Introduction of browse plants in the Sahelo-Sudanian zone**

P. Gosseye

Associate expert (Agronomy), FAO/ILCA, Bamako, Mali

1. Introduction

2. General remarks

3. Nursery studies

3.1. Experimental procedure

3.2. Results and discussion

4. Planting on the ranch

4.1. Experimental procedure

4.2. Results

5. Conclusions

1. Introduction

The importance of browse plants in the Sahel and Sudano-Sahelian zones for animal feeds no longer needs to be demonstrated. Within the general context of the study of these ligneous plants, whether it concerns their ecology, their productivity as related to utilization methods, their period of utilization or their nutritive contribution for different ruminants, there was a need to test different propagation and planting methods and to try out the introduction of various local and exotic varieties.

Nursery propagation methods include drilling seeds or planting rootstocks and various kinds of cuttings. Starting procedures are either potting at the end of the dry season (May–June) followed by planting out at the beginning of the rains (July), or else direct drilling. The question of how to treat the seeds in order to increase their germination rate also arises.

Methods have to be developed which, ideally, can be applied by arable farmers or livestock producers, taking into account the amount of labour required in their major farming activities. For arable farmers, the end of the dry season is taken up with clearing the fields, an occupation which leaves some time free. As soon as the rains come in sufficient quantities to make the soil workable, all the available time is devoted to subsistence crops, either the dry land crops of the Sahelo-Sudanian zone (mid July-mid August), or else for crops grown in rice plots.

2. General remarks

The nursery trials were carried out at the SERZS¹, in Niono, on the Office du Niger (ODN) irrigated perimeter. The planting out trials were carried out on the SERZS ranch situated 7 km east of the station and 4 km south of N'tila. The rainfall station on the ranch is 5 km northeast of the trial areas. The rainfall data are presented in Table 1.

Table 1. Monthly rainfall for browse plant sites on the ranch, 1977–1979^a_.

^a L. Stroosnijder, unpublished date

¹Station d'Elevage et de Recherches Zootechniques du Sahel.

The soils consist of dunes formed by wind erosion during the present epoch and maximum slopes of 3%. The upper layer consists of very fine loamy sand. At a deeper level there is very fine sand loam (Table 2). Drainage is excellent but the surface has a marked tendency to form a crust owing to its granular composition and the presence of blue-green algae. For rainfall in excess of 20 mm this situation can give rise to runoff losses amounting to 50%. The soils are classified as ochric arenosols (FAO-UNESCO, 1974).

Table 2. Granulometric composition of dune soils on the ranch.

Texture	depth (cm).
Clay 2 um	0-10	30	80	150
	6	13	9	10
Alluvium 2-5 um	20	18	15	14
Sand 50-105 um	60	56	64	66
Sand 105 um	14	13	12	10

^aL. Stroosnijder, unpublished data.

3. Nursery studies

3.1. Experimental procedure

These trials involved seeds, rootstocks and cuttings.

Seeds. In 1977 sowing was carried out in pots made of plastic wrappers filled with a mixture consisting of an equal volume of earth, coarse sand and manure. Each pot was planted with three seeds between 28/5/77 and 3/6/77 and placed under shade with regular watering. Some species of seed were treated with hot water. This method consisted of immersing the seeds in boiling water, leaving the mixture to cool and recovering the seeds 24 hrs. later, followed by immediate sowing.

In 1978 various methods of treating seeds were tried out: these included 20% sulphuric acid, a non-titrated solution of caustic soda, mechanical scarification, polishing with an emery cloth, soaking in boiling water from 30 seconds to minutes, followed by immersion in cold water for between 48 and 72 hours, depending on the size of the seeds. Only the latter treatment was retained and subsequently applied to all the varieties used, except for Atriplex. As soon as soaking was finished the seeds were sown at a rate of 4 per pot, between 5/4/78 and 5/5/78, placed under shade and regularly watered. The pots and the mixture used were the same as for the 1977 trial. In 1979 a germination trial on untreated seeds was carried out. Four batches of 110 seeds per variety were placed between two wads of cotton wool and covered with jute sacks. Plentiful watering and a sorting operation to eliminate germinated seeds were carried out each morning.

Rootstocks. Dressed rootstocks were taken from local trees on the ranch and planted in the nursery, without shading but with regular watering. This operation was carried out in June 1977.

Cuttings. In 1977 propagation from cuttings involved three different parts of the plant: non-lignified cuttings, lignified cuttings and heel cuttings. All three kinds were dressed in a mixture of earth and cow-dung and planted in nurseries without shading but with regular watering.

On 10/7/78 a propagation trial was carried out. The parts used were sprigs which had become well lignified. The cuttings were planted in the same type of pot, filled with the same mixture; shading was provided and there was regular watering. The trial involved ten varieties and 8,698 cuttings.

Development in the nursery. Following drilling in 1978, as soon as the younger plants put out their second leaf the containers were heeled in, placed in full daylight and regularly watered. After a fresh wave of mortality the plants were thinned out to one stock per pot.

3.2. Results and discussion

Germination. The results of the germination trials on seeds in pots (1977) and between cotton wads (1978) are shown in Table 3. Only the more promising species are given. Most of them are of local origin. The results for 1978 are presented in Table 5. It should be noted that exotic ligneous species were more successful in 1978 than 1977. The legumes in particular showed a fairly satisfactory survival rate.

The following list gives the species which showed a germination rate of between 2 and 3%. The seeds were treated with hot water, with the exception of Dalbergia sissoo, Acacia cyanophilla, A. ligulata, A. salicina, A. tortilis, Dalbergia sissoo, Prosopis cineraria, P. juliflora and Ziziphus nummularia.

In the case of the species given in the following list, either there was no germination or else the young plants quickly died off: Acacia angustissima, A. currassavica, Atriplex canescens, A. halimus, A. nummularia, A. paludosa, A. undulata and A. versicaria.

On examining Tables 3 and 5 we shall be able to classify seed treatments according to their effectiveness with regard to germination, under controlled conditions with watering and protection against predators.

1. Soaking in boiling water from 30 seconds to 2 minutes, followed by bathing in cold water from 48 to 72 hrs according to the size of the type of seed.
2. Soaking in boiling water followed by bathing in the same water which gradually grows colder over 24 hours.
3. No treatment at all.

Only Combretum aculeatum seeds in good condition germinate better without any treatment at all. The non-emergence of species shown in the list following Table 5 could be due to a loss of germinating power following prolonged storage in conditions which were not ideal. Some varieties did germinate but rapidly perished by damping-off.

Table 4 shows three categories of recovery rate for rootstocks. It should be noted that the rootstocks were placed in favourable conditions and watered. Unfortunately there is nothing to be said on the strike rate for these root-stocks under dry conditions. Hence these operations should be carried out only when rains have well and truly set in.

Table 3. Germination rate (%) for species sown in pots and between cotton wads.

^a The following did not germinate: Pterocarpus lucens, Cadaba farinosa.

^b Seeds treated with hot water.

Table 4. Strike rate (%) for root-stocks and various types of cuttings.

Table 5. Classification of species according to nursery survival rate and relative emergence.

The propagation method is slow and requires plenty of manpower.

Cuttings. Table 4 shows that the utilization of cuttings is disappointing. The experiment used several different diameters of wood. Only the larger diameters (2–2.5 cm) provided any real chances of recovery. If we had used only the thicker diameters for these trials, the strike rates for cuttings in Table 4 would have been approximately double.

During this experiment we found that the cuttings showed an apparent recovery since dormant buds opened out with the growth of sprigs left on the stalk. As reserves were gradually exhausted the cuttings died off without putting out any roots. Root development appeared to be very slow and as a result planting out during the same season has been ruled out. Consequently we have to keep the plants throughout a dry season before planting them out at the beginning of the following rainy season.

The final number of live shrubs from potted cuttings is accounted for by the low strike rate (3%), losses from storage and the final percentage of recovery for stocks planted out. This method could be used for propagating seed-bearers for tree species where seeds are not yet available.

Owing to the slow progress of root development and the needs for shade and watering, this method can be ruled out for direct use in dry zones (350–500 mm rainfall).

Nursery development. Table 5 gives the result for development in the nursery of the young plants after drilling. The percentage emergence rate is relative and not absolute. Each pot was sown with four seeds, but a pot with one or four young plants when removed from shading has the same value for the calculation of the percentage emergence. In other words a 70% emergence rate means that for 100 pots sown, i.e. 400 seeds, 70 pots were successful, producing 70 and 280 seedlings. On the other hand, a nursery survival rate of 70% means that for 100 pots sown 70 young plants were suitable for planting out on emergence, once the early growth had died off and the plants had been thinned out to one stock per pot.

The following species did not emerge at all: Acacia cambage, A. currasavica, A. estrophiolata, A. farnesiana, A. georginae, A. stenophylla, Atalaya hemiglanca, Atriplex atacanensis, A. breweri, A. canescens, A. glauca, A. vesicaria, Brachychiton gregorii, Colospermum mopane sv. Impala, Combretum heroresence, C. imberbe, Eremophila longifolia, Geijera parvifolia, Grewia flava, Hererodendron olgifolium, Prosopis tamarugo, Terminalia sericea and Ventilago vimenalis.

4. Planting on the ranch

4.1. Experimental procedure

Plants grown in pots. In August 1978 the plants in the nursery were planted out on the livestock ranch. For the purposes of this operation furrows 30 cm deep were dug at 10 m intervals with the displaced soil on the down slope side and as far as possible following the contour lines. The plastic wrapper was carefully split all the way up the pot and the stocks were then placed at the bottom of the furrow, at intervals of 10 m. The planting pattern was arranged so that the humidity gradient between the top of the slope and the bottom was represented for each species. It should be noted that out of 419 mm of rain during 1978, 278 mm had already fallen before planting, 24 mm fell during planting and 116 mm came after planting was over.

Counting operations were carried out after the rainy season, at the beginning of the 1979 rains and again in December 1979. The experiment required the use of 14, 122 pots for 48 varieties, of which only 1,897, representing 19 varieties, were actually tried out on the ranch.

Direct drilling. In 1979 the emphasis was placed on direct drilling in situ at the end of the dry season. The advantages of this method were its rapidity, the fact that it could take place during a period which was not too busy and that it was very easy to carry out in comparison with the method of sowing in pots followed by planting out.

The seeds were given two treatments. The first consisted of planting them as they were. The second consisted of scalding them from 30 seconds to 2 minutes according to the size of the seed, washing them in cold water and then leaving them to dry in the shade. The 18 furrows for the 1978 trial were used again without reploughing. The seeds were drilled between the bottom of the furrow and the top of the ridge formed by the displaced soil. Each variety was established in a group of ten pockets with three seeds per pocket and 30 cm between each pocket, with each being marked by a peg. One metre was left between each group. Each variety and treatment were represented in the humidity gradient by the slope. Two replicates were mounted, i.e. a total of 7,920 pockets, at a rate of 720 pockets per variety. Drilling was carried out from 7 to 13 June 1979, when 39 mm out of a total of 389 mm of rain had already fallen, while 23 fell during planting operations and 327 mm occurred after planting (Table 1).

4.2. Results

Plants grown in pots. Survival trends during the 18 months following planting again showed that local species were more successful than exotic varieties (Table 6). However, for all species mortality during the early dry season is fairly serious and the survival rate continues to fall off during the following six months.

Table 6. Strike rate (7%) for browse plants planted out in pots on the ranch at three observation dates.

Acacia aeneura, A. brachystachya, A. ligulata, A. senegal, A. seyal, Atriplex nummularia, Ziziphus nummularia did not survive the first rainy season.

From Table 7 it can be seen that emergence falls to very low levels in the natural environment and that scalding the seeds does not seem to encourage emergence. A more efficient method could be used, such as scalding followed by prolonged soaking in cold water, but this would imply drilling in moist soil to ensure continued germination, which would delay drilling until mid-July or the beginning of August, a period devoted to food crops. It may also be seen that without treatment germination spreads out in time. Only in the case of Combretum aculeatum seeds in good condition is there complete germination in under a fortnight.

Table 7. Emergence rate (%) at end of October 1979 for direct in situ drilling on the ranch.

^a seeds immersed in boiling water for 0.5-2.0 minutes.

5. Conclusions

The method involving the use of pots, a nursery and planting out has the disadvantages of being expensive, of requiring plenty of manpower and means of transport.

Nursery survival rates are highly variable but generally good. Mortality reaches its highest point a few days after the plants have been brought out into full sunshine. To reduce this mortality it will be necessary to harden the young plants by a process of gradual exposure. Our plant shelters were not designed with removable roofs, but if they had been they could have been opened during the morning and evening.

Direct in situ drilling is far more economical but requires more seeds on account of the losses due to predators. It also appears that to succeed this method requires a year which is not only rainy but also one in which the rains are well distributed, a factor which can never be catered for in advance. The results obtained by this method could be improved by a proper treatment of the seeds, if possible by a dry method so as to avoid triggering off a germination process which will afterwards be thwarted by lack of water. If a wet method is used drilling must be carried out in moist soil. The young plants obtained by in situ drilling are also very sensitive to competition and require hoeing. In situ drilling for windbreaks or hedges might be tried out for seeds which have passed through the digestive system of an animal. Whatever method is used, restricted grazing and protection from burning for at least three rainy seasons are necessary.

We were also able to establish that local varieties were more successful. These varieties are self-propagating under conditions which suit them. In this case also, minimum management of rangeland is required to avoid deforestation and ensure natural regeneration.

Reforestation with local browse plants would be necessary in cases where deforestation is too advanced. Exotic varieties however, need not be ruled out altogether.