H. M. Shelton

Adoption of tropical legume technology around the world: analysis of success

1. Examples of successful adoption of forage legumes are reported from all continents, where they delivered profi tability and often provided multipurpose benefi ts to farmers. 2. Factors vital to successful adoption were: meeting the needs of farmers; building relevant partnerships; understanding the socioeconomic context and skills of farmers; partici patory involvement with rural communities; and long-term involvement of champions. 3. Organisation of seed supply, achieving scale-up and forming partnerships to implement adoption are key features. 4. Legumes remain an important but under-exploited resource for tropical farming systems. The alternative to legumes will be greater and more costly use of N-fertilisers and purchased protein concentrates. 5. The R&D organisations will need to provide longterm support and greater investment for legume technologies to deliver benefi ts to farmers. Support will be needed for training and education programs to overcome declining availability of forage legume expertise and lack of awareness of oppor tunity for use of tropical forage legumes.

Growth, morphology and nutritive quality of shaded Stenotaphrum secundatum, Axonopus compressus and Pennisetum clandestinum

The response to shade in growth and nutritive quality of buffalo grass (Stenotaphrum secundatum) was compared with that of two other stoloniferous grasses, mat grass (Axonopus compressus) and kikuyu grass (Pennisetum clandestinum). The grasses were grown outdoors in Brisbane, Australia, in soil in pots over two growing seasons in 1985/86. Treatments in the first season (Expt 1) were shading at 0 (full sun), 42, 59 and 68 % with moderate N (two spaced applications of 50 kg/ha) and, in the second season (Expt 2), 0 and 59 % shade with high (50 kg/ha every 2 weeks) and low (5 kg/ha every 2 weeks) N. Unusually for tropical grasses, the top yield of all species was higher under shade than in full sun, except for S. secundatum at high N in Expt 2. 5. secundatum and A. compressus showed an increase in top yield up to 68 % shading whereas P. clandestinum reached a maximum at 42 % shading. Shade increased shoot: root ratio and specific leaf area in all species but had little effect on leaf: stem ratio and the proportion of dead material in the tops. Stubble yield (stem bases and stolons) was reduced under shade but to a much smaller extent than root yield. S. secundatum and A. compressus had similar morphological characteristics. Their yield was lower than that of P. clandestinum in Expt 1 (autumn growth) but was similar to that of P. clandestinum in Expt 2 (summer growth). These grasses had a higher leaf: stem ratio and lower proportion of dead material in tops than P. clandestinum. Shade generally increased herbage nutritive value through an increase in N cncentration and dry-matter digestibility. The latter response was somewhat variable and mostly small (1-3% units). Under shade, the general trend was towards a decrease in total nonstructural carbohydrates, cell-wall content and insoluble ash, and an increase in lignin content. A. compressus was higher in digestibility and soluble carbohydrates than the other grasses. S. secundatum was not clearly superior in response to shade and its nutritive quality was lower than that of A. compressus.

The scope and potential of tree legumes in agroforestry

Tree legumes play a vital role in many agroforestry systems currently in use throughout the world. Because of their multipurpose nature they can be used to provide high quality fodder for livestock, nutrient rich mulch for crops, fuelwood and timber, microenvironment amelioration, ecosystem stability, and human food.Tree legumes are increasingly being used to provide fodder for livestock, as they have a number of unique characteristics which make them attractive for both smallholder and largescale livestock enterprises. Research and development efforts have concentrated on broadening the resource base by evaluating a greater range of tree legume genera, defining optimum management strategies, and developing appropriate systems which capitalize on the advantages of these species.This paper reviews the role of tree legumes in agroforestry, especially for fodder purposes, outlines the areas of current research focus, and endeavors to highlight some gaps in our knowledge which require further research effort.

Voluntary feed intake by sheep and digestibility of shaded Stenotaphrum secundatum and Pennisetum clandestinum herbage

The effects of shade (50% of ambient light transmission) on the quality of established swards of buffalo grass (Stenotaphrum secundatum) and kikuyu grass (Pennisetum clandestinum) were examined at Wollongbar Research Station, northern New South Wales. Sarlon cloth was used to shade plots of the two species, after cutting to 8 cm, from November 1985 to March 1986. Herbage was harvested from shaded and adjacent unshaded plots for chemical analyses and for feeding as dry chaff to penned sheep at the University of Queensland research farm at Mt Cotton. Shade markedly decreased the voluntary feed intake (VI) of P. clandestinum by 30.7 % over the two harvests while VI of S. secundatum was not significantly affected. Shade had no effect on the in vitro and in vivo digestibility of P. clandestinum herbage but marginally increased that of S. secundatum. These shade effects resulted in a 14.2 % increase for S. secundatum and a 34% decrease for P. clandestinum in the intake of digestible dry matter of shaded compared with unshaded herbage. The reduced VI of the shade-grown herbage of P. clandestinum was associated with a higher proportion of stem (21 v. 8% for unshaded herbage), a lower proportion of leaf (48 v. 54%), and a lower concentration of total nonstructural carbohydrates (TNC) (0.9 v. 2.1 %). In S. secundatum, there were similar, but smaller, changes in proportion of stem and TNC percentage in shade-grown herbage but, in contrast to P. clandestinum, these potential detriments to VI were offset by an increase in proportion of leaf and a decrease in the proportion of dead material. Average moisture content of shaded herbage (76%) was higher than that of unshaded herbage (70%). Nitrogen and potassium concentrations in plant tissue were higher in the shaded herbage. The higher yielding capacity and maintenance of nutritive quality of shaded S. secundatum compared with shaded P. clandestinum indicates the potential usefulness of S. secundatum for plantation agriculture.

Prevalence of mimosine and DHP toxicity in cattle grazing Leucaena leucocephala pastures in Queensland, Australia

A postal survey of the level of awareness of leucaena toxicity and an on-farm study of the toxicity status of Queensland cattle herds grazing leucaena were conducted to investigate the prevalence of mimosine and dihydroxypyridine (DHP) toxicity in Queensland. In total, 195 of 356 graziers surveyed responded to the postal survey. Sixty-three percent had inoculated their cattle with in vitro Synergistes jonesii inoculum (produced in an anaerobic fermenter) and 30% of these had inoculated more than once. The remainder used inappropriate procedures. Many graziers (43%) had occasionally observed toxicity symptoms of hair loss and poor animal growth rates. In the on-farm study, the toxicity status of 385 animals in 44 individually managed herds on 36 properties was determined by urine analysis of mimosine and DHP concentrations. No animals were experiencing mimosine toxicity, based on low concentrations of this compound found in the urine. Using the criterion that average herd urine concentrations of DHP >100 μg/mL was indicative of subclinical toxicity, 48% of herds were exposed to subclinical toxicity due to dominant 3,4-DHP (21%) or dominant 2,3-DHP (27%) toxicity; many of these herds had been inoculated with S. jonesii and were thought to be protected. The finding that 27% of herds were excreting high concentrations of 2,3-DHP was unexpected. Statistical analysis of herd-management data revealed that the method used by graziers to inoculate their herds was significantly (P < 0.05) but weakly linked to herd protection status. It was concluded that subclinical 3,4-DHP and 2,3-DHP toxicity remains a problem in Queensland and is likely to be limiting animal production in a significant number of cattle grazing leucaena-grass pastures.

A grazier survey of the long-term productivity of leucaena (Leucaena leucocephala)-grass pastures in Queensland

Leucaena leucocephala subsp. glabrata (leucaena)-grass pastures are productive, perennial and long-lived (40 years). However, little is known about changes in the productivity of these pastures as they age even though they are grazed intensively and are rarely fertilised. A postal survey of beef cattle producers in Queensland who grow leucaena pastures was conducted. The questionnaire gathered information regarding: property location; extent and age of leucaena pastures; soil type; leucaena and grass establishment methodology; grazing and fertiliser management; and grazier perceptions of changes over time in leucaena productivity, grass growth and ground cover, prevalence of undesirable grasses and weeds, and livestock productivity. Graziers were asked to report on both young (≤10 years old) and aging (10 years old) pastures under their management. Eighty-eight graziers responded describing 124 leucaena paddocks covering 11750 ha. The survey results described the typical physical and management characteristics of leucaena pastures in Queensland. Graziers reported a decline in leucaena productivity in 58% of aging pastures, and declines in grass growth (32%) and livestock productivity (42%) associated with declining leucaena growth. Leucaena decline was greater in soil types of marginal initial fertility, particularly brigalow clay, soft wood scrub, downs and duplex soils. Maintenance fertiliser was not applied to most (98%) leucaena pastures surveyed despite significant amounts of nutrient removal, particularly phosphorus and sulphur, occurring over prolonged periods of moderate to high grazing pressure. It is predicted that large areas of leucaena pasture will continue to suffer soil nutrient depletion under current management practices. Research is needed to develop ameliorative actions to reinvigorate pasture productivity.

Genotypic variation in proanthocyanidin status in the Leucaena genus

The proanthocyanidin (PA) status of 116 accessions from the Leucaena genus representing 21 species, 6 subspecies, 3 varieties and 4 interspecific hybrids was evaluated under uniform environmental and experimental conditions at Redland Bay, Queensland, Australia in October 1997. The PA content of lyophilized youngest fully expanded leaves was measured spectrophotometrically by the butanol/HCl assay referenced to L. leucocephala ssp. glabrata standard PA and expressed as L. leucocephala ssp. glabrata PA equivalents (LLPAE). Considerable interspecific variation in PA concentration existed within the genus, ranging from 0-339 g LLPAE/kg dry matter (DM). Taxa including L. confertiflora, L. cuspidata, L. esculenta and L. greggii contained very high (> 180 g LLPAE/kg DM) PA concentrations. Similarly, many agronomically superior accessions from L. diversifolia, L. pallida and L. trichandra contained extremely high (up to 250 g LLPAE/kg DM) PA concentrations, although these taxa exhibited wide intraspecific variation in PA content offering the potential to select accessions with lower (120-160 g LLPAE/kg DM) PA content. Commercial cultivars of L. leucocephala ssp. glabrata, known to produce forage of superior quality, contained low amounts of PA (33-39 g LLPAE/kg DM). Artificial interspecific hybrids had PA contents intermediate to those of both parents, Lesser-known taxa. including L. collinsii, L. lanceolata, L. lempirana, L. macrophylla, L. magnifica, L. multicapitula, L. salvadorensis and L. trichodes, contained undetectable to low (0-36 g LLPAE/kg DM) quantities of PA and have potential as parents to breed interspecific hybrids of low PA status and superior forage quality. Extractable PA was the dominant PA component, accounting for 91% of total PA within the genus. Regression analysis of accession ranks from different experiments compared to these results indicated that genetic regulation of Leucaena spp. PA content was consistent (P < 0.01) under different edapho-climatic environments. The distribution of PA within the Leucaena genus did not concur with the predictions of various evolutionary and phylogenetic plant defence theories.

Forage dry matter yields and psyllid resistance of thirty-one leucaena selections in Hawaii.

Forage yields ofL. leucocephala (Lam). de Wit have been reduced as the result of psyllid damage sinceHeteropsylla cubana Crawford invaded the Hawaiian Islands in 1984. The forage productivity and psyllid resistance of 31Leucaena species and interspecific hybrids were assessed from 5 harvests in Hawaii during 1991 and 1992. The trial consisted of an augmented randomized complete block with 22Leucaena selections in all 4 replicates, 2 selections in 3 replicates and 7 selections in 1 or 2 replicates.Forage (leaves and stems < 6 mm diam.) dry matter (DM) biomass yield over a 13-month period ranged from 1.4 to 34 Mg ha−1 from total DM ranging from 1.9 to 63.7 Mg ha−1. Percent forage fractions ranged from 49 to 78% (forage DM/total DM). The 10 selections in this trial of eitherL. pallida Britton & Rose, and its hybrids withL. leucocephala consistently produced both the highest forage and total DM yields averaging 22 and 40 Mg ha−1, respectively. This represented a three-fold increase in forage production when compared toL. leucocephala K636 (a standard around the world).The excellent performance of theL. pallida lines was attributed to high psyllid resistance and seedling vigor. AllL. pallida selections with the exception of K953 exhibited good psyllid resistance.Leucaena diversifolia Benth. K749,L. pallida K376, andL. esculenta (Moc. & Sesse) Benth. K950 had the highest psyllid resistance (p<0.05). Psyllid damage was negatively correlated to forage DM yield at both harvest 2 and 5 (r=−0.55,p<0.01,n=94). Forage DM was positively correlated to seedling vigor for the first harvest (r=0.83,p<0.001,n=74) and combined harvests (r=0.88,p<0.001,n=74).

New forage species for integration of sheep in rubber plantations

The productivity and persistence of 91 accessions of grasses, legumes and broad-leaved weeds were evaluated under young rubber at the Rubber Research Institute of Malaysia Experimental Station at Sungai Buloh near Kuala Lumpur. The species were chosen from world collections for their reported shade tolerance. Two small plot experiments were conducted from 1989 to 1990 and regular harvests were used to assess yield and persistence. The soil at the site was acidic (pH 4.7) and infertile, and light transmission declined from 65 to 20 % in Expt 1 and from 90 to 50 % in Expt 2. Pattern and cluster analyses were used to allocate accessions to groups with similar yield responses. Grasses which demonstrated both high yield and persistence of yield were Brachiaria decumbens MARDI accession, B. humidicola cv. Tully and Panicum maximum cvs Gatton, Riversdale and Vencedor. High yielding and persistent legumes were Stylosanthes scabra cv. Seca and S. guianensis CIAT 184. Survival of species monitored one year after completion of the experiment indicated the good longer term persistence of grasses Stenotaphrum secundatum, Panicum laxum, Paspalum notatum, Paspalum wettsteinii, Brachiaria humidicola and Panicum maximum cv. Vencedor, and legumes Arachis repens and A. pintoi.

Factors affecting the early growth of Leucaena leucocephala. 2. Importance of arbuscular mycorrhizal fungi, grass competition and phosphorus application on yield and nodulation of leucaena in pots.

Slow seedling growth is a limitation to the more widespread adoption of the tree legume, leucaena (Leucaena leucocephala). Three glasshouse trials examined the role of arbuscular mycorrhiza and phosphorus (P) nutrition in determining early growth and nodulation of leucaena. Treatments included soil types, inoculation with arbuscular mycorrhizal (AM) fungi, P application, grass competition and fumigation with methyl bromide, an anti-fungal agent. Plant measurements included colonisation by AM fungi, nodule weight, tissue nitrogen (N) and P concentrations. Slower early growth of leucaena in a soil from Mt Cotton than in soils from Gayndah or Theodore was due to slow colonisation of roots by AM fungi. Sequential harvests of plants revealed that rate of colonisation in the Mt Cotton soil was only half that in the Theodore soil prior to 28 days after planting resulting in subcritical P concentrations 21 days after sowing and an approximate halving of top dry weight 41 days after sowing. However, following increased infection, tissue P concentration and final plant growth 98 days after sowing were similar in both soils. Early seedling growth in the Mt Cotton soil was increased by inoculating the soil with mulch containing AM fungi but not with soil collected from beneath established leucaena added at a lower rate. Phosphorus application significantly increased growth of leucaena seedlings, but only the highest rate of 1200 kg P/ha was able to prevent early P deficiency. Final growth was reduced by 50% in the presence of Panicum maximum as a result of increased competition for N and P and by 90% in fumigated soil as a result of P deficiency. The results of these experiments confirm the important role of AM fungi on early seedling growth of leucaena. However, the potential to increase early growth using a soil or mulch inoculum containing AM fungi or P fertiliser may be limited by the high rates of application needed. More work is needed to determine whether slow rate of infection is a significant limitation in soils other than the Mt Cotton soil in the field.