tree, CATIE, legume trees, Gliricidia sepium, Erythrina poeppigiana, legume tree, biomass production, king grass, African star grass, intercropping, milk production, soybean meal, intake, Tropical Agricultural Research and Training Centre, levels, protein source, body weight, Response variable, milk yields, nutrients, C. Devendra, farm animals, ruminants, forage system, nitrogen fixation, agroforestry system, The International Development Research Centre, protein supplement, weight gain, Universidad de Costa Rica, IDRC, Nitrogen-Fixing Tree Association, Turrialba, Turrialba, Costa Rica, Bali, Indonesia, Costa Rica, the Parliament of Canada, shade trees, pruning, International Development Research Centre, research and development, coffee plantation, coffee plantations, les animaux, Denpasar, Agroforestry systems, tree foliage, Canada K1G 3H9 Devendra, Denpasar, Indonesia, supplements
DEVEND TREE FODDERS
49268 OR FARM ANIMALS
- PROCEEDINGS OF A WORKSHOP IN DENPASAR, INDONESIA, 24 29 JULY 1989
The International Development Research Centre is a public corporation created by the Parliament of Canada in 1970 to support research design
ed to adapt science and technology to the needs of developing countries
. The Centre's activity is concentrated in six sectors; agriculture, food and nutrition sciences; health sciences; information sciences; social sciences
; earth and engineering sciences; and communications. IDRC is financed solely by the Parliament of Canada; its policies, however, are set by an international Board of Governors. The Centre's headquarters are in Ottawa, Canada. Regional offices are located in Africa, Asia, Latin America
, and the Middle East. Cover photo: D. Brian Palmer.
Shrubs and tree fodders for farm animals Proceedings of a workshop in Denpasar, Indonesia, 24-29 July 1989
Editor: C. Devendra
IDRC CA N AD A
© International Development Research Centre 1990 PO Box 8500, Ottawa, Ont., Canada K1G 3H9
Devendra, C. IDRC. Regional Office for Southeast and East Asia, Singapore SG IDRC-276e Shrubs and tree fodders for farm animals: proceedings of a workshop in Denpasar, Indonesia, 24-29 July 1989. Ottawa, Ont., IDRC, 1990. xii + 349 pp.:ill. /Trees/, /fodder/, /feed/, /animal nutrition/, /developing countries/ -/animal husbandry/, /agroforestry/, /ruminants/, /research and development/, /forage/, /Case Studies
/, /conference papers/.
Technical editor: WM. Carman
A microfiche edition is available.
The views expressed in this publication are those ofthe authors and do not necessarily reflect the views ofthe Centre. Mention ofaproprietary name does not constitute endorsement ofthe product and is given only for information.
Abstract This publication presents the results of an international meeting held in Denpasar, Bali, Indonesia, 24-29 July 1989, that focused on the use of shrubs and tree fodders by farm animals. Through 26 papers, the workshop addressed feed-resource availability, use by ruminants and nonruminants, processing methodology, economics, and development issues. These aspects and the current knowledge on shrubs and tree fodders were further highlighted by country case studies detailing prevailing situations and policy matters. A special session was held to discuss the successful development and results achieved in the three-strata forage system in Indonesia. The workshop concluded with important working group discussions on the priorities for further research and development, and on the potential for the wider use of shrubs and tree fodders in the developing world
. Resume Cette publication pr6sente les r6sultats dune rencontre internationale tenue i Denpasar, Bali, Indonesie, du 24 au 29 juillet 1989 et qui a porte sur 1'utilisation des arbustes et fourrages veg6taux par les animaux d'61evage. Les 26 communications qui y ont ete presentees traitaient de la disponibilit6 des ressources alimentaires pour les animaux, de leur utilisation par les ruminants et les non-ruminants, des m6thodes de transformation, des aspects economiques et des questions du developpement. Ces sujets et les connaissances dsraacqe'recauebcstdsehuurtsseseetyttrlieelocsdersthn'mse6esemtistulenuieddtl6rsieedelpssdseeaesdfardoeuap6duxrrcv'rboiarempudaslosugoptdlecpeioetstspriidqtoeveaimunetnvegldetese6eern.sssttaUffppuodonaouxiuseyurrdcrsrrsuaaaee1gpsgaxse'seanispsovcloeneevvssnseeaspiggpndr6a66teeyttcstlaaseiusugasuexxlrrencoioeuladrennpc6sptoetvropsneeoronslditosstseeaspusnisptibturctreaeiaremlvtis6letaae6tpsi6nmlsate6rs.dnittusi'uucrIednutleiidelelnisiosbsnpprapeerltlaiisusoocisderneie.ate6eLcf1tsoh'aladenartgsecdeiulerdi6neadsrepeutssltlaleetss Resumen enpeugcyassDlnieirosrttnviiapecdumovloennieieaEirelzojcpesicerarsadtaraiusrtslocooarsaaa.m.srinndaE6oplEdreeyianlul,asissnlsunbBetaodpaeelb2mtlaleeioxxyr6lcclspniaieiofassr,,tyrlioetcrntibmeaIearascioinnrodnbuoleucaditnotalntdeitjetoosjmieacsasolpndsdlidasiyeruzeoeoeienasymlpdlnscsoiacoreaeuteebgimeo,nleoxrcfosedstnoqcaieiarttcessaunnaioudlomledtsdpoiraen2oomsesopdsea4sosreoopsprrseaaebsrtaoddiolssulpareeauetcctl2sanulaaieltpl9rmetdnseanstaoeiiastzddeaiorlemmnaeincyosstddvtjiitsieuucnaddaoeedarraalesinsaibiseslrttooremic.iocobgosnuualSudeae,nssyyIensescoianltcooitlscod6r1eeausneesnop9neuselupno8yesosetyant9iebrlbdilriise,roiprtormezioilyebnanacauldioeillucoscpiaonesleinnsoeassodsacdtnsangnoefueretosrrpereandeuenerolaslnrsepcalrlsaicoolcoaa6ildjoeissnoesbneonatsosyodeslnqtemirrammeduamss6drpeolefritaaearseobcsrasanmduceolryerblteeilnboaoaaaeddu.rltfejsbteeeooeslraptnsptmraapoaasrcaoldooarrslainraaartojbsieoeyrrnmredonutleidalnincmraeosellbilena.caeiostsuaElslrlanetqseidsttruseoeess iii
Session I: The Resources
.................................. The diversity and potential value of shrubs and tree fodders
............................ Shrubs and tree fodders in farming systems in Asia
Major characteristics, agronomic features, and nutritional value
shrubs and D.A. Ivory Discussion
Session II: Use by Farm Animals
............................... The useof shrubs and tree fodders by ruminants
......................... The use of shrubs and tree fodders by nonruminants
.................... ... J Toxic factors and problems: methods ofalleviating them inanimals
Session III: The Three-Strata Forage System
..... The concept and development of the three-strata forage system I.M. Nitis, K. Lana, W. Sukanten, M. Suarna, and S. Putra
Research protocols appropriate to the development of methodology for
..... the three-strata forage system
K. Lana, I.M. Nitis, M. Suarna, S. Putra, and W. Sukanten
................................ Socioeconomic aspects of the three-strata forage system in Bali
Communication aspects and research extension linkages of the tNDhi.rsKece.u-:ssNtsruiaortnaainf.oir.ag..e..s..y.s.t.e.m...i.n..B..a.l.i........................................1.3. 1036
Session IV: Country Case Studies
.............................. Availability and use of fodder shrubs and trees in tropical Africa
.. ... Potential of legume tree fodders as animal feed in Central America
D. Pezo, M. Kass, J. Benavides, F. Romero, and C. Chaves
......... Availability and use of shrubs and tree fodders in Pakistan M. Akram, S.H. Hanjra, M.A. Qazi, and J.A. Bhatti
.................................. Agrosilvipasture systems in India
.............................. Availability and use of shrubs and tree fodders in India
........................ Availability and use of shrubs and tree fodders in Nepal
N.P. Joshi and S.B. Singh
............................... Availability and use of shrubs and tree fodders in Bangladesh
............................. Availability and use of shrubs and tree fodders in Sri Lanka
............................... Availability and use ofshrubs and tree fodders in Thailand M. Wanapat
................................ Availability and use of shrubs and tree fodders in Malaysia
Wong C.C .
.............. Availability and use of shrubs and tree fodders in Indonesia
M. Rangkuti, M.E. Siregar, and A. Roesyat
........................... .... Availability and use of shrubs and tree fodders in the Philippines L.T. Trung
Availability and use of shrubs and tree fodders in China
................................ Xu Zaichun
Session V: Processing, Methodology, and Economics
................................ Opportunities forprocessing and using shrubs and tree fodders M.R. Reddy
Development and evaluation of agroforestry systems for fodder
.................... production A.N. Abd. Ghani and K. Awang
................................. ..................................331 Economic aspects ofusing shrubs and tree fodders to feed farm animals P.Amir
..................... Conclusions and Recommendations
Potential of legume tree fodders as animal feed in Central America D. Pezo, M. Kass, J. Benavides, F. Romero, and C. Chaves Tropical Agricultural Research and Training Centre, Turrialba, Costa Rica Abstract -Agroforestry systems are promising alternatives for sustainable agricultural production in the subhumid and humid tropics ofCentral America. Intercropping Eryduina poeppigiana with coffee, African star grass (Cynodon nlemfuensis), or king grass (Pennisetum purpureum x Pennisetum typhoides); livefences ofE. poeppigiana or Gliricidia sepium; and monocultures ofthe same two legume tree species (managed as protein banks) are systems analyzed in this paper, in terms of theirforage biomass production potential, herbage quality, and nutrient cycling. Data generated in the Central American
isthmus on the potential of legume treefoliages (mainly E. poeppigiana and G. sepium) as protein supplements for ruminants are also presented. Some ofthe aspects covered are chemical composition, in vitro dry matter digestibility, rumen degradability, intake, live weight gain, and milk production responses in goats, sheep, beef, and dairy cattle. Also, some information on the conservation ofthe foliage of G. sepium as silage is included. Resume -Les systemes d'agroforesterie sort prometteurs de solutions de rechange allant dans le sens de 1'agriculture durable pour les tropiques sous-humides et humides de 1'Amerique centrale. Les auteurs analysent la culture intercalaire de Erythrina poeppigiana et du cafe, de Cynodon nlemfuensis ou de Pennisetum purpureum x Pennisetum typhoides; la plantation de haies vives de E. poeppigiana ou de Gliricidia sepium; et la monoculture des deux memes especes d'arbres legumineux (geres comme banques de proteins) enfonction de leur potentiel deproduction de biomasse fourragere, de la qualite des herbages et du cyclage des nutrients. Its presentent les donnees obtenues dans 1'isthme de 1'Amerique centrale sur le potentiel des feuilles d'arbres leguminux (principalement E. poeppigiana et G. sepium) comme complement proteique a la ration alimentaire des ruminants. Its abordent notamment la composition chimique, la digestibility in vitro des matieres s&hes, la degradability du rumen, !'ingestion, le gain depoids et la production de lait chez les chevres, les moutons, le boeufet la vache laitiere. tgalement, ils donnent des information sur la conservation des feuilles de G. sepiwn pour 1'enilage. Resumen -Los sistemas de agroforestaci6n, en los tr6picos hdmedos y semihumedos de America Central, son alternativas promisorias para mantenr la produccion agricola. El cultivo inlercalado o intermedio de Erythrina poeppigiana con cafe, grama africana (Cynodon nlemfuensis) o grama rey, (Pennisetum purpureum x Pennisetum typhoides), setos vivos de E. poeppigiana o Gliricidia sepium y monocultivos de las mismas dos especies de arboles leguminosos (explotados comp banos proteinicos) son sistemas que se analizan en este estudio, en funci6n de su potenial de producci6n de biomasa forrajera, calidad de pastizales y ciclo de nutrientes. Se presentan los datos obtenidos en el Istmo de America Central con relaci6n al potenial de 163
follajes de drboles leguminosos (principalmente E. poeppigiana y G. sepium) comp aditivos proteinicos para rumiantes. Entre los aspectos considerados se encuentran: composici6n quimica, digestibilidad de la materia seca in vitro, la capacidad de separaci6n de alimentos del rumen, consumo, aumento enpeso vivo y las respuestas de produccOn lechera en cabras, ovejas, ganado vacuno y lechero. Se incluye tambien informaci6n relativa a la conservaci6n del follaje de G. sepium comp ensilaje. Introduction Agroforestry systems are a promising approach to developing sustainable, economical agricultural systems in the subhumid and humid tropics in Central America. Within this context, multipurpose nitrogen-fixing trees play an important role in smallholder farming systems. They provide fuelwood, live fences, shade for crops and animals, plant support, green manure, fodders, and medicinal and other products. Since the late 1970s, the Tropical Agricultural Research and Training Centre (CATIE) has characterized the agroforestry systems used in the Central American isthmus and the Dominican Republic
and developed more productive and sustainable systems using the agroforestry approach. CATIE has been studying several agroforestry alternatives using multipurpose nitrogen-fixing trees: intercropping trees in cocoa and coffee plantations as a source of shade and for recycling nutrients; alley cropping legume trees and annual crops; use of trees as support for trailing crops (e.g., black pepper and chayote squash); and as a source of fuelwood and posts. This paper emphasizes the potential use of Erythrina spp. and Gliricidia sepium as fodder sources for ruminants. Forage biomass production Forage biomass yield has been evaluated for five agroforestry systems that include nitrogen-fixing trees: intercropping of Erythrina poeppigiana with coffee, African star grass (Cynodon nlemfuensis), or king grass (Pennisetum purpureum x Pennisetum typhoides); live fences of E. poeppigiana or G. sepium; and monocultures of E. poeppigiana or G. sepium planted as "protein banks." The use of E. poeppigiana as a shade tree in cocoa and coffee plantations is a traditional agroforestry practice in many areas of the tropics (Budowski 1959; Cadima and Alvim 1967; Willey 1975). Several effects have been mentioned: soil and crop temperatures decrease; radiation is reduced, affecting photosynthesis, stomata opening, flowering, and fruit ripening; transpiration is diminished; nutrients are recycled into the soil; and nitrogen fixation occurs when legume trees are used. In this system, the biomass yields and the amount of nutrients recycled are a function of the pruning interval used for the shade tree. They also depend on the spacing of planted shade trees
and the age of the plantation. Total biomass yield increases as the pruning frequency decreases. The reverse occurs for leaf yields (Table 1), indicating that the proportion of edible biomass is greater with more frequent prunings (Russo 1984). A similar trend to that mentioned for total biomass occurs for the amount of nutrients recycled through pruned and fallen leaves of E. poeppigiana. The average 164
Table 1. Annual biomass production (kg/ha) from E. poeppigiana pruned at different frequencies in a Costa Rican coffee plantation.
Number of prunings per year
Source: Russo (1984). a In plantation, 280 trees/ha, 6 x 6 m spacing.
values for annual recycled nitrogen, phosphorus, potassium, calcium, and magnesium are 257, 22, 142, 180, and 55 kg/ha, respectively (Russo 1984). If all the edible biomass of E. poeppigiana is extracted for animal feeding, then the most easily mineralizable fractions are removed from the field. The sustainability of these systems is, therefore endangered, unless organic or chemical fertilizers are applied. The second agroforestry system studied was the intercropping of E. poeppigiana with African star grass. This was compared with intercropping African star grass with a timber tree (Cordia alliodora) and with a control (no trees). Annual grass dry matter yield was higher when associated with the legume tree than with the timber tree (9 311 vs 4 087 kg/ha). The lowest annual yield (2 632 kg/ha) was obtained in the control plots. Also, there were broad-leaved weeds in those plots with trees. Even though the presence of trees negatively affected solar radiation
reaching the grass canopy (60% of that measured in the unshaded plots), the nutrients recycled through fallen leaves and pruned branches, and nitrogen fixation occurring in the legume tree plots, favoured grass growth and prevented the invasion of weeds. A significant increase in the grass crude protein
(CP) content was also observed in the E. poeppigiana plots compared with those plots with C. alliodora and without trees (9.5, 6.4, and 6.1% CP, respectively). This suggests that the legume tree effectively fixed and transferred nitrogen to the grass (Bronstein 1984). The third agroforestry system evaluated was the intercropping of king grass (P, purpureum x P. ryphoides) and E. poeppigiana under a cut-and-carry approach (Rodriguez 1985). In this system, the annual grass yield was not affected by the presence of trees (17.7 and 16.8 t dry matter (DM)/ha for plots with 1 666 and 3 333 trees/ha, respectively), indicating that the nutrients transferred by the legume tree compensated for the lower radiation that reached the grass canopy. Other beneficial effects of this system were the extra amount of a nitrogen-rich foliage produced annually by the legume tree (7.6 and 11.6 t DM/ha for plots with 1 666 and 3 333 trees/ha, respectively) and the slight increase observed in the CP content of the grass growing under the trees. The productivity of this system declined with time, however, because of the high removal of nutrients through the grass and legume tree biomass, which, in general, was greater in the intercropped than in the monoculture grass plots. The average annual extractions of nitrogen, phosphorus, and potassium in the intercropped treatments were 444, 71, and 396 kg/ha, respectively; the corresponding values for the control plots were 151, 59, and 345 kg/ha. This experiment has now been modified to assess the sustainability of the system, when different proportions of the biomass obtained after pruning the legume tree are returned as green manure.
The fourth agroforestry system considered was the use of legume trees (G. sepium or Erythrina berteroana) as live fences, a common practice in the humid and subhumid tropics of Central America. Different pruning intervals have been evaluated at four sites in the lowland humid tropics of Costa Rica (CATIE 1987). Several factors are known to affect biomass production in this system (Fig. 1), including age of live fence, distance between trees, and pruning frequency. In general, the total biomass yield increased as the pruning interval was delayed (Table 2), but the proportion of edible biomass declined with age. Also, tree survival was negatively affected by frequent pruning. To obtain a sustained productivity of such systems, the pruning interval has to be at least 4 months (CATIE 1987). Because of the limited yield that can be obtained from the live fence agroforestry system, a new strategy has been tested. Stakes of G. sepium and E. berteroana have been planted similarly to sugarcane, to establish legume tree "protein banks." The average annual yields obtained during the 1 st year were 19.4 and 27.0 DM/ha for G. sepium and E. berteroana, respectively (CATIE, unpublished data). Whether or not these high yields can be sustained needs to be investigated. Fig. 1. Measuring foliage production from a live fence of Erythrina berieroana in the lowland humid tropics of Costa Rica. 166
Table 2. Annual biomass production (kg dry matter/km) from live fences of E. berteroana and G. sepium pruned at different frequencies in the lowland humid tropics of Costa Rica.
Pruning intervals (months) 2 4 6
Edible biomass E. berteroana 1058-2168 1769-3976 1435-4218
Total biomass 1058-2168 3132-6201 3189-8273
Note: Results are ranges of values obtained at four research sites. Source: CATIE (1987).
139-1244 1581-7771 589-7483
The edible biomass of legume trees is a useful protein supplement for ruminants.
Efforts have been made to characterize the nitrogen fraction of these feedstuffs
(Espinoza 1984). The CP content varies with plant portions (laminae green stem
petiole); however, the rumen degradability of the nitrogenous fraction does not
follow the same pattern (green stem petiole laminae). In general, these forages
have shown higher DM rumen degradability values when in situ digestion
techniques have been used (Table 3), compared with the traditional in vitro dry matter digestibility (IVDMD) method. This suggests that some antiquality factors
are present, especially in the more mature laminae (Table 4). Also, the solubility of
the nitrogenous fraction is particularly high in the green stem, but lower in the
petiole and laminae. In all represented by nonprotein
three portions, more than 75% of the soluble fraction nitrogen (NPN) compounds (Espinoza 1984).
Most of our work on the quality of legume tree foliages has concentrated on two
species: E. poeppigiana and G. sepium. More recently, however, the nutritional
diversity of the genus Erythrina has been studied (CATIE, unpublished data). The results obtained show important variations among and within species in terms of
CP content, protein solubility, acid-detergent insoluble nitrogen (ADIN) content, and IVDMD (Table 5). These results showed that the Erythrina species most
commonly included in the agroforestry systems previously evaluated (E. poeppigiana and E. berteroana) had a higher nutritional quality than the less
E. fusca value of
and E. costarricense, E. berteroana. These
notwithstanding the variability in the results suggest the need for a better
definition of the germ plasm used in different experiments, as well for genetic improvement of quality parameters in these species.
Another important aspect that needed clarification with respect to these gn(1oo9ant8tsr6,a)wdsiahtisoownasaeldhfiotghrhaagtaesthstehwaDatMsobthtianeitneaexkdteeonfoftroEtf.hptehoheeeparcpbciagecpieatonaubasi,lialtesygtubhmye easnoLilmeabfalolasrb.agBpeuernspoauuvrriecdueessfor (f3o.l2ia4geasn,din3.t3e9rm%sboofdtyhewirepigohtte,nrteisapl eicnttiavkeel,y)h.avSeombeeendidffeetreecntecdesbaymAorngguelellgoumeteal.tree
Table 3. Nitrogen fractions in different portions of the edible biomass of G. sepium and E. poeppigiana harvested at 3- and 5-month intervals.
Crude protein (CP) (%)
N rumen degradability (%)
CP solubility (%) Laminae Petiole Green stem
Nonprotein nitrogen (% soluble CP)
ADIN (% total N)8 Laminae Petiole Green stem
Source: Espinoza (1984). a ADIN, acid-detergent-insoluble nitrogen.
Table 4. Nutritive value of the edible biomass in E. poeppigiana, as a function of position in the branches.
Stratum (cm from branch apex)
0-40 40-80 80-120 120-160 160-200
31.9a 29.8b 28.8bc 26.5d 28.3c
12.4a 9.2b 8.6b 8.4b 9.2b
60.Oa 49.7bc 50.Obc 48.1c 52.8b
72.Oa 66.1b 67.Ob 63.3b 67.4b
Note: CP, crude protein; IVDMD, in vitro dry matter digestibility. Means followed by the same letter in the same column do not differ statistically (P <- 0.05).
Source: Russo (1984).
(1986) and Vargas and Elvira both studies, the foliage of G. however, several experiments have no palatability problems
(1987), using goat kids and dual-purpose cattle. In sepium was the most readily consumed (Table 6); performed at CATIE have indicated that ruminants with the commonly used species of Erythrina,
whereas problems of Studies to clarify the
have been detected in of this problem in G.
some studies with G. sepium. sepium are needed. There is some
evidence of genetic variability in the content of antiquality factors in this species;
but some seasonal and age of regrowth effects may also exist.
Table 5. Nutritive value of the edible biomass in different species and clones of the genus Erythrina.
No. of clones
Crude M protein
Protein solubility Ma
ADIN (fit b
E. fusca E. poeppigiana
(19-25) (20-49) (10-18)
E. cocleata E.lanceolata
b Solubility in a borate-phosphate buffer.
% ADIN, acid-detergent-insoluble nitrogen, of total nitrogen. C 1VDMD, in vitro dry matter digestibility.
(8-19) 20 14
Values in parentheses are the ranges of mean values for clones within species.
IVDMD Mc 49 (42-55) 44 (43-47) 52 (50-54) 46 (32-53) 47 50
Table 6. Chemical composition, in vitro dry matter digestibility (IVDMD), and voluntary intake of Leucaena leucocephala, Glirieidia sepium, and Guazuma ulmifolia.
Leucaena Gliricidia Guazuma
CP (%) CWC (%) ADF (%) ADIN (% of total N) IVDMD (%) DM intake' (% body weight)
25.0 47.8 28.2 7.4 47.8 0.512c
25.8 43.5 26.2 10.7 58.4 0.868a
14.7 49.5 31.4 25.2 43.0 0.709b
Note: CP, crude protein; CWC, cell wall constituents; ADF, acid-detergent fibre; ADIN, aciddetergent-insoluble nitrogen; DM, dry matter. Means followed by the same letter do not differ statistically (P 5 0.05). 'Source: Vargas and Elvira (1987). Ad libitum intake obtained in 2 h after the morning milking.
The need for protein supplementation in grazing animals is enhanced during the dry season in subhumid ecosystems. However, in many legume trees, the protein-rich leaves are unavailable for animal feeding during this part of the year because leaves are shed during the early dry season, when plants enter a flowering stage (Russo 1984). One of the strategies considered to overcome this restriction is conserving these foliages as silage. This task is not easy, however, because, in many cases, the fermentation process in legumes is dominated by Clostridia, leading to a butyrate-type silage. This has been attributed to three factors in the legumes: high buffering capacity, low water-soluble carbohydrates, and, often, low DM content. Based on these facts, different preconditioning treatments to reduce nutrient losses during silage fermentation have been tested. Kass and Rodriguez (1987) observed that a high proportion of the nitrogen fraction was in a volatile form (ammonia) and low levels of lactic acid were
detected in silages made of either fresh or wilted G. sepium foliage, without additives. The addition of readily fermentable carbohydrate sources, such as molasses or chopped sugarcane, decreased ammonia and increased lactic acid concentrations. Also, the addition of formic acid markedly reduced ammonia nitrogen, but did not affect lactic acid production. Based on these preliminary results, different levels of sugarcane molasses as an additive for G. sepium silages were evaluated (CATIE, unpublished data). The CP content tended to decline slightly (dilution effect), but IVDMD increased as higher levels of molasses were added. A decline in ammonia concentration occurred when 2% molasses (w/w) was added to ensiled G. sepium, but higher levels of molasses resulted in minor changes in ammonia concentration. Lactic acid concentration increased almost linearly with molasses additions, but the highest level of molasses (10%, w/w) resulted in a silage with only 2.6% lactic acid. In all cases, the butyric acid concentration was very low (0.04%). The voluntary intake of these silages has not been measured, but observations with young female goats showed that silages made of G. sepium with 2% molasses as an additive were readily consumed when fed as a supplement to chopped grass. Animal production responses Small ruminants Initial work has been done at CATIE, using goats and sheep, aimed at developing ruminant feeding systems based on the use of legume tree foliages as protein supplements. In most of these studies, the basal diet was chopped king grass (P. purpureum x P. typhoides); this has been supplemented with different energy sources, especially green bananas. Gutidrrez (1983) compared E. poeppigiana foliage plus green bananas with a commercial concentrate as supplements for dairy goats receiving chopped king grass as the basal forage source. Daily milk yield was greater with goats fed the commercial concentrate than with those fed E. poeppigiana and bananas (1.29 vs 1.08 kg/head); however, the former showed lower daily gains (9 vs 59 g/head), and the Economic Analysis
favoured the use of E. poeppigiana plus green bananas. These findings encouraged further research in this field. Samur (1984) examined whether the use of a starch source (green bananas) would result in a milk production higher than that obtained with a sugar-rich source (ripened bananas). The rationale was not only that starch promotes microbial protein synthesis in the rumen when readily degradable protein sources are used but also that some starch escapes ruminal degradation, therefore increasing glucose availability for milk synthesis and other purposes. The results supported this hypothesis: slightly higher daily milk yields were obtained when green rather than ripened bananas were used as the energy supplement (1.28 vs 1.23 kg/head, respectively). However, this effect was more evident in two other experiments done with weaned lambs (Benavides and Pezo 1986) and goat kids (Benavides and Esnaola 1986). In both studies, the advantages of using either starch- or a mixture of sugar- and starch-rich sources as energy supplements to diets based on legume tree foliage were demonstrated. 170
Table 7. Milk production and dry matter (DM) intake in goats supplemented with two levels of E. poeppigiana and plantain fruits.
Daily milk yield (g/head) Milk fat (%) Daily DM intake (kg/head) Chopped king grass E. poeppigiana Plantain fruits Daily LW gain (g/head)
1270a 3.4 0.77 0.45 0.60 170
1090b 3.3 0.93 0.32 0.60 50
1090b 3.3 0.96 0.45 0.37 20
1 110ab 3.1 0.97 0.32 0.37 50
Note: H, high; L, low; P, plantain fruits; E, E. poeppigiana; LW, live weight. Means followed by the same letter do not differ statistically (P < 0.05). Source: CATIE (unpublished data).
Esnaola and Rios (1986) offered different levels of supplementary E. poeppigiana foliage to goats. Milk yield increased linearly as a response to the level E. poeppigiana used. Also, a partial substitutive effect of the supplement on the intake of chopped king grass was detected, resulting in an additive effect on total DM intake. A recent factorial experiment (CATIE, unpublished data), in which two levels of both E. poeppigiana foliage and plantain fruits were evaluated as supplements to a chopped king grass basal diet for milking goats, showed that the highest milk yield corresponded to the combination of the highest levels of both supplements (Table 7). However, the interesting aspect of this study is that the second highest milk production was obtained for the combination of the lowest levels of both supplementary sources. As both treatments had about the same CP/DE (digestible energy) ratio, perhaps a given ratio of both nutrients should be maintained for efficient milk production. Most of the small ruminant feeding experiments carried out at CATIE using legume tree foliages have used E. poeppigiana. There is one direct comparison of this species with G. sepium (Rodriguez et al. 1987): higher milk yields were obtained when goats were fed E. poeppigiana. Both tree foliages showed similar values for chemical composition, but the lower DM intake observed in the case of G. sepium negatively affected daily milk production (1.26 and 1.11 kg/head for E. poeppigiana and G. sepium, respectively). It is clear that there is a need for a more detailed study on the causes of limited intake in some G. sepium foliages, because previous work from other regions (e.g., Chadhokar 1983) has not detected this type of problem. Cattle Few studies have been done at CATIE evaluating the use of legume tree foliages as protein supplements for cattle; again, E. poeppigiana has been the main species evaluated. The first experiment was done with weaned dairy heifers, evaluating different levels of substitution of a traditional protein source (soybean meal) with E. poeppigiana foliage (Pineda 1986). The replacement of soybean with increased levels of E. poeppigiana resulted in lower daily gains (Table 8), confirming that the protein quality of this tree foliage is less than that of soybean meal. From an
economic viewpoint, however, the best results were obtained when 67% of the protein requirements were provided by E. poeppigiana. Vargas (1987) fed growing bulls different levels of E. cocleata foliage and observed increased live weight gain when E. cocleata was offered at a level equivalent to 0.5% body weight (398 and 524 g/day, for the control and 0.5% body weight for E. cocleata). Also, the addition of an energy supplement (green banana fruits) to the same level of E. cocleata resulted in a slight increase in daily gain (579 vs 524 g/head). The use of E. poeppigiana and sugarcane molasses as supplements for milking cows grazing African star grass, instead of commercial concentrates (Abarca 1989), have consistently resulted in lower milk yields (8.4 and 8.0 kg/day, for concentrate and E. poeppigiana supplementation, respectively). Minor changes in milk composition were also observed. However, when E. poeppigiana was compared with fish meal as a protein supplement for milking cows, higher milk yields were obtained with the fish meal (Table 9), confirming that this legume tree foliage has a lower protein quality. Economic analyses of our dairy cattle experiments have indicated that there are no differences in the net profit obtained using the traditional (fish meal or commercial concentrates) or nontraditional supplements (legume tree
Table 8. Intake, in vivo digestibility, and daily live weight gain (LWG) in heifer calves fed different proportions of soybean meal (SMB) and E. poeppigiana (P) foliage as a protein supplement.
67% SBM, 33% SBM,
100% SBM 33% P
Intake (% body weight) Soybean meal E. poeppigiana King grass Sugarcane molasses Digestibility (%)a DM CWC LWG (g/head)
0.46 0.0 2.19 0.86 62 62 410
0.31 0.26 2.04 0.97 60 55 366
0.15 0.52 1.93 1.01 60 55 372
0.0 0.79 1.80 1.15 58 53 294
Source: Pineda (1986). a DM, dry matter; CWC, cell wall constituents.
Table 9. Dry matter (DM) intake and daily milk production in cows grazing African star grass, supplemented with either fish meal or E. poeppigiana foliage as the protein source.
Fish meal E. poeppigiana
DM intake (% body weight) African star grass Supplement Total Milk production (kg/head)
1.93a 1.08a 3.Ola 9.0
1.24b 1.55a 2.79a 8.2
Milk compostion (%) Total solids Protein Fat
13.4b 3.2a 4.1a
12.7a 3.3a 4.3a
Note: Sugarcane molasses was the energy supplement. Means in the same row followed by the same letter do not differ significantly (P <_ 0.05).
foliages plus locally available energy sources, such as green bananas). There is an important advantage favouring nontraditional supplements: i.e., lower cash costs. This is important for the small farmer because the availability of cash is one of the farmer's main constraints. Another experiment, carried out by Tob6n (1988), showed a linear increase in milk production as the level of E. poeppigiana foliage increased. The equation describing this linear relationship was Y = 8.75 + 1.29X, where Y is daily milk yield (kilograms per cow) and X is DM intake (percent body weight). In summary, the results obtained with grazing dairy cows under lowland humid tropical conditions suggest that the direct benefits on milk production derived from the use of E. poeppigiana foliage are small, but that there might be important indirect benefits. These benefits include increased carrying capacity as a result of substitutive effects on grass intake and increased recycling of nutrients in the areas grazed, through richer animal excretas. These aspects need to be investigated in future work. References Abarca, S. 1989. Efecto de la suplementaci6n con porn (Erythrina poeppigiana) y melaza sobre la producci6n de leche en vacas pastoreando estrella africana (Cynodon nlemfuensis). Tropical Agricultural Research and Training Centre, Turrialba, Costa Rica. 68 p. Arguello, R., Benavides, J., Esnaola, M.A. 1986. Evaluaci6n de las ganancias de peso y consumo de alimentos en cabritos alimentados con distintos follajes de irboles, suplementados con banano verde de desecho. In Resumen de las investigaciones realizadas con ruminantes menores en el proyecto de sistemas de producci6n animal. Tropical Agricultural Research and Training Centre, Turrialba, Costa Rica. Informe Tecnico No. 67, 28-32. Benavides, J. 1986. Utilizaci6n de follaje de por6 (Erythrina poeppigiana) para alimentar cabras bajo condiciones de tr6pico h6medo. In Memorias II Congreso Asociaci6n Mexicana de Zootecnistas y Tecnicos en Caprinocultura. Azteca, Mazatlin, Mexico. pp. 192-210. Benavides, J., Esnaola, M.A. 1986. Efecto de la suplementaci6n con distintas fuentes y niveles de energia a cabritos en crecimiento consumiendo una diets base de follaje de por6 (E. poeppigiana). In Resumen de las investigaciones realizadas con ruminantes menores en el proyecto de sistemas de producci6n animal. Tropical Agricultural Research and Training Centre, Turrialba, Costa Rica. Informe Tecnico No. 67, 23-28. Benavides, J., Pezo, D. 1986. Evaluaci6n del crecimiento y consumo de materia seca en corderos alimentados con follaje de por6 (E. poeppigiana) ad lib, suplementados con distintas fuentes de energia. In Resumen de las investigaciones realizadas con ruminantes menores en el proyecto de sistemas de producci6n animal. Tropical Agricultural Research and Training Centre, Turrialba, Costa Rica. Informe Tecnico No. 67, 43-47. Bronstein, G.E. 1984. Producci6n comparada de una pastura de Cynodon nlemfuensis asociada con irboles de Cordia alliodora, con irboles de E. poeppigiana y sin irboles. Magister Sciendae. Universidad de Costa Rica, Turrialba, Costa Rica. Thesis. 110 p. Budowski, G. 1959. Pricticas forestales de interes para el cultivo del cafe. Cafe, 1, 49-52. 173
Cadima, A., Alvim, P. de T. 1967. Influencia del arbol de sombra Erythrina glauca sobre algunos factores edafologicos relacionados con la production del cacaotero. Turrialba, 17,330-336. CATIE (Tropical Agricultural Research and Training Centre). 1987. Sistemas silvopastoriles para el trdpico htimedo bajo. Segundo Informe Anual. CATIE, Turrialba, Costa Rica. 123 p. Chadhokar, S. 1983. The effect of gliricidia-supplemented dry season foliage on milk yield and composition on MRY (Netherlands) cows in Sri Lanka. Tropical Grasslands, 17, 39-41. Esnaola, M.A., Rios, C. 1986. Hojas de porn (Erythrina poeppigiana) como suplemento proteico para cabras lactantes. In Resumen de las investigaciones realizadas con ruminantes menores en el proyecto de sistemas de production animal. Tropical Agricultural Research and Training Centre, Turrialba, Costa Rica. Inform Tecnico No. 67, 60-69. Espinoza, J. 1984. Producci6n y caracterizaci6n nutritiva de la fracci6n nitrogenada del forraje de madero negro (G. sepium) y porn (E. poeppigiana) a dos edades de rebrote. Universidad de Costa Rica, Turrialba, Costa Rica. Thesis. 95 p. Gutierrez, R.A. 1983. Follaje de porn (E. poeppigiana) y banano maduro de desecho (Musa sp. cv. Cavendish) como suplementos para cabras lecheras estabuladas. Informe problema especial. Universidad de Costa Rica, Turrialba, Costa Rica. 27 p. Kass, M.L., Rodriguez, G. 1987. Preliminary studies on silage making from G. sepium. In Withington, D., Glover, N., Brewbaker, J.L., ed., Gliricidia sepium (Jacq.) Walp.: Management and improvement. Proceedings of a workshop held at CATIE, Turrialba, Costa Rica, June 21-27,1987. Nitrogen-Fixing Tree Association, Honolulu, HI, USA. pp. 201-204. Pineda, O. 1986. Utilizacion del follaje de por6 (E. poeppigiana) en la alimentacion de temeras de lecheria. Universidad de Costa Rica, Turrialba, Costa Rica. Thesis. 65 p. Rodriguez, Z., Benavides, J.E., Chaves, C., Sanchez, G. 1987. Milk production in goats fed G. sepium and E. poeppigiana foliages, supplemented with plantain (Musa sp. cv. Pelipita) fruits. In Withington, D., Glover, N., Brewbaker, J.L., ed., Gliridicia sepium (Jacq.) Walp.: Management and improvement. Proceedings of a workshop held at CATIE, Turrialba, Costa Rica, June 21-27, 1987. Nitrogen-Fixing Tree Association, Honolulu, HI, USA. pp. 212-216. Russo, R.O. 1984. Studies on Erythrina poeppigiana (Walpers) O.F. Cook, a versatile tree in Costa Rican farms. Southeastern University, New Orleans, LA, USA. PhD thesis. 145 p. Samur, C. 1984. Production de leche en cabras alimentadas con king grass (P. purpureum x P. typhoides) y porn (E. poeppigiana), suplementadas con fruto de banano (Musa sp. cv. Cavendish). Universidad de Costa Rica, Turrialba, Costa Rica. Thesis. 51 p. Tobon, C.J. 1988. Efecto de la suplementacidn con cuatro niveles de follaje de por6 (Erythrina poeppigiana) sobre la production de leche en vacas en pastoreo. Tropical Agricultural Research and Training Centre, Turrialba, Costa Rica. 72 p. Vargas, A. 1987. Evaluacion del forraje de por6 (Erythrina cocleata) como suplemento proteico para toretes en pastoreo. Universidad de Costa Rica, Turrialba, Costa Rica. Thesis. 88 p. Vargas, H., Elvira, P. 1987. Chemical composition, digestibility and intake of Leucaena leucocephala, Gliricidia sepium, and Guazuma ulmifolia foliages. In Withington, D., Glover, N., Brewbaker, J.L., ed., Gliricidia sepium (Jacq.) Walp.: Management and 174
improvement. Proceedings of a workshop held at CATIE, Turrialba, Costa Rica, June 21-27,1987. Nitrogen-Fixing Tree Association, Honolulu, HI, USA. pp. 217-222. Willey, R.W. 1975. The use of shade in coffee, cocoa and tea. Horticultural Abstracts, 45, 791-798. 175