapon of protective legislation on the one hand, and the rationalization of the methods by which feed resources are taken on the other.

Legislation ranges from draconian regulations such as those prohibiting the mere carrying of a hatchet in some areas, to laws of little practical value, such as those listing protected species, or which cannot be enforced, such as the blanket prohibition of burning. Giving an account of these measures and the various commentaries made on them would involve us in lengthy or even endless diversions (on the question of burning for example) and for an example I shall take only the question of species protection.

Every country has its list of species which are protected, often without being clear as to what degree of protection is required, so that the freedom to interpret often allows a repressive official the chance to penalize offences according to his mood of the moment, even well intentioned offences such as those connected with medicinal purposes. What is more, it is common knowledge that palatable species can in fact be used, and how!

It is thus imperative to specify details such as these and to turn government officials into sources of information and extension of the acceptable methods enabling even protected species to be used. These protected species are used in any case by herdsmen, at least during periods of drought, since it is inconceivable to allow a herd to starve to death next to a crown in full leaf or fruit.

The question of technique is also important because there can be no doubt that excessive offtake from trees is a highly problematic aspect and a decisive factor in determining official attitudes towards protection. Here we should make a distinction between three contrasting types of methods:

The first type consists of felling and cutting back, methods which frequently lead to the death of the tree except when it is young, in which case the ability to put out new shoots and suckers remains high and the tree is replaced by a mass of short stems which are directly accessible to livestock. Where stocking rates are high these new shoots are inevitably condemned to extinction.

On the other hand, where stocking rates are controlled this method, which results in the relegation of all ligneous plants to a low stratum, provides an easy solution for the utilization of ligneous plants. In the Adamaoua our most utilized plots from the point of view of ligneous plants were all used in this way (Daniella olivieri, Piliostigma thonningii, etc.) with periods of alternate stocking and rest for the enclosures concerned. Under these conditions the drop in productivity is not very great, and plants which have grown too high can be felled once every 2 or 3 years by manual or mechanical means. In the Sahel this method should be used in moderation, since the tree is needed as such for protecting the environment against erosion, and for maintaining a certain type of grass beneath the tree cover, etc.

Lastly, the tree is often felled fairly high (1–1.5 m) and in this case it is to be hoped that some species will form clump growths which are once again directly utilizable by livestock. Our own trials in the Adamaoua showed that for adult Daniella trees there was some difficulty in obtaining re-growth, and in particular there was a considerable drop in the yield for each individual (from 1–20%). I would suggest that this technique is suitable for artificial stands, but with densities per hectare which will allow good productivity.

In any case, many trees in the Sahel (Balanites, Guiera, Bauhinia rufescens, etc.) are maintained in this form, consisting of low-level clumps, by livestock themselves, with or without initial intervention on the part of the herdsman. In areas with a high density of ligneous plants (depression, edges of swampland, etc.) this technique should be considered as a favourable one.

The second type of method includes lopping (partial or total) and pollarding: these are the traditional practices frequently held up for justifiable criticism, since they often lead to the death of the tree. The question has therefore been whether, by moderating the intensity of lopping, it would be possible to achieve a sound management method. In the Oudalan area, a trial plot was set up for Acacia seyal, in order to compare the following treatments on the basis of the periodic growth in the circumference of the stem:

a) a control batch which was not lopped;

b) lopping of one third of the crown, with complete separation of branches;

c) the same treatment, but leaving the branches attached at their base;

d) lopping of half of the crown, with separation of branches;

e) pollarding, with separation, of branches.

As the treatment was carried out every year, polloding was effected in 1, 2, or 3, years according to the treatment given, and the experiment lasted somewhat over 3 years altogether.

During the second year one batch of stems were cut at a height of 1 m and a further batch at ground level. The plantation consisted of adult trees which were all the same age.

It became apparent that the criterion of development of the circumference could be used for this type of study, but in the final analysis it is also necessary to take into account partial or complete losses through decay and, in quantitative terms, the number and vitality of new shoots and suckers.

The conclusions of the study were as follows: "The lopping of Acacia seyal in the Sahel zone becomes damaging if carried out too intensively or repeated in consecutive years on the same tree. The production of new shoots after lopping is plentiful during the first year, but afterwards rapidly diminishes. Lopping one third of the crown produces a higher growth in the diameter than that of the control batch during the first year, but the difference reduces and falls to a lower level than the control group in subsequent years. As regards growth in diameter, the least harmful lopping method during the 3 years of observations was that involving one third of the crown, with complete separation of the branches.

However, observations on Acacia seyal which had been subject to `umbrella-type pollarding' and were not included in the experiment, showed a high level of re-growth for shoots in the centre of the lopped part of the tree. Pollarding produces high mortality very quickly (40% of trees remain alive a year and a half after pollarding). Cutting Acacia seyal at ground level or at 1 m from the ground should be forbidden in the Sahel, since the stumps die off a few months afterwards.

"Hence the conclusion is that the pollarding of trees as generally practiced by herdsmen, although favourable to livestock feeding, gradually destroys the tree flora in the Sahel zone, and particularly affects adult Acacia seyal trees.

Moreover, since the study was extremely limited it did not really enable a different type of lopping to be indicated, i.e. one which would be able to guarantee the perenniality of individual trees. The crowns take a very long time to recover when this drastic form of intervention, namely cutting off the branches, is carried out, and annual activities of this kind, even if only partial, cannot be recommended. On the other hand, the pruning of smaller branches affords quite different possibilities".

The third type of method involves pruning and trimming (superficial elimination). These terms correspond more closely to what should be aimed for, particularly the second, trimming, which is associated with trees which are processed especially to supply foliage. Pruning is only synonymous with trimming when the rule is respected whereby only branches which do not yet contain heartwood are cut. In both cases only the cutting of small branches is involved, and at the threshold with which we are concerned the best method is without doubt to cut only the leaves (blades and leaf stalks), so as to preserve as far as possible the buds and meristematic system of the plant, on which its growth depends.

This method can often be implemented by simply beating the crown of the tree, and this is often done for harvesting the pods of Acacia or other seeds. Obviously this method is a more demanding one as regards working time than lopping or felling, but it is the only one to show prudent management of the environment, `en bon pére de famine', as the French legal expression so aptly puts it. Any other attitude simply shows irresponsible and exploitative behaviour completely lacking in respect towards an environment which, after all, belongs to `other people'. There is really no need to explain to herdsmen the harm which they are doing, since they are generally very well aware of it. What is needed is social education in the strict sense, whereby a sense of conservation of the common welfare, or even the good of other people, is inculcated.

This said, I repeat that trimming can usually by carried out with a stick, particularly where tall trees are concerned. Once again in Cameroon, using tall samples of Daniellia, the daily ration was distributed in 1 or 2 hours by two herdsmen for about 30 animals. A single herdsman can thus implement this kind of distribution, taking his time throughout the day, especially as from the point of view of the livestock this activity does not rule out the direct consumption of accessible supplies. In the last analysis, we may ask whether the tree pruner should not became the basic tool of the herdsman, instead of the machete or the hatchet.

Many observations have been made with regard to the details and ways and means of intervention, but there have also been experiments which mean that operations in this field need not be completely `blind'. Amongst the most recent trials still in progress I shall cite those reported by Le Houérou in the Niono area of Mali. They were aimed at establishing:

a) utilization patterns;

b) the effects of trimming;

c) the effects of cutting back.

The main conclusions concern Combretum aculeatum and Cadaba farinosa, are as follows:

a) the more frequent the pattern of utilization the lower productivity falls, and this tendency does not appear to be compensated by the higher feed value of the forage harvested;

b) total utilization has a damaging effect on yield, as compared with partial utilization;

c) protein content is in inverse proportion to the frequency of utilization.

We have already shown that adult leaves have a high protein content. As regards trimming and cutting back as applied to Pterocarpus, these methods can enable the utilization period to be prolonged into the second half of the dry season. Trimming at the outset of the cool, dry season has better effects than when carried out in the middle of the dry, hot season, whereas cutting back "is thought" to have the reverse effect. In any case it was always true that trimming gave higher yields than cutting back, although this emerged less clearly for Acacia seyal. With Faidherbia albida trimming proved to be a far better method than cutting the larger branches, and intervention at the outset of the dry season also appeared more favourable than when carried out at the height of the dry season.

3.1.2 Beneficial activities

I referred above to the need to promote the `tree component', and the milder methods of foliage utilization are relevant in this connection. Measures intended to replenish stands, as we shall see below, are of a similar nature and purpose. However, in this connection I am reminded of the vast amount of needed propaganda for promoting a respect for what one might call tree lore in general, but especially for trees naturally occurring "in the bush". A tree which has been introduced cannot have the same symbolical value, especially since there have been so many "days of the tree", which have been expensive without leading to any tomorrows.

3.1.3 Replanting

In the broadest sense the problem of whether to regenerate by natural means or not is a question of the tree component, and here a prerequisite should be indicated: "there can be no reconstitution of forest stands in the Sahel without restricted grazing". And these restrictions on grazing should be seen as nothing other than one of the components of the overall management of pastoral resources, which for the sake of the future of the environment should be a constant and basic principle of everyday management.

The grazing restriction should also be extremely flexible, varying from total protection for several consecutive years against burning and grazing, to temporary forms of protection at the end of the wet season or the dry season, for example, depending on the stage which regeneration has reached.

Stage of regeneration. Under natural conditions and provided that seeds are locally present, most new plants originate from seeds, as has already been said and as is clearly shown by grazing restriction. From this point of view it is a good thing if the rainy season can begin without the soil having an excessive grass cover, since this indicates that the land has been used by livestock, who will have brought seeds and dung and mixed the two together enough to stimulate germination.

A few short grazing spells by cattle may subsequently be tolerated if they can be controlled so as to reduce the grass cover, but small ruminants and especially goats should be completely banned. As I have already said (Sedagri/IEMVT seminar, 1975): "Livestock do not cause adult ligneous formations to decline and frequently even overgrazing can lead to a degree of shrub invasion which is harmful to the grass formation. Small ruminants should thus be considered primarily as the enemy of reforestation, whereas the forest conservationist is able to make the best of them and even to see them as an aid to cleaning up the lower strata of plant formations". Moreover, it is not beyond the bounds of imagination to believe that the process of regeneration can be helped along, and no attempt will be made here to list the many facts known about this subject. However, under relatively extensive systems the planting of seeds, whether treated or not, may be envisaged on soil which may be either tilled or not as the case may be, using the recognized indigenous species.

Attempts might also be made to establish cuttings, but it should be remembered that although many plants can propagate via cuttings under experimental conditions, the difficulties under real-life circumstances are enormous, and became increasingly acute as the environment grows in aridity.

Le Houérou mentions nursery trials which are still in progress in the Niono area of Mali, using various cuttings and root stock fragments of the following varieties:

Combretum aculeatum, Acacia seyal, Faidherbia albida, Sclerocarya birrea, Bauhinia rufescens, Ziziphus mauritiana, Pterocarpus lucens, Balanites aegyptiaca, Cadaba farinosa, Grewia bicolor and Boscia angustifolia.

For my own part, in the Oursi area (300 mm rainfall annually) during 3 years of experiments, trials were carried out on Euphorbia balsamifera and Commiphora africana with a view to building hedges which would also be able to provide some relief for the other palatable species. More specifically, activities focused on research into cutting sizes and optimum planting dates.

The conclusions of the trial were as follows: "The strike rate for cuttings, which was very low throughout the 3 years, may be linked with the lack of water in the soil owing to abnormally low rainfall during the last three rainy seasons (387.6 mm in 1976, 288.8 in 1977 and 298.1 in 1978, plus 250 in 1979), according to measurements taken at the places where the cuttings were planted". There is no doubt that under these conditions the prospects of carrying out a proper study are rather uncertain, and we shall make further attempts to gain more specific information on these topics under more favourable conditions (at Dori which has 500–600 mm). Cuttings planted near the camp in 1979, and watered regularly experienced no problems in striking.

Nevertheless, with regard to E. balsamifera it transpired that cuttings with small diameters take root better than those with larger diameters; however, it would be difficult to come to any conclusion on this subject for C. africana, which must be protected against termites. As regards the length of cuttings, the longer ones, more entrenched in the soil, show a better ability to take root.

Finally, as regards planting dates, it was clearly shown that very early dates are not necessarily counterproductive. For this particular area mid-May seems entirely suitable.

It was obvious that cuttings planted in August gave better immediate results as regards taking root, but the roots did not go deep enough and when the dry weather arrived, strikes quickly died off. This emphasizes the limits which may be expected as regards climatic conditions, and it should not be forgotten also that as regards density the determining factor is local evapotranspiration conditions. For example, without certain specific conditions (accessible groundwater, depression, etc.), it is hopeless expecting to achieve a cover of more than 25 to 30% with Faidherbia albida when rainfall is less than 500 mm.

Intermediate stage. Once this initial establishment phase, lasting 2 to 4 years, is over, a second intermediate phase occurs in which the terminal shoots are out of reach of cattle but the stems are still not thick enough. Small ruminants, and specially goats, should still be absolutely prohibited, since they are capable of nibbling the bark and cambia through to a depth of 2 cm (Markoye, northern Upper Volta, 1975, Eucalyptus camaldulensis, 30–35 cm in diameter). Similarly, I was able to measure bites taken by large goats at a height of 2 m. They are true acrobats.

Trees of 3 or 4 m in height can be bent by goats and sheep, which then destroy the crown and often the whole tree, which does not recover its upright position.

Adult stage. Once they have become sufficiently resistant the trees can be placed in contact with any animal, man alone being still capable of responsibility, or rather blame, for their demise.

However, during the previous two stages and especially the second, it will have been possible to "train" trees to become more accessible to livestock by forming coppice or clump growths at various levels, etc.

It should also be possible to start appropriate trimming and harvesting operations during the intermediate stage, and these may normally be continued at the adult stage.

3.2 Management of artificial stands

This subject is a large one to tackle and a vast amount of knowledge is already being applied to combat the enormous problems which remain to be solved.

Of course the supply of seeds and the scarification methods referred to earlier come under artificial methods, but in the present context I am thinking more specifically of the results of planting previously prepared seedlings or cuttings.

This kind of intervention is applicable to dry or irrigated land, and the choice of plant material will thus vary considerably. We shall deal only superficially with this aspect, by outlining a few general rules regarding reforestation in the dry inter-tropical zone:

1. There is the possibility of planting local or exotic species, wherever annual rainfall exceeds 800 mm.
2. Between 800 and 500-600 mm it is possible to plant local species and very occasionally exotic species.
3. Below this level, reforestation will involve only local varieties, propagated primarily by means of restricted grazing.
4. In all cases, costly schemes undertaken for production purposes should be carried out only on good soils. Soil and environmental protection measures will almost exclusively depend on restricted grazing.

This said, once planting has been carried out the initial measures taken to obtain the necessary annual harvest will occur at an early stage. The reforestation measures to be undertaken will depend on the following choices:

a) direct utilization by the animal, or not;

b) green feeding (in the area itself or in enclosures), or harvesting followed by preservation.

The planting methods (hedges, clumps, etc.) will depend on these initial choices, as also will the densities, although the latter are often determined by environmental conditions.

I am deliberately leaving aside the question of extensive or intensive utilization, because the first alternative would seem impossible for the management of a plant formation of this kind. Even direct utilization by livestock cannot be implemented without firm instructions as to utilization dates, maximum stocking densities and the type of livestock acceptable (e.g., excluding goats while the plants are still young). Generally speaking there are three remaining alternatives: tall stems, clump growth at high or low level and coppice growths.

In cases where the stems are tall harvesting should be carried out manually (leaves or fruit) by means of trimming or beating, assuming that the manpower and materials necessary are available. This method will usually be unsuitable for forage preservation, but better suited to on-the-spot distribution of the harvest.

This system should hardly ever be advocated unless it is hoped periodically to harvest timber for construction (rather than firewood) in addition to animal feed. The other methods enable firewood also to be gathered.

Clump growths permit either manual harvesting (the easiest method) or direct utilization by animals, in the case of low-level growths, but this presupposes effective control of the herd.

If this is not the case, a high-level clump growth should be used instead of coppice growths, since the latter, as we shall see, also presuppose effective control of herds, especially during the budding stage. This type of treatment requires early intervention and an initial cut at the desired height 2 or 3 years after planting. Approximately 1 year later the same treatment should be carried out for seedlings under natural regeneration conditions. Herds will be able to make fairly rapid inroads into clump formations, the heads of which can be thinned out by immediate distribution to the animals, while at a later stage the remaining shoots can be harvested and stored before the leaves reach maturity.

For our purposes coppice growth are neither more nor less than clump growths at ground level. As such they will be selected owing to their greater ease of utilization and in cases where there is total control over the rangeland. Only one example need be given of what is possible in this field, and it is an example which, despite the fact that it is from a temperate country, is completely transferable, having features resembling those of the areas with which we are concerned. In addition it will draw attention to the qualitative and quantitative advantages which ligneous plants may have, even in comparison with the classically accepted forage crops.

I am referring to the study on the nutritive protein value of poplars, including the economic prospects for their utilization, carried out by Carlier (1978).

In this introduction he points out that, out of the 18 base amino acids, 5 are absolutely indispensable to animal life but cannot be synthesized by animals. Out of these 5, which are leucine, lysine, methionine, phenylalanine and tryptophan, lysine, which is necessary for good growth, is the most important for herbivores because it is the only one for which there is a feed deficiency risk, since the content of this substance in plant protein is very variable and often low, in many cases lower than in animal protein, which contains 6–12%. Under intensive hothouse conditions 2,500 kg/ha of proteins have been obtained (over seven times more than in a crop of Lucerne), but only 200 kg (1,500 kg of dry leaves) were obtained under poor conditions (such as drought, harvesting while the leaves were still at the development stage rather than at maturity, 1-year-old plants which were still too young, etc).

Although it makes a good green feed for goats, this feed source, which is valued at 0.8/0.9 FU per kd of DM, is freely consumed dry without any digestive problems for sheep and cattle also. Here we again come across a factor which we have already indicated previously, namely the possibility of consuming dead leaves which, in contrast to dry grass, retain a protein value which is still significant.

As regards the study of proteins as such, Carlier specifies the origin of the samples used as follows: "The young leaves were harvested on 15 June 1976 on the '1214' plantation of the Domaine de l'Etancon. They had not yet reached full development; some leaves at the tips of branches had not yet fully opened and presented the reddish colour which is typical of very young blades in this variety. Older leaves from the '1214' plantation (from the same source) were harvested on 10 October 1976. The plantation has two quite different areas:

a) In the first area the soil is maintained in good condition by mechanical tilling once a year. It has been fertilized every 4 years since the zone was set up in 1966; the trees are about 12 m high, 18 cm in diameter (at 1.30 m) and have abundant supplies of leafy branches.

b) In the second area the soil, which is covered with tall grasses, has never been fertilized. The trees are 5 to 6 m high, stunted (diameter is 8 cm at 1.30 m) and their branches have only a small number of leaves.

The other samples were taken from the enclosure for stock plants at the Domaine on 12.10.76, when some of the leaves at the base of the branches were beginning to turn yellow. All the leaves and branches represented the year's growth on stock plants which are cut back to ground level each autumn. It was not considered appropriate to take samples during June, as was done for the '1214' trees, because the shoots growing from the stumps only begin to develop during this period of the year, reaching maximal development only a few weeks before the leaves fall. Work would have been carried out on non-lignified wood and leaves which were very underdeveloped. The number of leaves increases as the stem grows longer. The soil in the enclosure is tilled once a year with a small rotary cultivator and is fertilized almost every year".

The results are given in Table 4 and Figure 1, which show total DM, protein and lysine contents of different clones and species of poplar and other crops.

Table 4. Comparison of protein and lysine production (per ha/year)

Figure 1. Comparison of lysine content in protein from traditional crops and varieties of Poplar

The conclusions are as follows:

1. Protein feed value of poplars. According to the analysis carried out its has been shown: 

a) that the total protein content of dry poplar leaves lies between approximately 9% (Deltoides, Grammont) and 15-16% ('1214'), with an exceptional result of 19% (Koreans);

b) that the protein contains large quantities of vital amino-acids (methionine, leucine, phenylalanine and lysine); poplar leaves are thus a good supplementary feed for livestock, a very valuable source of whole protein (especially lysine, a lack of which provokes arrested development); it should be noted that sheep show little interest in green leaves, but they will freely eat leaves which have fallen and turned brown, apparently without any digestive problems;

c) that the protein content of stems appears generally low and that the protein does not contain methionine.

2. Economics of utilization of a high-density coppice of poplars. If the protein and lysine production of a coppice with a density of 10,000 stocks/ha is studied (and the density can be increased to higher levels), the following measures should be taken:

a) in the first year the cuttings should be planted and their shoots should be cut back after the leaves have fallen; 

b) during the second year fertilizers should be applied, the soil should be weeded and cutting back should be again, carried out without any appreciable crop being possible;

c) In the third year (or the fourth for some varieties), large quantities of leaves can be harvested for use either in direct feeding (for example in the form of pellets with a low water content, similar to Lucerne pellets), or as a source of whole protein, or as the raw material for lysine production.

The production of protein with a high nutritive value from a coppice in which the soil is maintained in good condition and fertilized annually with 75 to 100 units of N, P and K is —whatever the variety of tree used —always higher than the best results obtained from traditional agricultural crops. The quantity of lysine extractable from the protein is equal in the worst cases, and three times as great or more in the best cases, to that obtained from Soya or Lucerne.

The production of livestock feed, plant protein or lysine from a copse of poplars is thus an operation in which profitability is guaranteed. In particular it is this last part, more concerned with forestry, which concerns us, since the majority of varieties which we possess should perform well in copses.

Moreover, there is no shortage of varieties (Neem, Calotropis, Boscia, Eucalyptus, etc,) which in the green stage are less appreciated than poplar except by a few animals with "depraved", tastes, whereas they may be suitable for use when dry.

No doubt spacing problems will have to be reviewed and the first cut for establishment purposes will have to be adapted to the initial growth speed, but it will remain a basic principle that only leaf crops for development purposes will be gathered and the harvest will be annual even if we have to wait 2 years (which will be an exception) for the coppice growth to become vigorous enough to be cut. On this simple basis alone, a large field of experimentation lies open, with, for example, a possible beginning on Albizzia lebbeck in the Niayes area of Senegal.

Finally, it should be borne in mind that for some species with small leaves or foliages preservation for forage is complicated by the fact that they easily fall from the tree, so that delicate handling is required.

4. Research topics

The whole of the present paper inevitably points to the discovery that there is a considerable lack of information, although in some fields this is very far from being the case.

Scientific publications throughout the world, for example contain large quantities of chemical analyses, frequently interpreted for feed value, and which adequately establish the qualitative value of tree feeds, so that this aspect has become a commonplace. From the simple point of view of forestry, few species pose any difficult technical problems as regards introduction, and the ecological limits are generally well understood. Even the problem of species introduction has been thoroughly tackled almost everywhere, although in strict terms it is an unlimited subject, especially as concerns the question of varieties or even that of the origin of plant materials. However, the quantitative aspects are very little understood, and in more specific terms knowledge is lacking on:

a) overall production of ligneous plants in individual species and formations (biomass in the broad sense of the term);

b) the intake capacities for different types of domestic animal, in as closely defined a relationship as possible with the grazing time available, and the question whether the animal feeds directly on the plant or is assisted.

As regards the qualitative aspect, it is now necessary by some means or other to raise the level of sophistication in these studies. Thus, the amino acid, and especially the lysine, contents of protein should be studied in greater depth, and amongst other things this might be a determining factor in the choice of species used in artificial plantations on dry or irrigated land.

Lastly, from the point of view of forestry, I mentioned above that the questions of variety and species should remain a constant preoccupation, but the details of treatment for each species also remain to be clearly defined. In particular, under extensive conditions where lopping will for many years remain the rule, the trimming methods which will be put forward as a substitute should be more clearly defined, both from the point of view of timing and also from that of the intensity with which these measures are carried out.

References

Boudet G. (1956), Manuel sur les paturages tropicaux et 1 es cultures fourragéres Maisons-Alfort, IEMVT.

Carlier, G. 'Valeur nutritive des protéines de peupliers et perspectives économiques de leur utilisation'. AFOCEL 78

CTPT–IEMVT (1978), 'Etude sur les arbres fourragers dens la zone sahélienne (Oudalan Voltaique)'. Maisons-Alfort, IELMVT.

Gillet H., et Moissang. (1969) 'Le Parc National de Zakouma (Tchad)'. Rome, FAO.

Granier P. (1977) Rapport d'activités agropastorales Republique du Niger. s.l., s.n.

ILCA (1975), Actes du colloque de Bamako Evaluation and Mapping of Tropical African Rangelands, Addis Ababa, ILCA.

Le Houérou H.N., (1979) 'Le rôle des arbreset arbustes dens les pâturages sahéliens'. s.l., s.n.

Monnier, F, Piot J. (1964) 'Problémes de pâturages dens l'Adamaoua (Cameroun)'. Bois et Forêts des tropiqttes, no. 97 et 98.

Piot J. (1966) `Etudes pastorales en Adamaoua Camerounais'. Rev. Elev. Méd. Vét. Pays Trop.

Piot J. (1966) 'Végétaux ligneux et pâturage des savanes de l'Adamaoua Cameroun'. Rev. Elev. Méd. Vét. Pys. Trop., 4.

Piot J. (1970) 'Pâturage aerien au Cameroun - Utilisation des ligneux par les bovins'. Rev. Elev. Méd. Vét. Pays. Trop., 4.

Piot J. (1975) 'Petits ruminants et forestiers'. Séminaire Sedagri. Maisons Alfort, IEMVT.

Poupon H. (1979) 'Structure et dynamique de la strate ligneuse d,une steppe sahélienne au Nord du Sénégal'. Thése de doctorat (ORSTOM).