Cover Image
close this book Use of Trees by Livestock : Prosopis - Acacia - Gliricidia - Anti-Nutritive Factors - Quercus - Ficus - Calliandra - Erythrina
close this folder Use of Trees by Livestock : Erythrina
View the document Contents
View the document Acknowledgements
View the document Foreword
View the document Genus Erythrina
View the document Summary
View the document Description and distribution
View the document Fodder characteristics
View the document Anti-nutritive factors
View the document Management
View the document Alternative uses
View the document References and further reading


The seeds of most Erythrina spp. are fertile and show little sign of dormancy (Rao and Singh, 1987), some, such as E. abyssinica may be hard seeded, a condition which can be rectified by scarification in either hot water or concentrated sulphuric acid (Laurent and Chamshama, 1987). Mechanical scarification is recommended for E. tahitensis (Powell and Nakao, 1992). Vegetative propagation is often favoured since most species root easily (Napier, 1988). Stakes of a minimum 2.5 cm diameter and 30 cm long will usually gro, but planting is often carried out using larger stakes of 10 cm diameter and some 2-2.5 m in length. Root formation, which usually takes place readily, can be further stimulated by the use of hormonal preparations, or by stripping the bark from the lower end of the part to be buried in the soil (Teketay, 1990). This end is cut obliquely and the upper end is protected against both rain and desiccation by the application of tar or lime. In a number of countries, apicormic shoots (large shoots that grow vertically upwards from horizontal branches) of E. costaricensis, E. variegata, E. poeppigiana and E. senegalensis have proved to be particularly successful as planting material (Jolin and Torquebiau, 1992).

Planting densities in the range of 1200-10 000 trees/ha are common. Lower densities are used where groups of trees are intended to provide shade and browse within a pasture, while at the upper end of the range, the trees would form a pure stand with very little understorey. When large stakes are used, growth is rapid and the first harvest can often take place within some 6-8 months of planting (Preston and Murgeitio, 1987).

Erythrina spp. form large, spherical nodules, which tend to be clustered on the central tap root. They are infected promiscuously by Bradyrhizobium spp. of the cowpea group, and therefore rarely require inoculation. The list of species which have been shown to nodulate includes all of those which are commonly used in animal production (Allen and Allen, 1981). At the low tree densities employed as shade for coffee and cacao, annual nitrogen fixation rates are 12-40 kg/ha (Budowski et al., 1986). Active mycorrhizal associations are formed which are valuable in soils with low phosphorus contents (Powell and Nakao, 1992).

Grass-based pasture growing under seven-yearold trees of E. poeppigiana established as shade for grazing animals (about 60 trees/ha) was compared with similar pasture growing either without shade or under the tree species Cordia alliodora, Albizia saman (syn. Samanea saman) or Gliricidia septum in Costa Rica, grass yields were similar in all cases, but the fibre content of the grass was higher in the unshaded areas. The CP content of the grass was highest under E. poeppigiana, and lowest under C. alliodora (Daccarett and Blydenstein, 1968). While direct measurements of the nitrogen fixing capacity of Erythrina spp. are lacking for pasture systems, these data would suggest that it is at least comparable to that of G. septum, a much-used fodder tree in tropical regions.

Frequent cutting appears to be advantageous when Erythrina spp. are grown for animal production. In a coffee plantation in the humid tropics of Costa Rica (2600 mm annual rainfall) where E. poeppigiana was planted for shade at a density of 280 trees/ha, the legume was cut at intervals of 4,6 or 12 months. The most frequent pollarding produced the lowest total yield of dry matter (woody stems, edible stems and leaves), but the highest yield of edible stems and leaves. With frequent cutting, the edible fraction was some 64% of the total biomass production, and contained about 85% of the total crude protein. The total amount of nitrogen in the cut material was 170 kg/ha at three cuts per year, compared with 230 kg/ha under more frequent cutting systems (Rodriguez, 1985; Russo and Budowski, 1986).

In a study conducted over two years in a similar area of Costa Rica E. poeppigiana was planted in association with King grass and neither weeded nor fertilized. The trees were established as large stakes at either 1667 or 3333 trees/ha, and cut three or four times a year. When harvested at varying intervals, on attainment of a height of 2 m, grass production was not reduced by the presence of the trees at either density, but the crude protein content (average over two years) was increased from 4.7 to 6.1 %. The higher tree density almost doubled the yield of tree fodder. More frequent cutting of the trees slightly increased tree leaf yield, but had little effect on crude protein content, which averaged 26.3% for the leaf and 11.7% for the stem (Benavides et al., 1989). Since the stems of the trees are of limited feeding value, and leaves may be shed in response to dry weather, livestock production would benefit from management under frequent cutting, although the growth rate of the trees, as influenced by climatic, edaphic and biotic factors, would govern the optimum interval between harvests. A heavy pruning of Erythrina spp. not only promotes the yield of leaves and prevents loss of feed resources resulting from leaf fall, it also appears to prevent the early senescence that is sometimes observed in trees after a few years of active growth (Seibert, 1987).