Session 1 Botany and Distribution

Session Papers

Germplasm Collection of Faidherbia albida in Eastern and Southern Africa

C. W. Fagg1

Abstract

The Oxford Forestry Institute is making systematic range-wide collections of Faidherbia albida (Del.) A. Chev. in undercollected areas of eastern and southern Africa to assemble germplasm for international provenance trials. The wide and discontinuous distribution suggests there may be complex patterns of genetic variation. There is also variation in morphology. Highly significant differences in size and weight of pods were found between trees, within and between provenances, indicative of an outcrossing species. The large variation in phenology is also expected to be under genetic control. The wide use of the trees in farming systems in parts of its southern range and the influence of human disturbance on its distribution introduce complications that must be taken into account. A sampling strategy has been devised. In collaboration with other institutes a broad range-wide seed collection is becoming available for testing.

Introduction

Faidherbia albida (Del.) A. Chev., syn. Acacia albida, is widely distributed throughout the dry zones of Africa and parts of the Middle East. It occupies a wide range of different habitats, indicating wide patterns of genetic variation. It is one of the four African acacia species identified—with A. tortilis, A. nilotica, and A. Senegal—as having particular potential for community forestry in dry areas (Palmberg 1981). They are priority species within the project sponsored by the Food and Agricultural Organization of the United Nations and the International Union of Forestry Research Organization (FAO/IUFRO), for exploration, evaluation, and conservation of genetic resources of dry zone species in 16 countries in the Sahelian and northern Sudanian zones (IUFRO 1988). While some seed collections of these species have been made, particularly in West Africa, the Oxford Forestry Institute (OFI) has initiated a project aimed at systematic range-wide collections in under-collected areas, notably in eastern and southern Africa, and assembling range-wide germplasm for international provenance trials and related research. Extensive field trips have been made in the natural range of these species in eastern and southern Africa, and a large number of provenances identified.

Taxonomy and Variation

F. albida has many characters which in combination make it a very distinct species, and for this reason it has been placed in the monotypic genus Faidherbia (Wood 1992). The most notable variation is in pod morphology. Pods can vary considerably in size, color, and form (curled in a circle, falcate, or twisted) between neighboring trees in a population, but appear to be unrelated to other morphological characters (Ross 1966). To test the variation in this character, 4 provenances from Malawi and Zimbabwe were sampled. Ten pods were randomly sampled from about 20 selected tree canopies. Length, taken from the central axis; mean width, taken at one-third and two-thirds of the length; and mass were measured. Highly significant differences were found between trees and provenances for pod dimensions and mass. Variation has also been found in seed mass between eastern, western, and southern African provenances (Sniezko and Stewart 1989).

Leaf phenology of F. albida is very variable (Wickens 1969) both within the same locality and between localities, so that some trees in the same locality can be in full leaf and completely leafless. In the arid Namibian desert, trees rarely lose their leaves except for about a 2-week period every year. Evidence from trials both in West Africa and Zimbabwe indicates large differences in flushing times between provenances, suggesting genetic control of this trait.

Distribution and Ecology

After a tentative distribution of the species was deduced by logging a large number of herbarium specimens, suitable populations were identified during collaborative fieldwork in most eastern and southern African countries; the provenances identified are shown in Figure 1.

The species occurs in a wide range of habitats and is most successful either on sandy alluvium or where land has been disturbed by man or livestock. It frequently colonizes sandbanks in rivers because of its ability to withstand flooding. It will also sucker, especially in loose, sandy soils, as has been noted in Namibia and Zambia. While it occurs more commonly in riverine and lakeshore habitats, nonriverine, hilly, and plateau populations have also been found.

The tree is found on a wide range of alluvial soil types. For example in Malawi, the majority of provenances occur on alluvial calcimorphic soils, but populations can also be found on ferruginous soils (sandy clays), weathered ferallitic soils, Vertisols, and gleys. Trees have been noted in saline areas such as near Lake Chilwa (Malawi) and near the Sowa Pan (Botswana).

In Somalia, Kenya, Uganda, and northern Tanzania, race A (Brenan 1983; Wood 1992) is not generally gregarious, and it is usually found thinly scattered in riverine woodland, whereas race B, which occurs from southern Tanzania southwards, can form large extensive monospecific stands over parts of its range (Malawi and Zambia) similar to those found around Wadi Azum in western Sudan. These areas support large populations, and the trees may have spread in part due to the influence of man and his livestock.

Cultivation

While it has a very widespread distribution throughout Africa and the Middle East, F. albida was known primarily as a Sahelian farm tree, but it has also been used in traditional farming systems in many countries in eastern Africa. Little tree planting takes place, but naturally regenerating seedlings are pruned and encouraged to develop in farmer's fields. Seeds can regenerate near the mother trees, or can be spread by animals or water.

In Zambia, the largest populations are found on alluvial soils in the fringes and in the surrounding areas of the Kafue flats. A survey of the area indicated that F. albida farmed parkland covers approximately 770 km² (Pullen 1974). Larger stands can be found in a triangle from Namwala to Masabuka and down towards Choma. Traditionally, the Tonga farmers own large herds of cattle, which graze on the floodplain grasses of the Kafue flats when the floodwater recedes and return to the higher ground, where the trees are frequent, during the flooded period. Then the farmers also cultivate beneath the trees. Around Namwala, considerable regeneration takes place even within the kraals, dispersal being aided by the livestock.

In Malawi, the benefits of cultivation under F. albida trees are well known and appreciated (Saka et al. 1990). In some areas the trees are known to provide "God-given fertilizer", and maize yields are known by experience to be better under the trees. More importantly, in "poor rainfall years, usually some harvest occurs under the trees where there may be failure in open fields." Most Malawian soils are impoverished, particularly in nitrogen and phosphorus, and preliminary research (Saka and Bunderson 1989) has shown similar increased maize yields under the trees in Ethiopia (Poschen 1986). Some of the largest populations are found along the lakeshore plains surrounding Lake Malawi, or on the medium altitude plateau.

In Kenya, distribution is more scattered, and only a large population around Tot appears to be integrated into the farming system. The traditional irrigation canals of the Pokot, which divert water from the watershed of the Cherangani hills, feed into fields with scattered trees of F. albida. The trees are also heavily lopped for fodder.

Cultivation is also common under the trees at the Usungu flats, in southern Tanzania. More recently,  the trees have been used as shade over coffee in the Arusha district of northern Tanzania.

Figure 1. Provenances of Faidherbia albida in eastern and southern Africa.

In parts of southern Africa, outside the native range of the species, farmers and gardeners have planted individual specimens. Once established, the trees flourish, although there are indications that they do not necessarily flower in these offsite plantings. Human influence does have a significant effect on the genetic distribution of the species and must be taken into account in the sampling strategy.

Seed Collections

Sampling Strategy

The methodology for seed collection of nonindustrial tree species has been developed over the past decade at the OFI and is outlined by Hughes (1987). Collections are as broad-based as possible to gain the widest variability for the multiple traits that the recipient farmer may require. A number of broad guidelines have been adopted specifically for the seed collection strategy of F. albida:

Provenance Selection

•  Geographical discontinuity from other populations.

•  Sites with differences in climate, altitude, soils, and ecology.

•  Selection from both undisturbed natural populations and populations in agricultural systems where there is a choice.

•  Distinctive morphological and phenological attributes.

•  Cover differences suggested by isoenzyme studies and established trials.

Seed Collection in Provenances

•  General provenance collection. 25 or more trees, spaced at least 100 m apart, from which comparable amounts are bulked to provide 5-10 kg of seed.

•  Progeny collection. 25 or more individual trees randomly selected, 100 m apart, pods from all parts of the crown.

•  For breeding system studies, collecting every tree in a defined population of 100-200 trees.

•  For genetic variation in isozyme studies, 100 individual trees randomly selected more than 30 m apart, 10 pods from over the crown, providing 50 viable seeds.

A pragmatic approach to seed collection is needed as it is not always possible to follow the ideal strategy because of poor seed years, difficulties in locating sufficient trees with pods, logistical constraints, and timing.

As well as seed and botanical voucher specimens, other materials, and associated site information (exact location, vegetation, soil, and tree characteristics) are also collected. Nodules have been collected from young seedlings, placed in small glass vials with a desiccant, and sent to the University of Dundee (Prof. J. Sprent), for isolating and culturing the rhizobial strains. Nodules can be easily collected only in the rainy season, and as seed collections occur during the dry season, only those seedlings in or beside rivers have been found with nodules. An alternative method, germinating the seed in a nursery with soil collected under the trees, has proved much easier for isolating the nodules. Wood and insect pollinators and predators have also been collected.

One problem in seed collection is the predation of seed by bruchid beetles, though infestation rates on F. albida appear to be less than with other acacia species, such as A. tortills. The species of bruchids also vary depending on the taxa they attack, and specimens of them and their parasites are being collected, to identify the number of species involved. Variation in susceptiblity between trees and between provenances is seen in the field. The adult bruchids lay their eggs on or in the young green pods, and the larvae enter the young seed a few weeks after to pupate. Some will emerge before collection, but even if seed is meticulously cleaned during extraction, inevitably, further emergence of the weevils from within apparently "viable" seed can occur up to 16 months after collection (Ernst et al. 1989). Seed can be x-rayed to determine the infestation, and later cleaned on a gravity table separator at the seed bank.

Conservation Status

While the F. albida species is not under direct threat, a number of its populations have been eliminated or severely depleted. In Malawi, where there are very high demographic pressures, the demand for dugout canoes has reduced the provenances to a few solitary individuals in the eastern shores of Lake Malawi. Large numbers of Mozambican refugees have contributed to deforestation in the Lower Shire, and very few trees now remain. In other areas of Malawi, ex- tensive populations appear to be protected and not threatened to the same extent. Some of the populations also suffer from little regeneration in other parts of the range, particularly where the wild life and game pressures are severe.

Seed Availability

Although collections are still in progress, some seed is now available for the establishment of species field trials (Table 1). For species trials, several provenances should be tried, as a single locality is unlikely to provide a good indication of the potential of the species. A choice of at least three provenances widely distributed geographically is recommended: (1) a site that closely matches the trial site, (2) an area where it reaches its finest development, and (3) a more marginal site. Seed should be accompanied by detailed site description forms.

Range-wide provenance collections are expected to be available for trials within a year, including sites from West Africa.

Acknowledgment. Fieldwork for this paper has been carried out both by Dr R. Barnes and the author in different areas of eastern and southern Africa. The work has been dependent on the support and collab-oration of the following institutions: the National Range Authority (NRA) and the British Forestry Pro-ject (BFP) in Somalia; the Kenya Forestry Research Institute (KEFRI) and the East African Herbarium in Kenya; the Tanzania Forest Research Institute (TAF-ORI) and the Department of Forestry and Beekeeping in Tanzania; the Soil Productivity Research Program (SPRP) and the National Parks and Wildlife, Division of Forestry Research, Department of Agriculture (Zambia); the Forestry Research Institute of Malawi (FRIM) and the Ministry of Agriculture and Lands (MAL) in Malawi; the Forestry Research Center (FRC), National Herbarium, and Department of National Parks (Zimbabwe); the Forestry Association of Botswana (FAB), the Gobabeb Research Station (Namibia), and the Botanical Research Institute, Pretoria, Republic of South Africa. Numerous individuals have contributed to this work for which we are very grateful. The work was financed by the UK Overseas Development Administration under research schemes R.4348 and R.4583 at OFI.

References

Brenan, J.P.M. 1983. Manual on taxonomy of Acacia species: present taxonomy of four species of Acacia (A. albida, A. Senegal, A. nilotica, A. tortilis). Rome, Italy: Food and Agricultural Organization of the United Nations. 47 pp.

Ernst, W.H.O., Tolsma, D.J., and Decelle, J. 1989. Predation of seeds of Acacia tortilis by insects. Oikos 54:294-300.

Hughes, C.E. 1987. Biological considerations in designing a seed collection strategy for Gliricidia sepium (Jacq.) Walp. (Leguminosae). Commonwealth Forestry Review 66(l):31-48.

IUFRO (International Union of Forestry Research Organizations). 1988. Tree improvement and silvopastoral management in Sahelian and North Sudanian Africa. IUFRO Special Program for Africa. Nairobi, Kenya: International Centre for Research in Agroforestry.

Palmberg, C. 1981. A vital fuelwood gene pool is in danger. Unasylva 33(133):22-30.

Poschen, P. 1986. An evaluation of the Acacia albida based agroforestry practices in the Hararghe highlands of eastern Ethiopia. Agroforestry Systems 4:129-143.

Pullen, R.A. 1974. Farmed parkland in Zambia. Zambian Geographical Association Magazine 26.17 pp.

Ross, J.H. 1966. Acacia albida Del. in Africa. Bole-tin de Sociedade Broteriana 2(40): 187-205.

Saka, A.R. and Bunderson, W.T. 1989. The potential of Acacia albida and Leucaena leucocephala for smallholder crop production systems in Malawi. Pages 150-161 in Trees for Development in Sub-Saharan Africa: proceedings of the IFS Regional Seminar, 20-25 Feb 1989, Nairobi, Kenya. Nairobi, Kenya: International Centre for Research in Agroforestry.

Saka, A.R., Bunderson, W.T., Mbekeani, Y., and Itimu, O.A. 1990. Planning and implementing agroforestry for small-holder farmers in Malawi. Planning for Agroforestry (Budd, W.D., ed.). Amsterdam, Netherlands: Elsevier Science Publishers.

Sniezko, R.A., and Stewart, H.T.L. 1989. Range-wide provenance variation in growth and nutrition of Acacia albida seedlings propagated in Zimbabwe. Forest Ecology and Management 27:179-197.

Wickens, G.E. 1969. A study of Acacia albida Del. (Mimosoideae). Kew Bulletin 23(2) 181-202.

Wood, P.J. 1992. The botany and distribution of Faidherbia albida. These proceedings.

Footnote__________

1. Oxford Forestry Institute, OX1 3RD, South Parks Rd, Oxford, UK.

Fagg, C.W. 1992. Germplasm collection of Faidherbia albida in eastern and southern Africa. Pages 19-24 in Faidherbia albida in the West African semi-arid tropics: proceedings of a workshop, 22-26 Apr 1991, Niamey, Niger (Vandenbeldt, R. J., ed.). Patancheru, A.P. 502 324, India: International Crops Research Institute for the Semi-Arid Tropics; and Nairobi, Kenya: International Centre for Research in Agroforestry.