Tải bản đầy đủ - 0trang
III. Agronomic Performance of Tree Legumes
GRAEME BLAIR ET A L .
(Partridge and Ranacou, 1973), 11.5 t ha-' in Japan (Kitamura, 1986),
12.9 t ha-' in Australia (Ferraris, 1979), 14.3 t ha-' in Hawaii (Evensen,
1984), 14.2 t ha-' in Thailand (Topark-Ngarm, 1983), 8.4 t ha-' in the first
year of growth in Indonesia (Ella et al., 1989), and 15.6 t ha-' ha-' over a
2.5-year period in Indonesia (Catchpoole and Blair, 1990a).
There are very few reports of forage yields from other species. Holm
(1972) obtained 7.2 t DM ha-' year-' from Sesbania grandijlora in
Thailand. Chadhokar and Lecamwasam (1982) reported Gliricidia leaf
fresh weight yield in Sri Lanka in excess of 40 t ha-', with dry matter percentage of leaf ranging between 20% and 23%. Gutteridge and
Akkasaeng (1985) compared the growth of 15 tree species during the first 6
months after planting. Sesbania formosa and Sesbania sesban gave the
best yields, whereas Sesbania grandijlora was comparable to Leucaena
leucocephala and Calliandra calothyrsus. Evans and Rotar (1987b) also
obtained comparable fodder yields from various perennial Sesbania species, Calliandra calothyrsus, and Leucaena leucocephala, during the first
year after establishment. Little information is available on the best means
of establishment and subsequent management practices for most species
(Maasdorp and Gutteridge, 1986), although researchers with Leucaena
have addressed such topics as rhizobiology (Diatloff, 1973; Jarvis, 1981;
Bushby, 1982; Manjunath et al., 1984; Sanginga et al., 1986), response to
pH (Ahmad and Ng, 1981; Norani and Ng, 1981),effect of CaC03 (Hutton
and Andrew, 1978; Tham et al., 1977),nutrition and symptoms of nutrient
deficiency (Gonzales et al., 1980; Haag and Mitidieri, 1980),and methods
of establishment (Pound et al., 1980a,b;Pound and Santana, 1980; Falvey,
1981; Cooksley, 1981; 1982; Oakes, 1984; Olvera and Blue, 1985; Olvera
and West, 1985).
In an evaluation of 16 tree legumes conducted over a 4-year period in
South Sumatra, Indonesia, Blair et al. (1988) found that after a 2-year
establishment period, leaf yields were highest in Acacia mangium; this
declined with time such that the highest total leaf production was recorded
in Cassia siamea (Fig. 1).
For another species comparison experiment conducted in South
Sulawesi, Indonesia, Catchpoole and Blair ( 1990a) recorded leaf yields
of 18.2, 19.2, 21.8, and 6.7 t DM ha-' in C. calothyrsus, G. sepium,
L . leucocephala, and S . grandi3ora over a 14-month period from establishment. The high yield in Leucaena was recorded prior to the arrival of
psyllids in this part of Indonesia.
Forage yields per hectare are likely to be maximized if the shrub legumes
are in a mixture with an understory of herbaceous pasture species, through
increased efficiency of utilization of light and soil moisture (Gray, 1970).
In an experiment conducted in Indonesia, Catchpoole and Blair (199Oa)
FORAGE TREE LEGUMES IN TROPICAL ENVIRONMENTS
0 Feb. '82 to Ian. '84
'84 to Fcb. '85
69 Feb. '85 to Apr. '86
FIG. 1. Leaf production (g DM per tree) of the ten most productive species for three
growth periods at Nakau, South Sumatra.
Dry Matter Yields (t ha-') over a 14-Month Period of TreelGrass Mixtures and
Monocultures at Gowa, South Sulawesi, Indonesia"
Total legume DM
" From Catchpoole and Blair (]!Ma).
" Means within a row with the same letter are not significantly different ( p < .05).
Edible DM = legume leaf DM and grass DM.
GRAEME BLAIR ET AL.
found that undersowing leguminous trees (Leucaena, Calliandra, Glirici&a) with Panicum maximum cv. Riverdale increased the dry matter yield
of edible forage from 18.7 to 24.7 t ha-' over a 14-month period (Table I).
In another Indonesian experiment, Horne and Blair (1990) found that
although the quantity of light penetrating to 20 cm above the soil surface
was similar in mixtures of Leucaena cut to 100 cm with Setaria sphacelata
(HLSET) or Pennisetum purpureum (HLPEN) and Leucaena cut to 30 cm
with Setaria (LLSET) or Pennisetum (LLPEN), the shape of the light
profile was markedly different (Fig. 2) and the canopy within each stratum
had a very different grass-legume composition.
Relallve Irradlance (100 I/Io)
4 0 20
Reletlve lrradlanae (100 1/10)
Relallve lrradiance (100
Relative lrradiance (100 I/lo)
C Leaf D.M. (kg ha-ld-')
C Leaf D.M. (kg lia-'d-')
FIG. 2. Mean light profiles of the mixtures and downward cumulative leaf dry matter
production. (Z leaf) in the canopies of (a) HLSET, (b) LLSET, (c)HLPEN, and (d) LLPEN.
Horizontal bars indicate standard errors of relative irradiance means.
FORAGE TREE LEGUMES IN TROPICAL ENVIRONMENTS
The poor performance of low-cut Leucaena in both mixtures resulted
largely from competitive interference for light. Pennisefum,although slow
to recover from cutting, was able to produce a tall canopy that severely
shaded low-cut Leucaena for most of each growth period.
Examination of the mixture and monoculture leaf yields in the experiment of Home and Blair (1990) illustrates the “trade-off” that occurs
between the legume and the grass in mixtures. The data presented in Table
I1 compare the total leaf yields obtained from the mixtures over the 1-year
experimental period with the potential yields obtained when the same land
area was subdivided evenly between monocultures of the component
species (the 50 : 50 monoculture option).
Forage from all four monoculture options consisted of approximately
30% Leucaena leaf, a percentage widely considered to be optimal for
ruminant production (Pound and Martinez-Cairo, 1983). Although total
forage yields were higher from the low-cut Leucaena mixtures than from
their monoculture options, Leucaena leaf made up no more than 12% of
the total forage. The reason for this low percentage contribution, compared with the monoculture options, is the unexpectedly good performance of low-cut Leucaena in monoculture and its poor performance in
mixtures. There are many reports in the literature that low cutting has an
adverse effect on Leucaena leaf and stem yields and these have been
summarized by Home e f al. (1986).
Although successful incorporation of shrub legumes into an existing
herbaceous pasture might prove difficult (Shaw, 1965), herbaceous species
could be added to an established stand of shrub legumes by oversowing
(Jones et al., 1982). A mixture of shrub legumes and grass also helps to
provide an appropriate mixed diet for grazing ruminants (Jones and Bray,
1983) and would help reduce erosion on sloping land.
In an experiment of Catchpoole and Blair (1990a), the undersowing of
tree legumes with Panicum maximum did not increase the edible N yield of
Total Leaf Dry Matter Yields (t ba-’ year-’) from the Mivtures and Potential Yields from
the 50 :50 Monoculture Option
50 : 50 monoculture
GRAEME BLAIR ET AL.
the system (652 kg ha-' year-'). Where grass was grown as a monoculture, the edible N yield was only 88 kg ha-'.
Another approach is to fence off stands of shrub legumes and to control
stocking of the shrub legume pastures and larger areas of herbaceous
pasture in a rotation system, to enable optimum use of each pasture
resource (Shaw, 1968; Blunt and Jones, 1977; Anonymous, 1981; Paterson
et al., 1982; Foster and Blight, 1983). Judicious use of the shrub legume
component would enable high-quality feed to be accumulated for use
during dry periods (Jones and Bray, 1983). Another grazing system proposed for Leucaena (Wildin, 1980)is to grow the legume as tall trees and to
allow grazing of the lower branches and seedlings under the canopy by
stock. However, much research is needed to work out the best management system for animals grazing shrub legume pastures, such as size and
arrangement of paddocks, time and frequency of grazing, stocking rates,
tree spacing, and fertilizer requirements. The ultimate measure of pasture
productivity is animal production itself.
IV. TREE LEGUME LEAF AS ANIMAL FEED
Many plant species probably form a part of the diet of domestic animals
at some time, but Table 111 can be used as a guide to the number of
leguminous shrub and tree species used regularly as a source of feed. Often
trees and shrubs are only used for feed during drought, or other times when
herbaceous material is unavailable (Gray, 1970). A notable exception is
Leucaena leucocephala, which can be harvested year-round to provide
forage for penned animals in a cut-and-carry feed system (NAS, 1984).
Nitrogen-Fixing Tree and Shrub Legumes Used for Forage"
After Brewbaker (1986).
FORAGE TREE LEGUMES IN TROPICAL ENVIRONMENTS
When animals are maintained in low-quality roughage such as rice
straw, the inclusion of high-protein tree legume leaf can have a significant
effect on animal performance.
An obstacle to the widespread acceptance of tree legumes as forage
sources has been low quality or palatability (Jones, 1979; Minson and
Wilson, 1980; D'Mello, 1982; Brewbaker, 1986). Chemical analyses and
digestibility data of leaf of Leucaena leucocephala, Gliricidia sepium,
and Sesbania grand;Jlora have been reported by Holm (l973), Adenaye
(1979), Minson and Wilson (1980), and Ekpenyong (1986). Vercoe (1989)
has presented data on 39 Australian tree and shrub species.
Palatability and acceptance by stock are generally not a problem with
Leucaena (Bray, 1984), but problems of stock acceptance of other tree
legume species have been reported (Skerman, 1977). Gliricidia has been
used as a feed supplement for sheep (Chadhokar and Kantharaju, 1980)
and dairy cattle (Chadhokar and Lecamwasam, 1982). Carew (1983)
showed that Gfiricidiaas the sole dry season diet for dwarf sheep was able
to maintain them for a continuous period of 21 weeks, despite an initial
drop in liveweight. There is a paucity of published information on the feed
quality of other species, despite their potential as sources of high-protein
feed for ruminant animals.
Most of the literature reporting liveweight gains of animals grazing shrub
legumes documents research predominantly undertaken in Queensland,
Australia, with Leucaena. Jones and Bray (1983) have summarized research in Australia on animal production from grazed Leucuena pastures.
Results from these experiments were obtained before Jones and Lowry
(1984) showed that the mimosine toxicity problem in Australian ruminants
could be overcome by infusion of rumen fluid from Indonesian ruminants.
Jones and Bray (1983) surmised that elimination of Leucaena toxicity
would result in a marked improvement in animal production. Despite the
toxicity problem, Falvey (1976) reported that a Leucaena-Cynodon dactylon pasture in the Northern Territory of Australia resulted in higher
liveweight gains than were obtained from heifers grazing Townsville
stylo-Panicum maximum pastures. Jones and Jones (1982) reported annual liveweight gains of cattle grazing Leucaena-Setaria pastures of 3 1 1
kg ha-' compared t o 200 kg ha-' from Siratro-Setariu pastures.
In Malaysia, Wong et af. (1983) reported that liveweight gains of cattle
rotationally grazing a Leucaena-Brachiaria decumbens mixture were
comparable to liveweight gains obtainable from nitrogen-fertilized pastures. In the Philippines, Moog (1983) reported that leaf production from
Leucaena-Zmperata cylindrica pastures is three times that from native
Imperata grasslands, and liveweight gains from Leucaena-Panicum maximum pastures have been of the order of 440 kg ha-'. However, nonru-