Output 3: Grass and legumes genotypes with superior adaptation to edaphic and climatic constrains are developed 3.1 Genotypes of Brachiaria and Arachis with adaptation to edaphic factors Highlights • Collaborative research conducted with Hokkaido University in Japan generated the first evidence based on 27Al NMR analysis that organic acids within root tissue help detoxify aluminum in non- accumulator species such as in the Brachiaria hybrid cv Mulato. • Showed that the high level of aluminum resistance in signal grass (Brachiaria decumbens) is part of a generic resistance mechanism that is effective against trivalent cations in general. • Showed that hematoxylin staining could be employed as a quick selection criterion to discard Al- sensitive genotypes in the Brachiaria breeding program, because most of the Al accumulates in the external layer of root meristems and should be readily stainable with hematoxylin in intact root apices. • Constructed a genetic linkage map for aluminum resistance in Brachiaria, evaluated 50% of the polymorphic SSRs and AFLPs in the F1 cross of B. decumbens x B. ruziziensis and found preliminary associations between markers SSRs and AFLPs and three phenotype root traits of aluminum resistance. • Using microarray technology, the 3’-UTR sequences of candidate genes associated with aluminum (Al) resistance in Brachiaria decumbens were identified by comparing expression levels between genotypes and treatments. • Using screening procedure to evaluate aluminum resistance, 3 sexual hybrids (SX03NO/0846, SX03NO/2367, SX03NO/0881) were identified with greater level of Al resistance than that of the sexual parent. One of the apomictic hybrids (BR02NO1372) was outstanding in its level of Al resistance and this hybrid is also resistant to spittlebug • Showed that the Brachiaria hybrid, FM9503-S046-024 (Mulato 2) performed well into the third year after establishment in the Llanos and its superior performance at 30 months after establishment was associated with its ability to acquire greater amounts of nutrients, particularly Ca and Mg from low fertility soil • Results from a 4-year field study in the Piedmont showed that the Arachis pintoi accessions CIAT 18744, 18751 and 22159 were superior to the commercial cultivar (CIAT 17434) in terms of persistence with low amounts of initial fertilizer application. • Collaborative research conducted in Goettingen, Germany, under controlled environmental conditions in a growth chamber, showed that the Arachis pintoi accessions CIAT 18744 was more efficient acquiring P from less available P-pools in a low-P oxisol than the commercial cultivar, CIAT 17434. This high P efficiency and the increase of P uptake were found to be due to a high P influx. The activity of acid phosphatase on root surface and exudation of organic acids (lactic and acetic) did not contribute to this increase in P influx. 53 3.1.1 Edaphic adaptation of Brachiaria Contributors: I. M. Rao, P. Wenzl, J. W. Miles, J. Tohme, M. Ishitani, J. Ricaurte, R. García, A. L. Chaves, M. E. Recio, A. Arango, D. F. Cortes, G. Gallego, E. Gaitán, and C. Plazas (CIAT), T. Watanabe, H. Yano, M. Osaki (Hokkaido University, Japan) Previous research on mechanisms of adaptation program. The use of improved screening methods of Brachiaria species to acid soil stress factors and identification of QTLs and candidate genes indicated that Brachiaria decumbens cv. responsible for Al resistance and adaptation to Basilisk is highly resistant to toxic levels of Al and low P supply will contribute toward development low supply of P. Based on this knowledge, rapid of superior genotypes that combine several and reliable screening procedure to evaluate Al desirable traits to improve pasture productivity on resistance was developed to improve the acid, infertile soils and to combat pasture efficiency of the on-going Brachiaria breeding degradation. 3.1.1.1 Investigating physiological and genetic aspects of aluminum resistance in Brachiaria As part of the restricted core project funded by to investigate physiological and genetic aspects of BMZ-GTZ of Germany, we continued our efforts aluminum resistance in Brachiaria. A) Aluminum resistance coincides with differential resistance to trivalent lanthanide cations in Brachiaria Contributors: P. Wenzl (CAMBIA, Australia), A.L. Chaves, M.E. Recio, J. Tohme and I.M. Rao (CIAT) Rationale Materials and Methods Signalgrass (Brachiaria decumbens) has Seeds were germinated in continuously aerated evolved a highly effective Al-resistance 200 M CaCl (pH 4.20) for 3-4 days. mechanism that does not appear to rely on 2 Homogeneous seedlings were selected and their chelation of Al ions with organic-acid anions. root lengths were recorded. The seedlings were Electrical charges at the external surface of root then transferred to continuously aerated cells generate an electrostatic potential that treatments solutions in the greenhouse (toxicant modulates cell-surface ion activities and hence + 200 M CaCl , pH 4.20; see left panel below). ion uptake and intoxication. We hypothesized that 2 the superior Al resistance of signalgrass After three days root length were measured compared to closely related ruzigrass again. Each experiment comprised six toxicant (B. ruziziensis) could be due to a less negative levels plus the toxicant-free basal solution. Three surface potential at root cells that are critical to independent experiments were performed for root growth and elongation. We tested this each toxicant. The concentration of a toxicant hypothesis by investigating whether Al resistance that inhibited relative root elongation (RRE) by of signalgrass was associated with superior 50 % (C ) was determined for each of the two 50 resistance to other cations toxicants and greater species after fitting a Weibull function to the susceptibility to anionic toxicants. pooled data from the three replicate experiments by using the Marquardt-Levenberg algorithm (right panel above). The SE of C was computed 50 based on error propagation rules. The C values 50 54 of the two species for a particular toxicant were These results suggest that Al resistance in considered to be different if their 95 % signalgrass is part of a more generic resistance confidence intervals did not overlap. mechanism that is effective against trivalent cations in general, a finding that confirms the Results and Discussion unique physiological basis of Al resistance in signalgrass. The superior Al resistance of signalgrass compared to ruzigrass was associated with The pattern of resistance to cationic and anionic greater resistance to all the trivalent lanthanide toxicants, however, is not consistent with the idea cations tested (Figure 11). If a lower root cell that a less negative root cell surface potential surface negativity was the cause for the greater confers resistance to cationic toxicants as a result lanthanide resistance of signalgrass, signalgrass of electrostatic interactions, that is, solely based should be more resistant to other cationic on the charge but not the structural properties of toxicants and more sensitive to anionic toxicants. a toxicant. The cross-resistance of signalgrass to The two species, however, were equally sensitive Al and other, mostly trivalent inorganic cations to the majority of divalent and monovalent cations may instead be based on interspecific differences (Figure 10b) and most anions (Figure 10c). Apart in critical cellular sites to which trivalent cations from lanthanides and Cd2+, signalgrass was more such as Al3+ and lanthanides bind. More work is resistant than ruzigrass only for those inorganic required to elucidate the nature of these sites and toxicants that are in redox equilibrium with a to develop biochemically-based strategies to trivalent cationic form: Fe2+ ↔ Fe3+, Co2+ ↔ isolate the underlying genes. It may be possible to Co3+, Cr O 2" 2 7 ↔ Cr3+. An organic trivalent cation use lanthanide cations as proxies for Al to (spermidine3+), by contrast, was equally toxic to circumvent some of the problems and ambiguities both species. caused by the difficulties to predict Al speciation. (a) Aluminum and lanthanides 140 Signalgrass 14 120 Ruzigrass 12 100 10 80 * * 8 (c) 60 6 * * * * * Anions 40 * 4 18 Signalgrass 1.8 Ruzigrass 20 * 2 16 1.6 14 1.4 0 0 3+ 3+ 3+ 3+ 3+ 3+ 3+ 3+ 3+l 12 1.2 A La Ce Pr Eu Gd Tb Er Yb 10 1.0 (b) 8 0.8 Other cations 6 0.6 70 Signalgrass 7 4 * 0.4 60 Ruzigrass 6 2 0.2 50 5 0 - 0 PB - 2- 2- 2- 40 * 4 T Cl O 3 O 7 O 4 4 eO Cr 2 Mo S 30 3 20 2 10 * * 1 0 0 2+ 2+ 2+ 2+ + P 3+ + b 2+ Zn Fe Co Cd TP ine R Ni mid pe r S Figure 11. Concentrations of cationic and anionic toxicants required for inhibiting root elongation of signalgrass and ruzigrass by 50 % (C50). 55 C (µM) for Zn2+, Fe2+ 3+ 50 , C50 (µM) for Al Co2+, Cd2+, TPP+ C50 (µM) for spermidine3+, Rb+, Ni2+ C50 (µM) for all but Al3+ C50 (µM) for TPB-, ClO -, Cr O 2-3 2 7 C50 (µM) for MoO 2-4 , SeO 2-4 B) Accumulation of callose and aluminum in root tips of Brachiaria spp. Contributors: A. Arango, P. Wenzl, I.M. Rao, and J. Tohme (CIAT) Rationale (w/v) hematoxylin and 0.01% (w/v) KIO3 for 20 min. Callose was visualized by staining with 0.1% The effects of aluminum (Al) toxicity on callose (w/v) aniline blue and 1M glycine NaOH (pH accumulation were evaluated in root tips of 9.5). Brachiaria populations previously evaluated for physiological and genetics response (IP-5 Annual Results and Discussion report 2002, 2003). Brachiaria decumbens (Al resistant ), B. ruziziensis (Al sensitive) and two Differential hematoxylin staining was observed contrasting B. ruziziensis x B. decumbens between sensitive and tolerant genotypes. Root hybrids were evaluated after 3, 12 and 21 days of apices of B. decumbens (Figure 12A) and an Al- Al treatment (200 µM AlCl ). resistant hybrid (Figure 12C) did not accumulate 3 much Al. By contrast, root apices of Materials and Methods B. ruziziensis (Figure 12B) and an Al-sensitive hybrid (Figure 12D) accumulated Al in the outer Rooted stem cuttings of B. decumbens, layer of root meristems. Aniline-blue staining of B. ruziziensis and two contrasting hybrids were histological sections revealed a higher content of cultivated as described previously (IP-5 Annual callose for Al-sensitive genotypes. Report, 2003). Root apices (5 mm), collected after different times of exposure to Al, were The pattern of callose accumulation was only fixed during 48 hours in 2.5 % glutaraldehyde to partially correlated with that of Al (visualized by detect callose with aniline blue, or in a 1:1 mixture hematoxylin). Al-tolerant B. decumbens of 3.7 % phormol (pH 7.4) and 2.5 % accumulated callose exclusively in the root cap glutaraldehyde to detect Al accumulation with and at the surface of the root meristem. Al- hematoxylin. sensitive B. ruziziensis accumulated a large amount of callose in cortical and vascular tissues, Samples were cut (7 µm) and processed and an area where little Al was detected in the aluminum was visualized by staining with 0.1% hematoxylin stain (Figure 13). Figure 13. Callose detection by aniline-blue staining Figure 12. Hematoxylin staining of Brachiaria.root apices. in histological sections of Brachiaria root apices. A, B. A. Al-tolerant parent (B. decumbens); B. Al-sensitive parent B. decumbens (without and with Al). C, D. (B. ruziziensis); C. Al-tolerant hybrid; D. Al-sensitive hybrid B. ruziziensis (without and with Al). Arrows designate (12 days of Al treatment). callose deposition (3 days of Al treatment). 56 Hematoxylin staining could be employed as a quick of the Al accumulates in the external layer of root selection criterion to discard Al-sensitive geotypes meristems and should be readily stainable with in the Brachiaria breeding program, because most hematoxylin in intact apices. C) Identification of molecular markers and QTLs associated with gene(s) conferring aluminum resistance in a Brachiaria ruziziensis × Brachiaria decumbens cross (F1) Contributors: J. Vargas, C. Quintero, C. Rosero, G. Gallego, M.C. Duque, M.E. Buitrago, A.L. Chaves, J.W. Miles, P. Wenzl, I.M. Rao and J. Tohme Rationale Scientific Instruments) and diluted at 4ng/ul for SSRs amplification and 25ng/ul for AFLPs Acid soils have been estimated to occur on about amplification. Methods for the isolation of 40% of the arable land (3.95 billions of ha). Plant microsatellites, and the methodology for PCR growth on these soils is constrained mainly by amplification and evaluation of polymorphism, aluminum (Al) toxicity and deficiencies of were as described previously (SB-2 Annual nutrients such as phosphorus (P), nitrogen (N), Report, 2000; 2001) with some modifications. and calcium (Ca). There is considerable variation The AFLP Analysis System I kit, and AFLP within and between plant species in their ability to Analysis System II Small Genome, from resist Al, and this variation within some species Invitrogenâ were used for AFLP amplification, has allowed breeders to develop genotypes that following the instructions, with some are able to grow on acid soils. Within the modifications. Silver staining (Promega Inc., Brachiaria genus, Al resistance of signal grass USA), was used to visualize allele segregation of (B. decumbens Stapf cv Basilisk), a widely sown the markers on 6% denaturing polyacrylamide tropical forage grass, is outstandingcompared gels with 5M Urea and 0.5X TBE. with the closely related ruzigrass (B. ruziziensis Germain and Evrard cv Common). The main Results and discussion objective of this work is to identify microsatellites, AFLPs and QTLs associated with the gene(s) Microsatellites: 73 SSRs were evaluates in the conferring aluminum resistance in a Brachiaria parental genotypes of which 40 were found to be ruziziensis × Brachiaria decumbens cross. polymorphic. When run in the progeny, three sets of primers did not amplify in 30% of the progeny, so they were discarded together with three more Materials and Methods monomorphic microsatellites. Ninety-seven polymorphic alleles were scored in the population, An F1 hybrid population of 263 individuals out of which 63 were found with the (B. ruziziensis x B. decumbens) was used for B. decumbens, Al-tolerant, 606 genotype, while this study. Young leaves were cut and placed in 34 carried the B. ruziziensis, Al-sensitive, 44-2 paper bags in an incubator previously set at 45- genotype (Figure 14). 50°C. Samples were allowed to dry for at least 20h, or until the leaves were dry enough that they AFLPs: 64 combinations of primers were as- break easily. Samples were stored at –80°C until sayed with the two parental genotypes. Among grinding. Dried leaf tissue was grinded with them, 12 having high number of polymorphic stainless steel spheres with vigorous shaking. bands, were chosen. To date 3 combinations Genomic DNA was extracted using a CTAB- (E-ACC/M-CAC; E-ACT/M-CTA and E-ACG/ Chloroform protocol with some modifications for M-CAG) were run in the progeny and yielded 63 small amounts of tissue. DNA was quantified on polymorphic bands distributed as follows: 50 were a DyNA Quant 200ä Flourometer (Hoefer found in the Al-tolerant parental and 13 in the Al- sensitive genotype. (Figure 15). 57 B GM 49 B A A A GM 97 A B B GM 105 A A B B A GM 6 A B B Figure 14. Segregation of four SSR alleles in some individuals of the F1 B. ruziziensis x B. decumbens progeny. A, Parental genotype 606 (Al-resistant) and B, parental genotype 44-2 (Al-sensitive). Linkage analysis: Segregation of markers in a linkage groups. These linkage groups span 445.3 1:1 ratio, as single dose restriction fragments cM, and the average marker density is one per (SDRF), will be determined by a Chi-square test. every 6.1 cM. The position of 78 markers is The data matrixes for presence or absence of shown in Figure 15, on the framework molecular bands were analyzed with MAPMAKER v3.0b genetic map of Brachiaria (LOD = 25 and tetha for PC, and MAPMAKER v2 for Macintosh. ( θ ) = 25). Map distances are shown in Kosambi Using 56 SSR and 50 AFLPs molecular markers, map units. we constructed a putative linkage map with 18 Population F1 PT PS PT PS Figure 15. Segregation of AFLPs bands in individuals of the F1 B. ruziziensis x B. decumbens progeny. PT, Parental genotype 606 (Al-resistant) and PS, parental genotype 44-2 (Al-sensitive). 58 Association between Molecular Markers and Al Table 15. Association analysis between molecular resistance: To find association with molecular markers and phenotipic variable for Al resistance markers, a preliminary analysis of 106 markers at in a Brachiaria decumbens x Brachiaria ruzizienses hybrid population. the 10% level was done using SAS. Putative associations were found between 78 SSRs and Linkage Phenotipic 2 AFLPs markers and the phenotypic Marker R Group Variable a characterization under greenhouse conditions. The GM 44d2 0.0267 B Rl b three phenotypic variables for Al resistance (root GM 44d2 0.0218 B Tips length, abundance of root tips, mean root diameter) C1b4 0.0181 R Tips c were analyzed for association with molecular GM 109c 0.0177 N Rd markers. We found 13 molecular markers with R2 C1b17 0.166 D Rd between 0.0124-0.0267 that explain the variance C2b17 0.0161 R Tips for these traits; molecular markers associated with C2b15 0.016 I Rd phenotypic characterization are in blue (Table 15 GM 109d 0.0156 N Rd and Figure 16). GM 79c 0.0156 A Rl GM 58a 0.0151 D Rd Further work is in progress: (i) to saturate linkage GM 44d1 0.0149 B Tips map of B. decumbens CIAT 606 parent with SSR C2b12 0.0145 B Rd and AFLPS; (ii) to analyze data for mapping for Al C1b12 0.014 R Tips resistance; (iii) to conduct QTLs analysis for Al GM 79d 0.013 D Rd resistance; and (iv) to design of SCARs for marker GM 44d2 0.0124 B Rd assisted selection. a Root Length b Abundance of root Tips c Root Diameter Figure 16. Preliminary Brachiaria linkage map of a B. ruziziensis × B. decumbens F1. Markers in blue indicate putative association with Al resistance. 59 D) Identification of candidate genes associated with aluminum resistance in Brachiaria decumbens. Contributors: A. Arango, G. Gallego, D. Bernal, P. Wenzl, I.M. Rao, M. Ishitani and J. Tohme (CIAT) Rationale the vector used for library preparation, unrelated genes such as insulin, Sp1 β-cell, HPH) and The 3’-UTR sequences of candidate genes positive controls (GADPH of Brachiaria, α- associated with aluminum (Al) resistance were tubulin, Spy genes). A total of 768 controls were identified by comparing expression levels spotted onto the array. The logarithms of the between genotypes and treatments. Subtractive crude ratios between the two channels were first libraries were prepared from root apices of normalized by using the lowess algorithm and contrasting genotypes grown in the presence and then analyzed with the Significance Analysis of absence of Al: Brachiaria decumbens (Al- Microarrays (SAM) software. resistant parent), three Al-resistant B. ruziziensis x B. decumbens hybrids, B. ruziziensis (Al- Differentially expressed clones were amplified, sensitive parent), and seven Al-sensitive hybrids purified (Qiagen kit) and sequenced (ABI (Annual Report 2003). Microarray technology BigDye terminator kit). The sequences obtained was then used to catalogue clones derived from were compared against those in the UTR genes that were differentially expressed between database at http://bighost.area.ba.cn.it/BIG/ samples. UTRHome. Materials and Methods Results and Discussion Subtractive libraries of 3’UTR fragments were prepared with the differential subtraction chain The microarray hybridizations identified a total of (DSC) method (IP-5 Annual Report, 2003). 35 3’-UTR fragments of candidate genes that Inserts were amplified and arrayed in duplicate were expressed differentially in the four on glass slides. Pairs of contrasting RNA comparisons between target and control pools populations (control, target) were hybridized to (Figure 17, Table 16). microarrays (Table 16). Two pairs of dye-swap hybridizations were performed per combination of Seven clones contained the post-transcriptional control and target. control sequence 15LOX-Dice (15-Lypoxigenase Differentiation Control Element), seven clones Microarray sample pools (MSPs) were contained the ribosomal regulatory element IRES synthesized from 2.4 to 240 ng of cDNA to cover (Internal Ribosome Entry Site), and one clone a 100-fold range of signal intensities. They were was homologous to the 3'-UTR of a Arabidopsis arrayed together with other controls, such as thaliana gene coding for the germination protein negative controls (spotting buffer, polylinker of Table 16. Populations evaluated in conditions of Al toxicity by Microarrays. Combination Control Treatment Target Treatment 1 Al-sensitive parent + AlCl3 (200 M) vs Al-resistant parent + AlCl3 (200 M) hybrids hybrids 2 Al-resistant parent + Al (0 M) vs Al-resistant parent + AlCl3 (200 M) hybrids hybrids 3 B. ruziziensis AlCl3 (200 M) vs B. decumbens AlCl3 (200 M) 4 B. decumbens Al (0 M) vs B. decumbens AlCl3 (200 M) 60 Figure 17. Differentially expressed genes associated with Al resistance (red points). A. Al-sensitive parent + hybrids in Al treatment vs Al-resistant parent + hybrids in Al treatment; B. Al-resistant parent + hybrids in control treatment vs Al-resistant parent + hybrids in Al treatment; C. Al-sensitive parent (B. ruziziensis) in Al treatment vs. Al-resistant parent (B. decumbens) in Al treatment; D. Al-resistant parent (B. decumbens) in control treatment vs. Al-resistant parent (B. decumbens) in Al treatment. GLP2 (EMBL: BT002170). Five clones had no corresponding full-length genes in a cDNA library match in the data base. The 3’-UTR clones of previously prepared from root apices of the Al- differentially expressed genes identified in these resistant parent (B. decumbens). experiments will be used as probes to isolate the E) Internal mechanisms of plant adaptation to aluminum toxicity and phosphorus starvation in three tropical forages Contributors: T. Watanabe, M. Osaki, H. Yano (Hokkaido University, Japan) and I. M. Rao (CIAT) Rationale prevent Al toxicity within the plant by creating Al complexes with organic acids or silicon. Soil acidity inhibits plant growth, principally because of toxicity from excess aluminum (Al) Phosphorus is a major nutrient that plays a role in and lack of nutrients, especially phosphorus (P). forming phospholipids, nucleic acids, nucleosides, Plant mechanisms for tolerating toxic levels of Al coenzymes, and phosphate esters in plants. are usually grouped into two: external Under P starvation, adapted plants produce more mechanisms to prevent Al invading root cells, or fine roots or root hairs and/or increase root mass internal mechanisms that provide tolerance to to acquire more P from soils. The roots release excess Al. One significant external mechanism is chelating compounds (e.g. organic acids) to for roots to exude organic acids into the mobilize and use insoluble phosphate compounds rhizosphere, where they then make stable (e.g. those with Al, Fe, and Ca). Roots also complexes with Al. Internal mechanisms are release enzymes (e.g. acid phosphatase or often found in Al-accumulator species, which APase) to use organic P. In addition, under P 61 deficiency, adapted plants have the strategy of Experiment 1: Tolerance of Al stress and P efficiently using P in cells. Phosphohydrolases starvation: After precultivation, seedlings were may function as a P-recycling mechanism in transferred to 36-litre containers containing plants. Inorganic P, liberated by APase or treatment solutions, which comprised: the ribonuclease (RNase) in old tissues are standard nutrient solution with 0.06 mM P retranslocated to young tissues. Induction of (control); the same solution but no P (-P, i.e. 0 APase and RNase in plants under P deficiency mM P); and the same solution but with Al added has been intensively studied. Another strategy— (+Al, i.e. 0.37 mM Al and 0.06 mM P). Al and P mycorrhizal symbiosis—is also important for were added as Al2 (SO4) 3 and NaH2PO4, many plant species. respectively. At the end of 10 days of treatment, the roots of seedlings from each treatment were Brachiaria species are adapted to low-fertility first washed with tap water, then with deionized acid soils in the tropics, tolerating both excess Al water, and transferred to polyethylene bottles and P starvation very well. However, their containing 0.1 mM CaCl2 (pH 4.0) for collecting mechanisms of high level of adaptation have yet root exudates. After being left overnight from to be defined. In B. decumbens, for example, Al- 17:00 to 8:00 (i.e. 15 h) in a greenhouse, the exclusion mechanisms, such as exudation of exudates were collected, the roots washed with organic acids and rhizosphere alkalinization, are deionized water, and the seedlings cut to separate not involved in its high level of resistance to Al. roots from shoots. The fresh weight of each set Many other tropical forage grasses and legumes of organs was determined. Half of the fresh also grow well in acid soils, adapting to excess Al samples were dried in a forced-air oven at 80°C and P starvation stresses by using mechanisms for 72 h, then weighed and digested with H2SO4- that are still unclear. We therefore studied the H2O2 for mineral analysis. The other half of the mechanisms of adaptation to Al toxicity and P samples was used to determine organic acids. starvation in three tropical forages: two grasses Acid phosphatase activity in root exudates was (a Brachiaria hybrid and Andropogon determined. One unit of acid phosphatase activity gayanus) and one legume (Arachis pintoi). was defined as the amount of enzyme that hydrolyzed 1 µmol of pNPP per minute. Soluble Materials and Methods organic acids in fresh samples were determined and the concentration of organic acids was For both experiments, we used seeds from three measured by capillary electrophoresis. The levels tropical forages: the grass Brachiaria hybrid of Al, K, Ca, and Mg in leaves and roots were (B. ruziziensis Ger. & Ev. clone 44-06 × determined. Concentrations in leaves and roots B. brizantha (A. Rich.) Stapf CIAT 36061, also were determined by the semi-micro Kjeldahl known as cv. Mulato); the grass Andropogon method for N and the vanado-molybdate yellow gayanus Kunth (CIAT 621); and the legume method for P. Arachis pintoi Krap. & Greg. (CIAT 17434). The seeds were sterilized with sodium Experiment 2: 27Al NMR study: Brachiaria hypochlorite for 10 min, washed with deionized seedlings were prepared as described above, and water, and sown on a moderately moist perlite- transferred to 36-litre containers carrying the vermiculite mixture (1:1, v/v) in a greenhouse at standard nutrient solution, but with 2.8 mM Al at Hokkaido University, Sapporo, Japan. Uniform pH 3.7 added, and left to grow for 1 month. The seedlings (shoot height = 10 cm) were much higher Al concentration was used to ensure transplanted to containers containing a standard clear peaks in the 27Al NMR spectrum. Even so, nutrient solution with 0.06 mM P at pH 4.0, and the Brachiaria seedlings grew well (data not grown for one week to adapt to hydroponic shown). After treatment, roots were removed conditions. from the seedlings and washed, first with tap water, then with deionized water. The roots were grouped into three: Fraction (a), roots given the 62 water washings only, and used to determine total Results and Discussion amounts of Al and organic acids; Fraction (b), roots were also washed with 0.1 N HCl for 5 min Experiment 1: Tolerance of Al stress and P to remove apoplastic, soluble or loosely bound, starvation. The effect of each treatment on components; and Fraction (c), roots were also plant growth was expressed as dry matter washed with 0.1 N HCl for 5 min, frozen at - accumulation in each treatment relative to that of 50°C for 1 h to rupture cell membranes, thawed, the control treatment (Figure 18). The +Al and washed again with 0.1 N HCl for 5 min to treatment (+0.37 mM Al) did not affect growth in remove symplastic, soluble or loosely bound any of the species used in the study. The -P components. Al and organic acid concentrations treatment did not inhibit growth in Brachiaria in each fraction were determined as described and A. pintoi, but did in A. gayanus. Relative for Experiment 1. Each fraction of Brachiaria growth rate (RGR) in the control treatment was roots was placed in a 10-mm-diameter NMR much higher in the two grasses than in the tube. AlCl3 (0.1 M) solution was used as an legume (Figure 18). The root-to-shoot ratio in external reference to calibrate the chemical shift Brachiaria and A. pintoi increased remarkably (0 ppm). 27Al NMR spectra were recorded, using with -P treatment, whereas it remained a Bruker MSL400 spectrometer at 104.262 MHz. unchanged in A. gayanus (Figure 18). The The spectra were obtained by using a frequency treatments hardly affected acid phosphatase range of 62.5 kHz, a pulse width of 12 µs, a delay activity in root exudates of seedlings, except in time of 0.16 ms, a cycle time of 0.5 s, and 4000 A. gayanus where it increased with +Al scans. treatment (Table 17). 1.6 Shoots 1.4 Roots 1.2 1 0.8 0.6 0.4 0.2 0 Brachiaria Andropogon Arachis hybrid gayanus pintoi Figure 18. Effects of P starvation and Al toxicity on plant growth in Experiment 1. Plant growth was expressed as the relative dry matter accumulation (i.e. [dry weight after treatment – initial dry weight in each treatment]/[dry weight after treatment – initial dry weight in control treatment]). Bar values indicate root-to-shoot ratios, and the range for each bar indicates the ± SE value. Relative growth rate (dry weight after treatment – initial dry weight)/initial dry weight) in control treatment was 10.4, 8.5, and 1.1 g g-1, in a Brachiaria hybrid, Andropogon gayanus, and Arachis pintoi, respectively. Table 17. Acid phosphatase activity in root exudates of forage seedlings grown under three treatments Acid phosphatase (mU g-1 f. wt 15 h-1) Species Control +Al -P Brachiaria hybrid 170.9 72.2 100.7 42.1 202.9 24.9 Andropogon gayanus 182.3 3.2 213.6 24.5 166.6 21.4 Arachis pintoi 76.1 2.9 76.6 7.0 54.6 1.0 63 Relative dry matter accumulation (g -g1) Control 0.275 Al 0.387 P 0.551 Control 0.549 Al 0.374 P 0.577 Control 0.283 Al 0.313 P 0.369 Mineral concentrations in leaves and roots, and were about 600 and 340 mg kg-1, respectively. total organic acid concentrations in roots are Organic acid concentrations in roots increased shown in Table 18. Nitrogen concentrations under the +Al treatment in the Brachiaria hybrid were hardly affected by the treatments. and A. gayanus. Phosphorus concentrations in leaves and roots of all species declined drastically in the -P treatment, Experiment 2: 27Al NMR study. Aluminum, especially in the Brachiaria hybrid and oxalate, malate, and citrate concentrations as A. gayanus. Phosphorus concentrations tended determined by the 27Al NMR spectrum are to decrease as K concentrations decreased. shown for each fraction in Figure 19. Aluminum Aluminum concentrations in Brachiaria leaves concentration in fraction (b) (roots after removing with the +Al treatment were less than 100 mg apoplastic soluble components) did not differ kg-1, whereas in A. gayanus and A. pintoi, they significantly from that in fraction (a) (intact Table 18. Concentrations of N, P, K, Ca, Mg (mg g-1d. wt), Al (mg kg-1d. wt) and total organic acid (µmol g-1f. wt) in organs of Brachiaria hybrid, Andropogon gayanus and Arachis pintoi. Values are the means of three replicates±SE. SEs were not shown in organic acid concentrations because of two replicates. ND = not determined. Species Leaf Root Control +Al - P Control + Al - P Brachiaria N 47.3 ±1.7 48.0 ±0.6 55.8 ±0.8 27.0 ±0.9 27.6 ±1.3 25.4 ±1.0 hybrid P 8.8 ±0.7 14.3 ±0.3 0.8 ±0.0 5.6 ±0.2 8.8 ±0.5 1.1 ±0.0 K 24.9 ±0.5 25.3 ±0.8 9.7 ±0.0 33.7 ±0.4 33.1 ±2.2 16.1 ±3.5 Ca 4.5 ±0.6 2.8 0.3 2.3 ±0.2 1.3 ±0.1 1.3 ±0.0 0.7 ±0.1 Mg 3.7 ±0.3 3.7 ±0.3 3.3 ±0.1 3.5 ±0.1 4.5 ±0.3 2.3 ±0.1 Al 49 ±6 75 ±11 56 ±7 357 ±40 3285 ±76 687 ±24 Organic acid ND ND ND 0.79 4.71 0.23 Andropogon N 21.7 ±0.2 27.0 ±0.3 20.5 ±0.2 18.7 ±1.3 20.6 ±0.2 26.0 ±0.7 gayanus P 6.5 ±0.0 5.5 ±0.2 0.6 ±0.0 10.6 ±0.9 7.0 ±0.0 0.2 ±0.3 K 18.8 ±0.2 20.7 ±0.5 14.6 ±0.7 24.9 ±1.7 28.8 ±0.2 10.8 ±1.4 Ca 6.5 ±0.0 2.9 ±0.0 4.4 ±0.1 2.1 ±0.1 2.0 ±0.0 1.7 ±0.1 Mg 3.7 ±0.1 3.0 ±0.0 3.1 ±0.0 3.5 ±0.2 4.5 ±0.0 2.8 ±0.1 Al 39 ±3 594 ±22 23 ±4 219 ±15 5822 ±4 117 ±139 Organic acid ND ND ND 3.44 10.70 3.71 Arachis N 49.5 ±1.8 51.4 ±1.4 56.0 ±1.3 31.4 ±0.5 35.1 ±0.4 37.2 ±1.0 pintoi P 17.1 ±0.5 10.1 ±0.7 2.4 ±0.1 31.1 ±0.5 22.9 ±1.3 2.3 ±0.1 K 27.5 ±0.1 25.0 ±1.6 22.6 ±0.6 30.2 ±0.7 28.8 ±1.6 33.5 ±1.8 Ca 15.6 ±0.2 19.5 ±1.5 13.8 ±0.8 5.1 ±0.0 3.0 ±0.6 3.5 ±0.1 Mg 8.8 ±0.0 8.8 ±1.6 6.5 ±0.3 15.3 ±0.4 5.0 ±0.3 3.5 ±0.2 Al 83 ±6 339 ±15 58 ±4 605 ±1 4402 ±143 356 ±6 Organic acid ND ND ND 0.47 0.53 1.19 64 roots). The differences in organic acid concentrations between fractions A and B were also small. 27Al NMR spectra obtained from fractions A and B were very similar, and showed several peaks downfield and a small peak at 0 ppm. In contrast, in fraction (c) (roots after removing apoplastic soluble components, followed by removing symplastic soluble components), Al concentration was much lower than in fractions A and B. The 27Al NMR spectrum was also different, with the resonance peaks downfield decreasing and a peak at 0 ppm becoming higher and sharper. Of the three organic acids, malate decreased drastically in fraction (c) (Figure 19). All three tropical forages used in the present study were highly tolerant of Al and no growth reduction was observed with the +Al treatment (+0.37 mM Al) (Figure 18). Concentrations of Ca and Mg, the uptake of which appeared inhibited by Al, decreased in the +Al treatment (Table 18). However, concentrations of other major nutrients were not affected by Al application (Table 18), suggesting that Al did not affect root function for these nutrients. To prevent or evade Al toxicity, plants growing in acid soils have developed mechanisms to exclude Al from roots or to tolerate high Al concentra- tions in tissues. Although organic acid exudation from roots is a major mechanism for excluding Al from roots, this mechanism was not evident in Brachiaria species. Previous research showed that Al application increases organic acid concen- Figure 19. 27Al NMR spectra and concentrations of Al trations in roots of Brachiaria species. In our (mg g-1 d. wt) and organic acids (µmol g-1 f. wt) in three root study, organic acid concentrations increased with fractions of a Brachiaria hybrid. Fraction (a) = intact roots; Al application, but not with P starvation in fraction (b) = roots after removing soluble or loosely bound apoplastic components; fraction (c) = roots after removing Brachiaria and A. gayanus (Table 18). soluble or loosely bound apoplastic and symplastic components. AlCl3 (0.1 mM) was used as an external We speculated before that the increase in organic reference to calibrate the chemical shift (0 ppm). acids in roots suggested that they play a role in the internal detoxification of roots from Al by acting as ligands for Al. In many Al-accumulator An important result of our study was to species, leaves with high concentrations of Al are demonstrate that Al-ligand complexes occur detoxified by organic ligands, such as Al-oxalate, inside the root cells of a Brachiaria hybrid, an Al Al-catechin, and Al-citrate. The same mecha- non-accumulator genotype. We applied simple nisms are considered possible in roots. fractionation of Al and 27Al NMR to roots of the Brachiaria hybrid to evaluate the occurrence of Al-ligand complexes in root tissues. The 27Al 65 NMR spectrum obtained from intact roots Although APase activity in root exudates, which (fraction a) showed several peaks downfield at contributes to the use of organic P in the 10-20 ppm, suggesting that most of the soluble Al rhizosphere, was hardly affected by P deficiency in roots makes octahedral complexes, presumably in any of the species (Table 17), it was higher with organic acids. Both the 27Al NMR spectrum than that of other forage species. While the and Al concentrations in fraction (b), in which ability to use organic P in the rhizosphere did not soluble or loosely bound apoplastic Al was affect plant growth in the solution culture removed, were almost the same as those of intact experiment, high APase activity in root exudates roots (fraction a), indicating that root apoplast is may significantly affect P acquisition in soils. In not a primary site for Al accumulation in addition, the higher root-to-shoot ratio under P Brachiaria hybrid (Figure 19). In contrast, in starvation in Brachiaria hybrid and A. pintoi fraction (c), the peaks downfield in the 27Al would increase P uptake in soils (Figure 18). NMR spectrum, Al concentrations, and malate concentrations all drastically decreased after This study showed that the Brachiaria hybrid, removing symplastic components (Figure 19), Andropogon gayanus, and Arachis pintoi adapt indicating that most of the chelated Al occurs in well to acid soils with low P and excess Al. The root symplast, and that malate may participate in results indicate that the Brachiaria hybrid makes formation of Al complexes. chelating complexes of Al (possibly with organic acids) inside root cells (symplast) to detoxify the In Al-accumulator species, Al that penetrates the cells. Until now, Al detoxification with organic plasma membrane of root cells is immediately acids in plant tissue has been observed mainly in transferred to shoots. In Brachiaria, however, Al Al-accumulator species. accumulated only in roots and Al concentrations in shoots were kept very low (Table 18). In This study provides direct evidence for the first shoots of buckwheat, an Al-accumulator species, time that organic acids inside root tissues help Al-ligand complexes are mostly isolated in detoxify Al in non-accumulator species of Al. In vacuoles. Al in the Brachiaria hybrid may also Al non-accumulator species, Al is assumed to be compartmentalize in vacuoles and, thus, not be detoxified outside the root cells, in the rhizosphere translocated from roots to shoots. and/or root apoplast, by exuded organic acids. However, the Brachiaria hybrid, an Al non- Of the three species tested in this study, the accumulator, accumulates large amounts of Al in Brachiaria hybrid and A. pintoi were extremely the root symplast (Figure 19), probably in the tolerant of low P stress (Figure 18). The vacuoles. Thus, in some cases, Al accumulation Brachiaria hybrid grows well under P starvation, and detoxification in roots involve the same despite very low P concentrations in its leaves mechanism in Al-accumulators and non- (Table 18). This suggests that Brachiaria can accumulators. Differences in characteristics of reuse P more efficiently. In A. gayanus, growth Al accumulation in shoots between Al- under the -P treatment was inferior to that of the accumulators and non-accumulators may be control treatment, probably because of a higher caused by the existence of ligands for controlling relative growth rate (RGR). Based on values of translocation from roots to shoots. P-use efficiency (the reciprocal of P concentration in plant), both A. gayanus and the Brachiaria hybrid appear to be tolerant of P Conclusions starvation (Table 18). Higher P-use efficiency was observed before in Brachiaria humidicola We tested how Al toxicity and P starvation affect (dictyneura) than in forage legumes, including growth, concentrations of minerals and organic A. pintoi, in acid Oxisols. acids, and acid phosphatase activity in root exudates of Brachiaria hybrid, Andropogon gayanus and Arachis pintoi. The two tropical 66 grasses tolerated high levels of Al toxicity and P are probably adapted to P starvation through high starvation, with the Brachiaria hybrid P-use efficiency. These experiments provide the maintaining very low levels of Al concentration in first evidence we know of that organic acids shoots. 27Al NMR analysis revealed that, in within root tissue help detoxify Al in non- Brachiaria hybrid, Al makes complexes with accumulator species such as the Brachiaria some ligands in root symplast. The three forages hybrid. 3.1.1.2 Screening of Brachiaria hybrids for resistance to aluminum Contributors: I. M. Rao, J. W. Miles, R. Garcia and J. Ricaurte (CIAT) Rationale were screened initially using 200 µM Al treatment to eliminate the Al sensitive hybrids. From this For the last three years, we have implemented initial screening, a total of 124 sexual hybrids screening procedure to identify Al-resistant were selected to test under both treatments of Brachiaria hybrids that were preselected for with Al (6 experiments) and without Al (3 spittlebug resistance. In 2002, we have identified experiments) in solution. Out of these 124 sexual 2 sexual hybrids (SX 01NO3178 and hybrids, data were obtained for 86 sexuals for SX01NO7249) and one apomictic hybrid both treatments. Mean values from all the (BR99NO/4132) with greater level of Al experiments (ranging from 3 to 20 observations) resistance than that of the sexual parent, BRUZ/ are reported. Stem cuttings of hybrids and checks 44-02. Last year, we have identified 2 hybrids were rooted in a low ionic strength nutrient ((BR02NO1372 and BR02NO1621) with greater solution in the glasshouse for 9 days. Equal level of Al resistance than that of the most numbers of stem cuttings with about 5 cm long hybrids generated from the Brachiaria breeding roots were transferred into a solution containing program. With the partial support of BMZ-GTZ 200 µM CaCl2 pH 4.2 (reference treatment) and of Germany and Papalotla (seed company) of a solution containing 200 µM CaCl2 and 200 µM Mexico to the Brachiaria improvement project, AlCl3 pH 4.2 (Al treatment). The solutions were this year we evaluated Al resistance of a sexual changed every second day to minimize pH drifts. population of 745 hybrids along with 14 checks. At harvest on day 21 after transfer, the dry The increase in Al resistance of the sexual weight of stems was measured. Roots were hybrids has been very marked compared with the stained and scanned on a flatbed scanner. Image sexual population of 2001. analysis software (WinRHIZO) was used to determine root length and average root diameter. Materials and Methods A total of 745 sexual hybrids generated from Results and Discussion 2003 population together with 14 checks including 3 parents (B. decumbens CIAT 606, B. As reported for the past 3 years, higher values of brizantha CIAT 6294 and B. ruziziensis 44-02) total root length per plant and lower values of were included for evaluation of Al resistance. All mean root diameter after exposure to 21 days the new sexual hybrids were screened for with or without toxic level of Al in solution spittlebug resistance (C. Cardona, personal indicate that the parent B. decumbens CIAT 606 communication; also see Output 2 and Activity is outstanding in its level of Al resistance (Figures 2.2 of this report). All the hybrids that rooted well 20 and 21). 67 12 1.0 0 µM Al; LSD = 0.081 0 µM Al; LSD0.05 = 2.55 0.05 200 µM Al; LSD0.05 = 0.093 200 µM Al; LSD 0.05 = 0.84 0.8 9 Parents Hybrids apomictic Parents Hybrids sexual apomictic sexual 0.6 6 0.4 3 0.2 0 0.0 06 94 -02 6 2 o 1 72 21 6 7 1 T 6 4 at /13 /16 /08 4 6 8 6 4 2 1 2 1 6 7 1 T 4 l /23 /08 T60 29 -0 o 37 62 84 36 6 88 IA A z u O O O O O IA AT z 4 4 ulat /1 /1 /0 /2 /0 I O O c C C ru M N 02 2N 3N 3N 3N C CI ru M 2N 2N 3NO NO NO de riz B 3 3 B b BR BR0 X0 X0 X0 ec iz B 0 0 X0 X0 X0 B S S S Bd br BR BR S S S Genotypes B Genotypes Figure 20. Al resistant apomictic and sexual hybrids of Brachiaria based on total root length and mean root diameter. Total root length and mean root diameter were measured after exposure to 0 or 200 ìM AlCl3 with 200 ìM CaCl2 (pH 4.2) for 21 days. Among the 745 sexual hybrids and checks tested, root length of the three sexual hybrids, both in the 3 sexual hybrids (SX03NO/0846, SX03NO/2367, presence or absence of Al, was markedly SX03NO/0881) and 3 apomictic hybrids (Mulato, superior to the sexual parent, B. ruziziensis BR02NO1372 and BR02NO1621) showed (Figure 20). Among the hybrids and the checks greater level of Al resistance based on total root tested, B. humidicola (dictyoneura) CIAT 6133 length per plant (Figure 20; Table 19). Among showed the lowest values of mean root diameter these promising hybrids, BR02NO1372 showed in the presence or absence of Al in solution greater fine root development than CIAT 606 in (Table 19; Figure 22). This could be a desirable the absence of Al in solution (Figure 20). Total attribute for persistence under infertile acid soil conditions. 300 300 MEAN = 351.9 MEAN = 351.9 1372 1372 250 2367 250 2367 200 36061 2365 200 36061 2365 606 0846 606 0846 6133 1079 0881 11682210 0893 0881 1820 6133 1079 1621 0893 1167 1167 150 02942303 02942303 1877 22525424 3178 150 1877 22525424 3178 1225 2822 3335 1225 2822 0923 3335 11033211 0946 2296 3096 MEAN = 119.9 0923 11033211 0946 2296 3096 MEAN = 119.9 096900310239 2203076 2768 0969 082602968 003102 2203076629267868 0273 100 027339 0820 100 0194 0961 0903 0194 0961 42607903 2045 4267 3374 0378 2045 3374 0378 3930 093284348951 3930 093284348951 1938 2086294 1938 2086294 07059 0810 07059 0810 50 44-02 23909254 0581 50 44-02 23909254 0581 S046-024 S046-024 0508 0827 0508 0827 0 0 0 100 200 300 400 500 600 700 800 0 100 200 300 400 500 600 700 800 Root length without Al (cm plant-1) Root length without Al (cm plant-1) Figure 21. Relationship between total root length and Figure 22. Relationship between total root length with Al mean root diameter of 78 genotypes of Brachiaria with and total root length without Al of 78 genotypes of presence or absence of aluminum in solution. Gentoypes Brachiaria with presence or absence of aluminum in that develop finer root system were identified in the upper solution. Genotypes that develop greater root length were box of the left hand side. identified in the upper box of the right hand side. 68 Root length with Al (cm plant-1) Total root length (m plant-1) Root length with Al (cm plant-1) Mean root diameter (mm) Table 19. Root length and mean root diameter of Brachiaria sexual hybrids evaluated with (200 µM Al) and without Al (0 µM Al) in solution in comparison with their parents and other checks. Root length Root diameter Root length Root diameter (m plant-1) (mm) (m plant-1) (mm) 0 µM 200 µM 0 µM 200 µM 0 µM 200 µM 0 µM 200 µM Genotypes Al Al Al Al Genotypes Al Al Al Al Sexual hybrids Sexual hybrids SX03NO/0846 5.115 2.575 0.349 0.465 SX03NO/2087 2.168 1.337 0.505 0.658 SX03NO/2367 4.756 2.365 0.346 0.490 SX03NO/0969 2.135 1.294 0.399 0.561 SX03NO/0881 3.912 2.314 0.474 0.564 SX03NO/0194 2.191 1.292 0.420 0.555 SX03NO/2433 4.323 2.170 0.314 0.448 SX03NO/1225 2.294 1.292 0.430 0.476 SX03NO/2365 4.976 2.170 0.332 0.456 SX03NO/2376 3.337 1.286 0.397 0.590 SX03NO/1167 5.685 2.069 0.349 0.505 SX03NO/3374 4.679 1.263 0.367 0.609 SX03NO/1079 3.192 1.992 0.351 0.479 SX03NO/1877 3.231 1.249 0.480 0.692 SX03NO/0311 3.573 1.969 0.350 0.483 SX03NO/0031 2.481 1.237 0.463 0.535 SX03NO/2968 3.718 1.966 0.397 0.578 SX03NO/0239 2.825 1.185 0.443 0.664 SX03NO/2424 5.109 1.855 0.389 0.521 SX03NO/2822 3.459 1.181 0.393 0.576 SX03NO/0893 5.765 1.804 0.373 0.555 SX03NO/4267 4.026 1.180 0.404 0.543 SX03NO/2255 4.896 1.797 0.390 0.584 SX03NO/0820 3.418 1.108 0.393 0.516 SX03NO/1820 4.404 1.734 0.341 0.419 SX03NO/0508 1.565 1.096 0.454 0.550 SX03NO/0273 2.731 1.726 0.394 0.502 SX03NO/1938 3.371 1.065 0.417 0.649 SX03NO/0903 4.241 1.703 0.473 0.457 SX03NO/3485 2.415 1.025 0.478 0.698 SX03NO/2296 4.839 1.634 0.317 0.532 SX03NO/3024 2.034 1.009 0.400 0.575 SX03NO/1132 3.429 1.634 0.378 0.494 SX03NO/0581 2.953 0.977 0.390 0.636 SX03NO/0961 2.946 1.593 0.386 0.554 SX03NO/0984 2.308 0.944 0.419 0.548 SX03NO/2995 2.005 1.578 0.426 0.636 SX03NO/0827 2.033 0.764 0.446 0.656 SX03NO/3096 4.983 1.535 0.320 0.475 Parents SX03NO/2391 2.441 1.509 0.454 0.580 B.decumbens CIAT606 4.625 2.496 0.259 0.360 SX03NO/2006 3.305 1.486 0.375 0.532 B. brizantha CIAT6294 2.444 0.979 0.348 0.570 SX03NO/2768 3.855 1.480 0.404 0.562 B.ruziziensis 44-02 1.310 0.705 0.461 0.634 SX03NO/0946 4.060 1.454 0.405 0.549 Checks SX03NO/0059 2.428 1.450 0.357 0.528 BR02NO/1372 7.557 2.395 0.313 0.383 SX03NO/2045 2.749 1.449 0.360 0.539 Mulato 3.688 2.361 0.441 0.525 SX03NO/0378 5.153 1.440 0.396 0.702 BR02NO/1621 4.377 1.980 0.321 0.458 B.humidicola (dictyoneura) SX03NO/0920 4.249 1.400 0.357 0.554 CIAT6133 2.732 1.922 0.221 0.287 SX03NO/3930 1.770 1.399 0.555 0.724 SX01NO/3178 5.794 1.341 0.394 0.653 FM9503-S046-024 (CIAT 36087-Mulato SX03NO/0810 3.079 1.398 0.371 0.610 2) 1.609 0.934 0.512 0.765 SX03NO/0923 2.426 1.356 0.436 0.579 Means 3.519 1.537 0.395 0.553 SX03NO/3335 3.973 1.338 0.426 0.625 LSD (P<0.05) 2.548 1.077 0.081 0.093 attribute for persistence under infertile acid soil the 78 genotypes from 352 to 120 cm plant-1 conditions. (Figure 22). Relationship between total root Relationship between root length and mean root length without Al and with Al in solution showed diameter with Al in solution showed that several that several apomictic and sexual hybrids were sexual hybrids were superior to the sexual parent, superior to the Al resistant parent, B. decumbens B.ruziziensis 44-02 (Figure 21). Exposure to Al CIAT 606 in the absence of Al in solution (Figure decreased the mean value of total root length of 22). The greater root vigor of these hybrids could 69 The sexual hybrids SX03NO/846 and SX03NO/ SX03/2694 was more sensitive to Al than the 0881 were found to be spittlebug resistant while sexual parent, B. ruziziensis 44-02. The SX03NO/0311 was resistant to Rhizoctonia foliar promising sexual hybrids that combine spittlebug blight. Two sexual hybrids that were found to be resistance with Al resistance and other desirable spittlebug resistant, SX03/2483 and SX03/1820, attributes are being used in recurrent selection to were also found to be moderately resistant to Al generate superior hybrids of Brachiaria. stress. Another spittlebug resistant sexual hybrid, 3.1.1.3 Field evaluation of most promising hybrids of Brachiaria in the Llanos of Colombia Contributors: I. M. Rao, J. Miles, C. Plazas and J. Ricaurte (CIAT) Rationale one spittlebug resistant genetic recombinant, FM 9503-S046-024 (CIAT 36087—Mulato 2) and one Evaluation of a large number of Brachiaria parent (CIAT 6294) (Table 20). With high initial hybrids for their resistance to spittlebug and fertilizer application also these two genotypes adaptation to infertile acid soils resulted in were outstanding in live forage yield (Figure 23). identification of a few promising Brachiaria Among the 4 hybrids tested, Mulato 2 was hybrids. We selected 4 of these hybrids for outstanding in its adaptation to low initial fertilizer further field-testing in comparison with their application. It is important to note that CIAT 6294 parents. The main objective was to evaluate had greater amount of dead biomass and stem growth and persistence with low nutrient supply biomass under low fertilizer application in soil at Matazul farm of the altillanura. (Figure 23). Materials and Methods Table 20. Correlation coefficients (r) between green forage yield (t/ha) and other shoot traits of Brachiaria genotypes A field trial was established at Matazul farm on grown with low or high initial fertilizer application in a sandy 31 May of 2001. The trial included 4 Brachiaria loam oxisol in Matazul, Colombia. hybrids (BR98NO/1251; BR99NO/4015; Low High BR99NO/4132; FM9503-S046-024) along with 2 Shoot traits fertilizer fertilizer parents (B. decumbens CIAT 606 and B. Total (live + dead) shoot biomass (t/ha) 0.84*** 0.91*** brizantha CIAT 6294). The trial was planted as Dead shoot biomass (t/ha) 0.54 0.76*** a randomized block in split-plot arrangement with Leaf biomass (t/ha) 0.94*** 0.94*** two levels of initial fertilizer application (low: kg/ Stem biomass (t/ha) 0.69*** 0.68*** ha of 20P, 20K, 33Ca, 14 Mg, 10S; and high: 80N, Leaf N content (%) -0.24 -0.35 50P, 100K, 66Ca, 28Mg, 20S and micronutrients) Leaf P content (%) 0.21 0.06 as main plots and genotypes as sub-plots with 3 Leaf TNC content (mg g-1) -0.25 -0.31 replications. The plot size was 5 x 2 m. A number of plant attributes including forage yield, dry Leaf ash content (%) 0.05 0.01 matter distribution, nutrient (N, P, K, Ca and Mg) Stem N content (%) -0.45 0.79*** uptake, leaf and stem total nonstructural Stem P content (%) -0.43 -0.31 -1 carbohydrate (TNC) content and leaf and stem Stem TNC content (mg g ) 0.45 0.12 ash (mineral) content were measured at 30 Stem ash content (%) -0.12 0.05 months after establishment (November 2003). Shoot N uptake (kg/ha) 0.93*** 0.85*** Shoot P uptake (kg/ha) 0.91*** 0.93*** Results and Discussion Shoot K uptake (kg/ha) 0.76*** 0.78*** Shoot Ca uptake (kg/ha) 0.85*** 0.81*** After 30 months of establishment, forage yield Shoot Mg uptake (kg/ha) 0.84*** 0.89*** with low fertilizer application was greater with *, **, *** Significant at the 0.05, 0.01 and 0.001 probability levels, respectively 70 As observed last year, results on shoot nutrient nutrient acquisition with low initial fertilizer uptake, particularly Ca and Mg, indicated that the application contributed to superior performance hybrid, Mulato 2 was superior to CIAT 606 under (Table 20). No significant correlations were low fertilizer application (Figure 24). Nutrient found between live forage yield and leaf and acquisition by Mulato 2 was also greater than the stem TNC or ash contents. rest of the hybrids with high initial fertilization. The performance of the 4 hybrids in comparison These results are consistent with the results with two parents with maintenance fertilizer reported last year from the same experiment. application will be monitored for next year in Correlation coefficients between live forage yield terms of forage yield and nutrient acquisition. and other plant attributes indicated that greater N ; LSD0.05 = 4.8 NS P ; LSD0.05 = 1.2 NS 60 K ; LSD0.05 = 4.1 NS 5000 Dead; LSD0.05 = 657 NS Ca ; LSD0.05 = 1.9 NS Mg ; LSD0.05 = 1.9 NS 4000 Stems ; LSD0.05 = 182 NS Leaves ; LSD = 213 318 HIGH FERTILIZER LOW FERTILIZER 0.05 40 3000 HIGH FERTILIZER LOW FERTILIZER 2000 20 1000 0 60 6 25 1 15 32 24 94 06 51 15 32 24 94 1 40 41 46 62 6 12 40 41 46 62 0 Genotype 06 51 15 32 24 94 06 1 5 6 2 0 1 6 2 6 25 01 13 2 24 94 1 4 4 4 6 1 4 4 46 62 Genotypes Figure 23. Genotypic variation as influenced by fertilizer Figure 24. Genotypic variation as influenced by fertilizer application in shoot biomass production (forage yield) of application in nutrient uptake (N, P, K, Ca and Mg) of two two parents (CIAT 606, 6294) and four genetic parents (CIAT 606, 6294) and four genetic recombinants recombinants (1251, 4015, 4132, 4624) of Brachiaria (1251, 4015, 4132, 4624) of Brachiaria grown in a sandy grown in a sandy loam oxisol at Matazul, Colombia. Plant loam oxisol at Matazul, Colombia. Plant attributes were attributes were measured at 30 months after establishment measured at 30 months after establishment (November (November 2003). LSD values are at the 0.05 probability 2003). LSD values are at the 0.05 probability level. NS = level. NS = not significant. not significant. 3.1.2 Participatory evaluation of Brachiaria accesions/hybrids in comparison with commercial cultivars in Nicaragua Contributors: A. Schmidt, C. Davis, M. Peters, J. Miles and I. M. Rao (CIAT) Rationale selecting promising forage germplasm, because their selection criteria are not necessarily the As part of the BMZ-GTZ project on developing same as those of researchers. Thus the main aluminum resistant Brachiaria hybrids, in 2002 objective of participatory evaluation was to we initiated field studies in Nicaragua for expose the promising hybrids to farmers and evaluation of new hybrids of Brachiaria along generate information on farmer selection criteria. with commercial Brachiaria cultivars with This information is highly useful to Brachiaria farmer participation. The opinion of farmers is improvement program to incorporate farmer very important in the process of identifying and perspectives on Brachiaria ideotypes for multiple uses in crop-livestock systems. 71 Shoot biomass (Kg ha-1) Shoot nutrient uptakes (kg ha-1) Materials and Methods Fertilization levels were adjusted to soil analysis results. Agronomic and participatory evaluations The experiment site was chosen in Ubú Norte with farmers were conducted in August 2003 (12° 58’ 44" N, 84° 54’ 23" E, 261 masl) in the (max. rainfall.), March 2004 (min. rainfall) and Region Autónoma del Atlántico Sur (RAAS) July 2004 (max. precip) in each case 6 weeks where acid soils are predominant. Soil after standardization cut (Figure 25). characteristics and rainfall distribution from January to December are presented in Table 21 and Figure 24, respectively. A total of 14 Brachiaria accessions and hybrids (CIAT No. 606, 654, 679, 6133, 6780, 16322, 26110, 26124, 26318, 26646, 26990, 36061, 36062, and “Mixe” (no CIAT No.) were sown in three replicates in a split-plot design with fertility levels as main plots and genotypes as subplots. Site preparation was initiated in September and plots (5x4m) were sown early October 2002 upon the beginning of the second rainy season. Table 21. Soil characteristics of experimental site Ubú Norte. Photo 14. Brachiaria accessions and hybrids tested on acid soils in Ubú Norte, Nicaragua. SOM pH P-Bray K Ca Mg (%) (ppm) (meq) (meq) (meq) Na (meq) Results and Discussion 6.61 5.66 3.22 0.55 5.50 2.42 0.09 Evaluation 2003: Results from agronomic evaluations showed no significant fertilizer effect on plant height, soil cover and dry matter yield The establishment of grasses in the plots was (Table 22). Significant differences were found heavily affected by unusual high precipitation, among accessions/ hybrids, but no fertilizer x leaving Ubú with 1/3 more precipitation as the accession/hybrid interactions were detected. average of the last 20 years and causing floods in the area. Most of the plots established did not Data obtained in 2003 showed Brachiaria germinate and thus plots were replanted in May brizantha cv. Toledo (CIAT 26110) and 2003 (Photo 14). Two fertilization treatments Brachiaria hybrid cv. Mulato (CIAT 36061) as (high/low) were applied upon plot establishment. the top ranking accessions with regard to dry matter yield, followed by B. brizantha CIAT 26124 and B. brizantha cv. Marandú (CIAT 450 6780). The lowest yields were obtained with 400 350 B. humidicola CIAT 679 and B. humidicola 300 (dictyoneura) CIAT 6133 due to their slow 250 200 establishment. Best soil cover was observed in 150 plots with B. hybrid cv. Mulato (CIAT 36061), 100 B. brizantha “Mixe”, B. brizantha CIAT 26124, 50 0 and B. brizantha cv. Marandú (CIAT 6780). 1 2 3 4 5 6 7 8 9 10 11 12 Plant height was greater with B. brizantha cv. Months Toledo (CIAT 26110) and B. brizantha cv. Marandú (CIAT 6780). B. humidicola CIAT 679 Figure 25. Rainfall distribution at the experimental site of did not recover well from the standardization cut Ubú Norte, Nicaragua. and was discarded from the experiment in 2004. 72 Precipitation (mm) Table 22. Plant height, soil cover and DM yield of 14 Brachiaria accessions and hybrids in Ubú Norte, Nicaragua, 2003-2004. Fertility 2003 2004 Levels/ Max Min Max Accessions Plant Soil DM Plant Soil DM Plant Soil DM height Cover yield height Cover yield height Cover yield (cm) (%) (g/m2) (cm) (%) (g/m2) (cm) (%) (g/m2) Fert. Level - High * 88 64 802 56 73 310 76 75 338 - Low 82 60 747 54 69 283 71 67 303 No. CIAT 606 90 76 578 56 73 244 73 82 329 654 71 48 538 39 52 239 55 37 188 679 32 17 - - - - - - - 6133 55 30 339 37 46 234 55 65 347 6780 126 89 1000 40 78 171 65 91 244 16322 98 88 646 72 84 252 94 88 348 26110 108 65 1682 72 70 375 76 64 363 26124 99 89 1239 64 67 264 75 66 320 26318 90 34 854 62 47 359 61 17 342 26646 75 48 505 65 70 305 76 76 275 26990 87 60 695 64 53 357 73 48 277 36061 99 93 1546 52 74 305 69 81 308 36062 70 32 440 52 76 322 54 49 253 “Mixe” 89 90 775 56 81 331 77 86 355 Mean Acc. 85 61 834 56 67 289 69 65 304 LSD (0.05) 31 30 645 13 24 190 21 30 263 (*Differences at fertilization levels were not significant at p < 0.05) Results obtained during the period of low rainfall 16322, and B. brizantha CIAT 26318. showed no fertilizer effect on agronomic Brachiaria hybrid cv. Mulato showed good soil parameters. Best performing accessions were cover and above average dry matter yield. B. brizantha cv. Toledo (CIAT 26110), B. brizantha cv. Marandú (CIAT 6780), widely B. brizantha CIAT 26318, B. brizantha CIAT planted in the area of Ubú Norte, established well 26990, and B. brizantha “Mixe”. In early 2004, and showed good adaptation, soil cover and Rhizoctonia foliar blight detected in Brachiaria height, especially in the first year, but in hybrid cv. Mulato, had a negative effect on dry subsequent evaluations dry matter yield was matter production. Plots of B. brizantha CIAT below average. Apart from the Rhizoctonia foliar 16322 developed very well in the period of low blight attacks no significant or limiting pest and rainfall and outperformed other accessions in disease incidents were recorded throughout the plant height and soil cover, indicating good experiment. adaptation to prevailing environmental conditions. Dry matter yield, however, was below average Participatory evaluations with farmers: Prior because of the very fine leaf structure of this to agronomic data collection, farmer groups accession. evaluated the plots in accordance with their own criteria. Their preference rankings are Evaluation 2004: Due to severe reduction in summarized in Table 23. rainfall during the first semester of 2004 in Nicaragua, dry matter yields were significantly The main criteria applied by farmers throughout lower than in 2003. The highest yields were the experiment were: plant height, soil cover, recorded for B. brizantha cv. Toledo (CIAT foliage production, and ease of establishment, leaf 26110), B. brizantha “Mixe”, B. brizantha CIAT size and color. While the high ranking of 73 Table 23. Preference ranking of Brachiaria accessions and hybrids at Ubú Norte, Nicaragua (2003-2004) (Max = Maximun rainfall; Min = Minimum rainfall). Max 2003 Min 2003 Max 2004 B. brizantha cv Marandú B. brizantha cv. Toledo B. brizantha cv. Toledo B. brizantha cv. Toledo B. brizantha cv Marandú B. hybrid cv. Mulato B. hybrid cv. Mulato B. brizantha “Mixe” B. brizantha cv Marandú B. brizantha CIAT 26990 B. hybrid cv. Mulato B. brizantha “Mixe” B. brizantha CIAT 26124 B. brizantha CIAT 16322 B. brizantha CIAT 16322 B. brizantha CIAT 16322 B. brizantha CIAT 26646 B. brizantha CIAT 26318 B. brizantha cv. Marandú (CIAT 6780) was difficulties to keep track of all plots. In the future, somewhat expected, the cultivar is known in the smaller number of materials should be presented area for years and well-adapted to the prevailing to farmers in order to avoid confusion, especially conditions, B. brizantha cv. Toledo (CIAT with farmers who do not have experiences with 26110), Brachiaria hybrid cv. Mulato (CIAT ranking of forage germplasm. 36061) were preferred because of their abundant foliage with bright green leaves. The fact that Although the chosen experimental site both materials were sold on the seed market represented an acid soil region of Nicaragua, soil could have influenced the ranking. Accessions pH and aluminum contents were not limiting such as B. brizantha CIAT 26990, 26124, 16322 factors for the tested accessions/hybrids. Since were often classified as less productive because real drought conditions did not prevail throughout of their leaf size. Most other materials were rated the experiment, hybrids such as cv. Mulato could low due to low soil cover or plant height. not express their full potential, but did show susceptibility to Rhizoctonia foliar blight. As mentioned earlier, farmer rankings could have Nevertheless farmers seem to appreciate the been influenced to some extent by the active Mulato hybrid, since seed is increasingly available presence of a livestock project in the area to them. Brachiaria brizantha accession CIAT promoting B. brizantha cv. Toledo and B. hybrid 16322 and Brachiaria brizantha “Mixe” are cv. Mulato. The main difficulty during the high-potential materials due to their excellent participatory evaluation was the large number of adaptation and growth. Their small leafs might be accessions to be ranked. Some farmers had a decisive factor for their adoption by farmers. 3.1.3 Edaphic adaptation of Arachis pintoi 3.1.3.1 Field evaluation of most promising accessions of Arachis pintoi in the Llanos of Colombia Contributors: I. M. Rao, M. Peters, C. Plazas and J. Ricaurte (CIAT) Rationale close to La Libertad (CORPOICA Experimental Station) and the soils in this region are relatively Based on field studies conducted in Caqueta, more fertile than in the Altillanura. The main Colombia and the data collected from objective of this work was to identify plant multilocational evaluation, we have assembled a attributes related to superior adaptation of the set of 8 genotypes for further testing at Piedmont most promising accessions for the llanos of in the Llanos of Colombia. The site in Piedmont is Colombia. 74 Materials and Methods Shoot ash (mineral) content was higher in CIAT 22172 under both low and high initial fertilizer A field trial was established in Piedmont in May, application (Tables 24 and 25). 2001 as monoculture. The trial included 8 accessions of Arachis pintoi (CIAT 17434; The superior performance of CIAT 18744 was 18744; 18747; 18748; 18751; 22159; 22160 and associated with greater TNC (Total Non- 22172). The trial was planted as randomized structural Carbohydrates) and lower ash content block in split-plot arrangement with two levels of in the shoot biomass under low initial fertilization initial fertilizer application (low: kg/ha of 20P, indicating greater nutrient use efficiency for 20K, 33Ca, 14 Mg, 10S; and high: 80N, 50P, producing green forage. Measurements of 100K, 66Ca, 28Mg, 20S and micronutrients) as nutrient uptake indicated that CIAT 22160 was main plots and genotypes as sub-plots with 3 outstanding with high initial fertilizer application replications. Genotypic differences in agronomic performance were determined at 42 months after Table 24 establishment (December 2004) at low and high . Shoot biomass, cover, vigor and shoot total nonstructural carbohydrates (TNC) content and shoot ash initial fertilizer application. Maintenance fertilizer content in 8 Arachis pintoi accessions evaluated with high initial at half the level of initial applications was applied fertilizer application to low fertility acid soils at La Libertad (December 2003). at 26 months after establishment (August 2003). Shoot Shoot biomass Cover TNC Shoot ash Results Accession (kg ha-1) (%) Vigor (mg g-1) (%) 17434 2384 98 3.0 79.3 8.073 At 42 months after establishment of the trial, the 18744 2853 100 4.3 79.3 7.867 response in terms of shoot biomass production 18747 3339 100 4.3 56.0 8.367 with fertilizer application was greater with CIAT 18748 2959 100 4.0 81.3 7.693 22160 (Figure 26). Overall the performance of 18751 3095 100 4.0 102.1 7.527 CIAT 18744 and CIAT 18751 and CIAT 22159 22159 2193 100 4.0 63.1 7.967 was better than the other accessions. These 22160 4464 100 4.3 73.2 7.600 results are consistent with the observations made 22172 3240 98 4.0 67.1 8.400 last year. Shoot TNC content was markedly Mean 3066 99.5 4.0 75.2 7.937 greater in CIAT 18751 under both low and high LSD(P>0.05) 1331 NS 0.66 NS NS initial fertilizer application (Tables 24 and 25). 5000 Table 25. Shoot biomass, cover, vigor and total nonstructural High F. ; LSD = 1331 carbohydrates and ash content in 8 Arachis pintoi accessions 0.05 Low F. ; LSD0.05 = 920 evaluated with low initial fertilizer application to low fertilizer on 4000 low fertility acid soils at La Libertad (December 2003). 3000 Shoot Shoot biomass Cover Vigor TNC Shoot ash Accession (kg ha-1) (%) (mg g-1) (%) 2000 17434 2276 100 3.3 90.7 8.033 1000 18744 3114 100 4.0 135.9 7.240 18747 2088 98 3.7 64.2 7.673 0 34 44 47 8 4 7 7 74 75 1 15 9 60 721 1 18748 2878 97 3.3 84.3 7.927 17 18 18 18 18 22 22 22 18751 2525 100 4.0 125.1 7.040 Genotypes 22159 3028 100 4.0 69.7 7.560 Figure 26. Genotypic variation in forage yield of 8 22160 3021 100 4.3 100.8 7.373 accessions of Arachis pintoi at 42 months after establishment (December, 2003) in forage yield (kg/ha) as 22172 2043 97 3.0 65.1 8.187 influenced by initial level of fertilizer application to a clay Mean 2622 99 3.7 92.0 7.629 loam oxisol at La Libertad, Piedmont. LSD (P>0.05) 920 NS 0.71 NS 0.611 75 Shoot biomass (kg ha-1) while CIAT 22159 was superior with low initial Table 27. Shoot N, P, K, Ca and Mg uptake of 8 fertilizer application (Tables 26 and 27). Arachis pintoi accessions evaluated with low initial fertilizer application on low fertility acid soils at La Libertad (December 2003). Correlation coefficients between green forage yield and other shoot attributes indicated that the Accession Shoot nutrient uptake (kg ha-1) superior performance with low initial fertilizer N P K Ca Mg application was associated with lower level of Ca 17434 60.0 7.3 26.7 52.6 23.9 content in the shoot tissue (Table 28). 18744 79.8 9.7 31.5 55.8 36.8 18747 61.4 7.1 20.2 48.3 19.8 Results from this field study indicated that after 4 18748 78.0 8.4 19.9 70.0 32.1 years, the Arachis pintoi accessions CIAT 18751 64.3 8.2 27.5 47.2 21.8 18744, 18751 and 22159 are superior to the 22159 91.1 11.7 35.6 61.7 29.9 commercial cultivar (CIAT 17434) in terms of 22160 75.2 9.0 21.2 62.3 35.2 persistence with low amounts of initial fertilizer 22172 56.1 5.4 15.7 55.5 23.1 application. Mean 70.7 8.3 24.8 56.7 27.8 LSD (P>0.05) 24.7 3.0 14.0 NS 12.4 Table 26. Shoot N, P, K, Ca and Mg uptake of 8 Arachis pintoi accessions evaluated with high initial fertilizer application to low fertility acid soils at La Table 28. Correlation coefficients (r) between green Libertad (December 2003). forage yield (t/ha) and other shoot traits of Arachis pintoi genotypes grown with low or high initial Accession Shoot nutrient uptake (kg ha-1) fertilizer application in a low fertility Oxisol in La Libertad, Colombia. N P K Ca Mg 17434 65.5 7.5 32.5 51.0 23.3 Shoot traits Low High fertilizer 18744 76.6 8.5 35.8 55.3 27.9 fertilizer Soil cover (%) 0.20 -0.05 18747 88.6 10.1 46.9 84.8 34.2 Vigor (visual scale) 0.44* 0.55** 18748 75.2 7.9 34.2 68.3 28.5 Shoot N content (%) -0.13 -0.49** Shoot P content (%) -0.06 -0.58** 18751 80.0 8.0 31.0 70.7 28.0 Shoot K content (%) 0.11 0.28 22159 63.2 7.9 29.4 47.8 21.4 Shoot Ca content (%) -0.45* 0.04 22160 113.6 12.0 59.6 90.2 43.9 Shoot Mg content (%) 0.13 -0.17 Shoot TNC content (mg g-1) 0.05 -0.01 22172 90.6 10.0 47.0 75.5 31.5 Shoot ash content (%) -0.29 -0.02 Mean 81.7 9.0 39.5 67.9 29.8 *, **, *** Significant at the 0.05, 0.01 and 0.001 probability levels, respectively LSD(P>0.05) 28.8 3.1 23.5 NS 13.6 3.1.3.2 Genotypic variation in Arachis pintoi for tolerance to low P supply Contributors: N. Castañeda, N. Claassen (University of Goettingen, Germany) and I. M. Rao (CIAT) Rationale among the accessions CIAT 17434, CIAT 18744 and CIAT 22172. Therefore this P acquisition Several studies conducted in Colombia as well as was related to an increase of P soil solution in Goettingen confirmed that Arachis pintoi concentration with CIAT 18744 i.e. to P genotypes showed a different growth potential mobilization in the rhizosphere. These results when grown in Ultisols or Osixols with extremely indicated that the rapid establishment as well as low P availability. A growth chamber study the sustained yield of CIAT 18744 and 22172 was conducted last year in Goettingen indicated due to their great ability to acquire P from P- significant genotypic difference in P acquisition deficient soil per unit root length. There was no 76 significant difference among the genotypes in with 3 kg air dry sand (size 3-5 mm) in which P their ability to utilize acquired P. This year, we were removed using 5% HCl solution and then continued our efforts to determine the washing the sand till getting the same pH of physiological basis of differences in P influx distilled water. The sand compartment was between the 2 accessions, CIAT 17434 and watered three times per day with nutrient solution 18744. without P. Its composition (M) was: Ca (NO3) 2 5.0×10-3, KCl 5.0×10-4, K2SO4 3.5×10-3, MgSO4 Material and Methods 2.5×10-3, H BO 5.0×10-5 3 3 , MnSO4 5.0×10-6, ZnSO 2.5×10-6, CuSO -6 4 4 1.0×10 , (NH4) 6Mo7O24 Plant cultivation - Germinated seedlings in sand 5.0×10-8, and FeEDTA 1.0×10-4. This of Arachis pintoi CIAT 17434 and 18744 were compartment was allowed to percolate water pre-cultured in −P nutrient solution for 5 days and through a hole in the bottom which were later then the taproot was excised to develop lateral leached the root exudates. Two levels of P supply roots for additional 15 days till the lateral roots (0, and 1000 mg kg-1) as Ca (H2PO4) 2.H2O were had reached about 5 cm long. The 20-day-old given only in the soil compartments. Basal nutrients seedlings were transplanted to pots divided in were applied (mg kg-1) every 30 days to each soil three compartments (Figure 27) and placed on compartment: 50 K as K2SO4, 40 Mg as MgSO4, the dividing walls and their roots split so that 50% 0.2 B as H3BO3, 0.1 Mo as (NH4) 6Mo7O24 and 100 grew in soil and 50% in sand. N as Ca (NO3) 2. The soil surface in each soil compartment was covered with a layer of quartz Each one of the outer compartments was filled sand (1 to 2 cm) to avoid the formation of a with 2 kg of air dry clay loam fossil Oxisol (clay superficial crust due to the watering. Two weeks 50%, organic carbon 0.35%, pHCaCl2 5.1, pH before transplanting, water was added to get H2O 5.2, P-CAL 0.4 mg /100 g soil and P-Bray II 1.4 moisture content of 25% w/w. One pot for each P mg/100 g soil, Fe/Al-P 788 mg kg-1 and Ca-P 330 treatment was left unplanted to measure soil mg kg-1; in soil solution pH 4.9 and 0.1 µM P) moisture evaporation losses and as control of P from Lich in the Vogelsberg area (Hessen – dynamic under unplanted conditions. The pots Germany). The middle compartment was filled were watered daily and water was added to Slits for controlling nutrient solution level Soil Quartz sand Soil Soil Quartz sand Soil Drainage Windows Drainage Figure 27. Schematic representation of plant growing technique with split root system. 77 maintain the soil with 60% of its water holding mM KH2PO4 adjusted to pH 2.2 with H3PO4 was capacity by weight. The plants were grown in a used for isocratic elution, with a flow rate of 0.25 growth chamber, maintained at 25°C, with a photon mL min-1 at 24°C and UV detection at 210 nm. flux density of 300 µmol m-2 s-1 and 80% relative Identification of organic acids was performed by humidity during 16 h day and at 20 °C and 70% comparing retention times and absorption spectra relative humidity during 8 h night. with those of known standards. Determination of root surface APase activity There were four replicates for each treatment. - Excised segments (2-3 root tips at 1.5-2 cm) of The pots were completely randomized and re- root tips were harvested in the soil compartments positioned weekly to minimize any effect of and transferred to 1.5 mL Eppendorf reaction uneven environmental factors. vials. The root segments were washed 2 times with distilled water for 5 min to remove contents Collection of root exudates using a of wounded cells, followed by adding 0.5 mL percolation system - Prior to collection of root water, 0.4-mL Na-Ac buffer and 0.1 mL pNPP exudates, the sand compartment was thoroughly substrate. After a reaction time of 10 min at 25- washed with distilled water until removing 30°C, an aliquot of 0.8 mL of the reaction nutrient ions, specially NO3, which affects medium was taken out and mixed with 0.4 mL of accurate determination of organic acids by 0.5 M NaOH to terminate the reaction. The HPLC. The sand compartment drainage’s was absorption of reaction solution was measured at closed and then filled with bi-distilled water for 1 405 nm. The fresh weight of excised segments of hour. After that the leached root exudates were root tips was recorded after determination of collected to avoid O2 stress (deficiency) in the APase activity. roots and immediately the sand compartment was again filled with the collected root exudates for Quantitative determination of the pH values one hour. Finally, leaving the drainage open, the at the root surface with the antimony collected root exudates were once more added to electrode In the soil compartments, the pH the sand by a pressure bottle in order to leach all values at the root surface of young and old root as possible root exudates from the sand. The were measured potentiometrically with an collected root exudates were lyophilized to antimony electrode. concentrate them and after that the dry root exudates were diluted in 1 mL bi-distilled water Plant and soil harvests - The plants were and put into 1.5 mL Eppendorf reaction vials. harvested as separate roots and shoots at 30, 60 Thereafter, they were subjected to shaking for and 100 days after transplanting (DAT) into split- three-times (30 seconds each) for extraction of root pots. The dry weight was determined by carboxylates. The aliquot was centrifuged at drying the shoots and roots in an oven at 65% for 2000g for 10 min. The supernatant was collected 1 day and then at 105°C till constant weight. by a micropipette and stored at –20ºC for After grinding, the plant material was used for analysis of organic acid anions by HPLC. determination of nutrient composition. To determine P concentration in plant tissue, shoot Analysis of carboxylates - The organic acid and root samples were wet digested in HNO anions in root exudate samples were analyzed by 3 and P was determined with Molybdate-Vanadium reversed phase HPLC in the ion suppression method. mode. Separation was conducted on a 250 × 4 mm reversed phase column (LiChrospher 100 Shoot P uptake, shoot growth rate, P acquisition RP-18, 5 µm particle size) equipped with a 4 × 4 efficiency (mg of P uptake in shoot biomass per mm LiChrospher 100 RP-18 guard column unit root length), P use efficiency (g of forage (Merck, Darmstadt, Germany). Sample solutions production per g of total P acquisition), and P- (100 µL) were injected onto the column, and 18 Influx (the amount P taken up per unit of root and 78 time) were determined. Soil solution pH value 30 days 60 days 0.8 100 days was measured directly. Soil pH was determined P-0 a a a 0.6 b b in 0.01 M CaCl2 in soil to solution ratio of 1:2.5. a Data were subjected to an analysis of variance 0.4 using the SAS computer program. Least- 0.2 significant differences were calculated by an 0.0 tuckey-test. A probability level of 0.05 was 0.3 P-1000 a a considered statistically significant. 0.2 a a a a Results and Discussion 0.1 0.0 At high P supply (P-1000) the shoot yield (Figure 434 744 4 4 4 4 17 18 743 874 743 4 1 1 1 187 28) and the shoot P concentration of both Genotype genotypes was similar. i.e. the growth potential of Figure 29. Influence of phosphorus supply (P-0 and P- both genotypes is the same. But under low P 1000) on genotypic differences in P use efficiency of 2 availability, the shoot dry matter yield and shoot P accessions of Arachis pintoi grown for 30, 60 and 100 days concentration were different between the in a clay loam oxisol in a growth chamber. Means are different (P<0.05) if letters above bars are different within genotypes. Since under sufficient P condition the P supply level at a given age. genotypes show the same growth, the difference at P-0 is because they have different P efficiency (Figure 29). At 30 DAT, the two genotypes had 30 days 60 days 100 days 7 P-0 b different shoot P contents (Figure 30) under low 6 P availability and at 100 DAT the P content in the 5 4 shoot of CIAT 18744 is 5 times higher than that 3 2 of CIAT 17434. This shows, that the high shoot b a 1 a b a dry matter yield by CIAT 18744 was due to its 0 a high P uptake. 50 P-1000 a 40 a 30 b 30 days 20 60 days 100 days a a 5 P-0 b 10 4 0 434 7 8744 7434 44 34 44 3 1 1 1 187 174 187 Genotype 2 b a 1 b a Figure 30. Influence of phosphorus supply (P-0, P-50 and a 0 P-1000) on genotypic differences in shoot P content of 2 20 accessions of Arachis pintoi grown for 30, 60 and 100 days P-1000 a a in a clay loam oxisol in a growth chamber. Means are 15 different (P<0.05) if letters above bars are different within 10 P supply level at a given age. a b 5 a a 0 Differences in P uptake by a plant could be due 4 4 743 874 7434 744 8 7434 44 1 1 1 1 1 187 to differences in the size of the root system and/ Genotype or differences in P influx i.e. the amount taken up per unit of root and time. As a measure of the Figure 28. Influence of phosphorus supply (P-0 and P- 1000) on genotypic differences in shoot biomass (forage) size of the root system, we used the root length- production of 2 accessions of Arachis pintoi grown for 30, 60 shoot biomass ratio (Figure 31). CIAT 18744 and 100 days in a clay loam oxisol in a growth chamber. showed the lower ratio at each harvest. Thus the Means are different (P<0.05) if letters above bars are different high shoot P content, of the efficient genotype, within P supply level at a given age. DAP = days after was not associated with more roots but to a planting. greater P influx. The P influx (ability of plant P 79 Dry weight shoot, g plant-1 -1 P content, mg plant-1 P use efficiency, g DM mg Ptotal uptake per unit length of the root per unit time) 30 to 60 days 60 to 100 days 14 was determined for the period between 30 and 60 12 P-0 b days and 60 and 100 days. The P influx by the 10 CIAT 18744 was markedly greater than that of 8 CIAT 17434 (ca. 2 times and 3 times for the first 6 and second harvest, respectively). 4 a b 2 a 30 days 60 days 60 100 days 0 P-0 a a 50 P-1000 40 b 50 b a 30 a 40 20 b 10 30 a a b 0 20 25 P-1000 a 20 10 a 15 0 a 10 a 17434 744 434 744 18 17 18 a a 5 Genotype 0 4 4 4 4 4 Figure 32. Influence of phosphorus supply (P-0 and P- 743 874 743 874 743 8744 1 1 1 1 1 1 1000) on genotypic differences in P influx of 2 accessions Genotype of Arachis pintoi grown for 30, 60 and 100 days in a clay Figure 31. Influence of phosphorus supply (P-0 and P- loam oxisol in a growth chamber. Means are different 1000) on genotypic differences in root length/shoot (P<0.05) if letters above bars are different within P supply biomass ratio of 2 accessions of Arachis pintoi grown for level at a given age. 30, 60 and 100 days in a clay loam oxisol in a growth chamber. Means are different (P<0.05) if letters above bars Table 29. Influence of P supply on are different within P supply level at a given age. colonization of roots by mycorrhizae. P-level Genotype Colonization % With age, the P influx increases 4 times higher for 30 d 60 d 100 d CIAT 17434 and 6 times higher for CIAT 18744 P-0 17434 < 1 < 5 25 a (Figure 32). One possible reason for high P influx 18744 < 1 < 5 56 b could be due to increase in P absorbing surface area per cm root that was infected by hyphae of P-1000 17434 < 1 < 1 5 a 18744 < 1 < 1 12 a native arbuscular-mycorrhizae (AM). Only at 100 DAT and -under low P availability, CIAT 18744 had a higher colonization of native AM than CIAT Thus the APase activity may not have contrib- 17434 (Table 29), which may be the reason for the uted to either the high P influx of the efficient higher P influx of CIAT 18744, during the second genotype or the increase of P influx with age. growth period. However, the infection found after The P absorbing surface area and the hydrolysis 30 and 60 DAT cannot explain the P influx of organic P cannot explain the higher P influx by differences between the genotypes, at 30 DAT. CIAT 18744. To explore further possible reasons for differences Another possibility could be that the plants in P influx between the two accessions, we investi- mobilise P, i.e. increase P concentration in soil gated the acid phosphatase (APase) acitivity on the solution. P concentration in soil solution of root surface and exudation of organic acids from unplanted and planted soil was also determined roots. At each growth stage, CIAT 18744 had a (Figure 34). lower APase activity at the root surface than the CIAT 17434 (Figure 33). After 30 days of planting, the P concentrations for all were approximately 0.1 µM, which is 80 Root length / shoot ratio, m g-1 P influx, 10-15 mol cm-1 s-1 30 days 60 days 60 100 days 6.0 60 days 60 days 100 days P-0 0.8 50 a 5.6 6.0 40 a 5.2 30 b 0.6 b 4.8 20 a 4.4 b 5.6 10 4.0 0.4 0 3.6 25 6.0 P-1000 a 5.2 a 5.6 0.2 20 b b 5.2 15 a 4.8 0.0 4.8 10 b lant 434 4 t t 4.4 p 17 1874 plan 17434 744 lan 434 4 o o 18 o p 17 1874 n n n 5 4.0 Genotype 0 3.6 434 744 434 744 434 4 17 18 17 18 17 1874 Genotype Figure 33. Influence of phosphorus supply (P-0 and P- Figure 34. Influence of 3 accessions of Arachis pintoi and 1000) on genotypic differences on APase activity at the 1 accession of Arachis hypogea on P concentration in soil root surface and pH on the young root of 2 accessions of solution and pH in soil solution when grown with no Arachis pintoi grown for 30, 60 and 100 days in a clay loam phosphorus supply (P-0) for 30, 60 and 100 days in a clay oxisol in a growth chamber. Means are different (P<0.05) if loam oxisol in a growth chamber. Means are different letters above bars are different within P supply level at a (P<0.05) if letters above bars are different within P supply given age. level at a given age. below CLmin found for Arachis hypogea. At 60 Further research work is necessary to identify DAT, especially at 100 DAT, the P concentration the physiological mechanisms responsible for the of the planted soil increased markedly and the high P uptake observed with the Arachis pintoi efficient genotype showed the highest P CIAT 18744. concentration. It was 6 times higher than that of the unplanted soil Lactic acid Acetic acid P mobilization has very often been related to 0.14 P soil solution 18744 organic acids exuded by roots. Twelve different P soil solution 17434 organic acids were analyzed from the collected 0.8 0.12 root exudates in which only lactic acid and acetic 0.10 acid were detected in significant amounts with 0.6 the HPLC. Organic acids, like citric acid or malic 0.08 acid, that are often shown to be associated with P mobilization, were only found in trace amounts 0.4 0.06 (Figure 35). 0.04 0.2 The observed organic acids were related to the P 0.02 soil solution concentration. The highest exudation rate was associated with the lowest P 0.0 0.00 concentration in soil solution with CIAT 17434, 30 60 100 30 60 100 Days after transplanting and the increase of the P concentration with age was associated with a decrease of organic acid exudation. Thus the organic acid exudation by Figure 35. Influence of 3 accessions of Arachis pintoi and 1 the genotypes could not explain the increase of P accession of Arachis hypogea on P concentration in soil solution and organic acid exudation from roots when grown concentration in soil solution. Moreover, with no phosphorus supply (P-0) for 30, 60 and 100 days monocarboxylic acids such as lactic or acetic in a clay loam oxisol in a growth chamber. Means are acid have not been associated with P different (P<0.05) if letters above bars are different within mobilization. P supply level at a given age. 81 APase, µmol PNP h-1 g RFW pH on the young root P soil solution, µM P in soil solution, µM pH in soil solution Org. acid, nmol cm-1 h-1 3.2 Genotypes of Brachiaria with dry season tolerance Highlights • Found that total nonstructural carbohydrate (TNC) content in stems of Brachiaria genotypes increased with increasing drought stress in large plastic soil cylinders under greenhouse conditions. • Found that Brachiaria hybrid Mulato 2 is superior to cv. Mulato in terms of dry season tolerance under low fertility acid soil conditions in the Llanos of Colombia. This was associated with a greater proportion of fine roots in the top 5 cm of the soil profile. 3.2.1 Genotypic differences in root distribution and drought tolerance of 2 hybrids and 3 parents Contributors: J. Rincón, J. Polania, F. Feijoo, R. García, J. Ricaurte, J. W. Miles and I. M. Rao (CIAT) Rationale The trial was planted as a randomized block in split-plot arrangement with three levels of water Identification of shoot and root attributes that are supply: 100% field capacity (well-watered), 60% associated with superior drought adaptation will field capacity (moderate drought stress) and 30% help to develop rapid and reliable screening field capacity (severe drought stress) as main methods. These methods are needed to develop plots and genotypes as sub-plots. Soil was Brachiaria hybrids that combine drought fertilized with adequate level of nutrients (kg/ha adaptation with other desirable attributes. Field of 80 N, 50 P, 100 K, 66 Ca, 28 Mg, 20 S and studies conducted for the past few years in the micronutrients). Llanos of Colombia indicated that one of the Brachiaria hybrids, CIAT 36087 (FM9503-S046- Treatments of water stress were imposed after 024 or Mulato 2) is superior to its parents and three weeks of initial growth of plants established other hybrids. This hybrid maintained greater with seed. Water stress was maintained by proportion of green leaves during dry season weighing each cylinder every week and applying under field conditions. A greenhouse study was water to the soil at the top of the cylider. After 2 conducted to characterize shoot and root months of stress treatment (at 85 days after responses of this hybrid in comparison to its germination), shoot biomass distribution, root parents and another hybrid, cv. Mulato when subjected to moderate and severe drought stress conditions. Materials and Methods A greenhouse study was conducted using a sandy loam oxisol from Matazul farm in the Llanos of Colombia. The trial comprises 5 entries, including three parents (B. decumbens CIAT 606, B. brizantha CIAT 6780, B. ruziziensis 44-02) and two hybrids (cv. Mulato and CIAT 36087). Plants were grown in large plastic cylinders (100 cm long and 15 cm diameter) covered with PVC tubes (Photo 15). Photo 15. A large plastic cylinder showing root distribution across soil depth. 82 biomass and root length distribution in different bottom of the cylinder. We also noted problems of soil depths, and leaf and stem nutrient composi- compaction in some cylinders. Therefore we tion, ash content and TNC (total nonstructural conducted some additional studies to overcome carbohydrates) contents were determined. these problems and found that use of 2:1 of soil and sand in smaller plastic tubes (50 cm long and Results and Discussion 5 cm diameter) could overcome some of the problems encountered with large cylinders. Use Reducing the water supply to 30% of field of small cylinders will also facilitate evaluation of capacity markedly decreased the leaf, stem and a larger number of genotypes. root biomass of the three parents and the two hybrids (Figure 36). Leaf biomass of CIAT 606 Results on the determination of N, P, ash (min- was lower than the other genotypes tested at all eral) and TNC contents in leaves and stem tissue three levels of water supply. Differences be- indicated that water stress could markedly tween the two hybrids (CIAT 36061 and CIAT increase stem TNC content (Figure 38). Severe 36087) in leaf biomass were not significant within water stress also increased stem N, P and ash each level of water supply. contents. But leaf N, P, ash and TNC contents were not markedy influenced by water stress Results on root length distribution showed that the conditions. hybrid Mulato 2 had a greater proportion of fine roots in the top 5 cm of soil profile at all 3 levels Work is in progress to evaluate the usefulness of of water supply (Figure 37). Higher values of fine root production in top soil, root penetration in root length observed for 50 to 100 cm soil depth subsoil and % increase of stem TNC as in some genotypes at 60% and 100% of field indicators of drought tolerance and green leaf capacity indicates the growth of roots on the production in Brachiaria. surface of the plastic tube and reaching to the 30 30% Field Capacity 60% Field Capacity 100% Field Capacity Leaves; LSD0.05 : NS Leaves; LSD0.05 : 1.76 Leaves; LSD0.05 : NS 25 Stem; LSD : 1.48 Root 0.05 Stem Stem; LSD0.05 : NS Stem; LSD0.05 : NS Root; LSD0.05 : NS Leaves Root; LSD : NS Root; LSD0.05 : NS 0.05 20 15 10 5 0 36061 36087 AT 606 T T I T 6780 44-02 6061 3 36087 T 606 6780 -02 1 7 6 0 z T T IA T z 44 T 3606 08 60 78 T 36 IAT T 6 z 44-02 CIA CIA C CIA B. ru CIA CIA C CIA B. ru CIA CIA C CIA B. ru Genotypes Figure 36. Influence of three levels of water supply (100%, 60% and 30% of field capacity) on dry matter distribution among leaves, stem and roots of five Brachiaria genotypes. 83 Total biomass (g/plant) 30% FC 60% FC 100% FC 0 - 5 5 . 10 10. 20 20. 30 Total: 313 Total: 379 Total: 582 30. 40 40. 50 50. 100 CIAT 36061 CIAT 36061 CIAT 36061 0 - 5 5 . 10 10. 20 20. 30 Total: 380 Total: 463 Total: 773 30. 40 40. 50 50. 100 CIAT 36087 CIAT 36087 CIAT 36087 0 - 5 5 . 10 10. 20 20. 30 Total: 418 Total: 424 Total: 571 30. 40 40. 50 50. 100 CIAT 606 CIAT 606 CIAT 606 0 - 5 5 . 10 10. 20 20. 30 Total: 377 Total: 445 Total: 522 30. 40 40. 50 50. 100 CIAT 6780 CIAT 6780 CIAT 6780 0 - 5 5 . 10 10. 20 20. 30 Total: 308 Total: 555 Total: 354 30. 40 40. 50 B. ruz 44-02 B. ruz 44-02 B. ruz 44-02 50. 100 0 50 100 150 200 0 50 100 150 200 0 50 100 150 200 Root length (m/plant) Figure 37. Influence of three levels of water supply (100%, 60% and 30% of field capacity) on root length distribution across soil depth in five Brachiaria genotypes. 84 Soil depth (cm) Leaves Stem 3 2 1 0 0.16 0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00 30 30% FC 60% FC 100% FC 20 10 0 250 200 150 100 50 0 061 087 606 780 2 36 36 AT T 6 44-0 36061 087 606 780 4-02 T T I CIA CIA C CIA 6 6 B. ruz T T 3 T 4 CIA CIA CIA CIAT . ruz B Genotypes Figure 38. Influence of three levels of water supply (100%, 60% and 30% of field capacity) on leaf and stem N, P, ash and TNC contents of five Brachiaria genotypes. 85 TNC Content (mg/g) ASH Content (%) P Content (%) N Content (%) 3.2.2 Genotypic variation in dry season tolerance in Brachiaria accessions and hybrids in the Llanos of Colombia Contributors: I. M. Rao, J. W. Miles, C. Plazas, J. Ricaurte and R. Garcia (CIAT) Rationale germplasm accessions, CIAT 26110 was identified A major limitation to livestock productivity in from previous work in Atenas, Costa Rica as an subhumid regions of tropical America is quantity outstanding genotype for tolerance to long dry and quality of dry season feed. A field study is season (up to 6 months). The trial was planted as a completed this year at Matazul Farm in the Llanos randomized block in split-plot arrangement with of Colombia. The main objective was to evaluate two levels of initial fertilizer application (low: kg/ha genotypic differences in dry season (4 months of of 20 P, 20 K, 33 Ca, 14 Mg, 10 S; and high: 80 N, moderate drought stress) tolerance of most 50 P, 100 K, 66 Ca, 28 Mg, 20 S and promising genetic recombinants of Brachiaria. micronutrients) as main plots and genotypes as Results from this field study for the past 2 years sub-plots. Live and dead forage yield, shoot indicated that the superior performance of the nutrient composition, shoot nutrient uptake and leaf germplasm accession CIAT 26110 and the and stem TNC (total nonstructural carbohydrates) Brachiaria hybrid Mulato 2 (FM9503-S046-024), in six entries (2 hybrids, 3 parents and 1 accession) which maintained greater proportion of green were measured at the end of the dry season (56 leaves during moderate dry season in the llanos of months after establishment; March 10, 2004). Colombia, was associated with greater acquisition Maintenance fertlizer (half the levels of initial of nutrients under water deficit conditions. This application) was applied at the beginning of the wet year, we report results from the dry season seasons of 2001 and 2003. performance into fifth year after establishment. Results and Discussion Materials and Methods Because of the application of maintenance A field trial was established on a sandy loam oxisol fertilizer, forage yields with high fertilizer treatment at Matazul farm in the Llanos of Colombia in July, were greater than those with low fertilizer 1999. The trial comprises 12 entries, including six treatment (Figure 39; Table 30). At 56 months natural accessions (four parents) and six genetic after establishment (4 months after dry season – recombinants of Brachiaria. Among the March 10, 2004), live forage yield with low 3000 HIGH FERTILITY LOW FERTILITY 2500 Dead ; LSD Dead ; LSD 0.05 = 390 0.05 = 310 Stems; LSD0.05 = 163 Stems; LSD0.05 = 149 Leaves; LSD0.05 = 179 2000 Leaves; LSD0.05 = 202 1500 1000 500 0 06 73 02 24 94 10 6 3 2 4 4 0 6 18 44 46 62 61 60 7 18 44 0 46 2 62 9 11 2 26 Genotypes Genotypes Figure 39. Genotypic variation as influenced by fertilizer application in dry matter distribution among green leaves, stems and dead biomass of genetic recombinants, parents and other germplasm accessions of Brachiaria grown in a sandy loam oxisol at Matazul, Colombia. Plant attributes were measured at 56 months after establishment (at the end of the dry season – March 2004). LSD values are at the 0.05 probability level. 86 Shoot biomass (kg ha-1) fertilizer application ranged from 0 to 609 kg/ha shoot biomass, both with low and high fertilizer and the highest value of forage yield was application. As expected, the performance of observed with a germplasm accession CIAT one of the parents, BRUZ/44-02 was very poor 26110. This accession was released in Costa compared with other parents and genetic Rica as cultivar Toledo and is known for its dry recombinants as it produced almost no live forage season tolerance. Among the 3 parents, CIAT after dry season. The values of leaf to stem ratio 6294 was outstanding in green live forage and of the genetic recombinant, FM9503-S046-024, dead biomass production with low fertilizer were markedly superior to other genotypes under application. A spittlebug resistant genetic low and high levels of initial fertilizer application recombinant, Mulato 2 was superior among the (Table 31). genetic recombinants in terms of greater live Table 30. Genotypic variation as influenced by fertilizer application in live shoot biomass, leaf to stem ratio and total forage yield of genetic recombinants, parents and other germplasm accessions of Brachiaria grown in a sandy loam oxisol at Matazul, Colombia. Plant attributes were measured at 56 months after establishment (at the end of the dry season – March 10 2004). LSD values are at the 0.05 probability level. Genotype Live shoot biomass Leaf to stem ratio Total forage yield (kg ha-1) (%) (kg ha-1) Low High Low High Low High Fertilizer Fertilizer Fertilizer Fertilizer Fertilizer Fertilizer Recombinants: FM9201-1873 223 301 0.9 2.0 420 714 FM9503-5046-024 542 760 3.4 3.7 1190 1622 Parents: CIAT 606 447 665 1.3 0.9 1169 1723 CIAT 6294 582 605 2.4 2.4 1329 1442 BRUZ/44-02 0 0 . . 0 0 Accessions: CIAT 26110 609 872 2.1 1.8 1408 2019 Mean 401 534 919 1253 LSD (P>0.05) 327 318 582 667 Table 31. Genotypic variation as influenced by fertilizer application in leaf N content, stem N content and shoot N uptake of genetic recombinants, parents and other germplasm accessions of Brachiaria grown in a sandy loam oxisol at Matazul, Colombia. Plant attributes were measured at 56 months after establishment (at the end of the dry season – March 10 2004). LSD values are at the 0.05 probability level. Genotype Leaf N content Stem N content Shoot N uptake (%) (%) (kg ha-1) Low High Low High Low High Fertilizer Fertilizer Fertilizer Fertilizer Fertilizer Fertilizer Recombinants: FM9201-1873 1.05 1.11 0.72 0.89 4.04 3.12 FM9503-5046-024 0.99 0.86 0.67 0.48 4.94 5.94 Parents: CIAT 606 1.28 1.34 0.62 0.63 4.45 6.57 CIAT 6294 1.24 0.94 0.72 0.58 6.13 4.99 BRUZ/44-02 . . . . . . Accessions: CIAT 26110 1.21 0.97 0.61 0.49 6.12 6.61 Mean 1.15 1.04 0.67 0.61 5.14 5.45 LSD (P>0.05) 0.26 0.17 NS 0.35 NS 2.90 87 Results on leaf N content indicated significant greater for the hybrid Mulato 2 than the cv. differences among genetic recombinants, parents Mulato (Table 31; Figure 39). Leaf and stem N and accessions with both levels of fertilizer content and shoot N uptake values indicated that application (Table 31). Shoot N uptake with both the genetic recombinant Mulato 2 could use N low and high fertilizer application was markedly more efficiently to produce green forage in the dry season (Table 31). 3.2.3 Dry season tolerance of most promising hybrids of Brachiaria in the Llanos of Colombia Contributors: I. M. Rao, J. Miles, C. Plazas and J. Ricaurte (CIAT) Rationale Results and Discussion Evaluation of a large number of Brachiaria At 34 months after establishment, live forage yield hybrids for their resistance to spittlebug and with low fertilizer application was greater with one adaptation to infertile acid soils resulted in spittlebug resistant genetic recombinant Mulato 2 identification of a few promising Brachiaria and one parent (CIAT 6294) (Table 32). With high hybrids. We selected 4 of these hybrids for initial fertilizer application also these two genotypes further field-testing in comparison with their were outstanding in live forage yield (Figure 40). parents. The main objective was to evaluate growth and persistence in dry season with low nutrient supply in soil at Matazul farm of the Table 32. Correlation coefficients (r) between green forage yield (t/ha) and other shoot traits of Brachiaria genotypes altillanura. grown with low or high intial fertilizer application in a sandy loam oxisol in Matazul, Colombia. Materials and Methods Low High Shoot traits fertilizer fertilizer A field trial was established at Matazul farm on 31 May of 2001. The trial included 4 Brachiaria Total (live + dead) shoot biomass (t/ha) 0.81*** 0.87*** hybrids (BR98NO/1251; BR99NO/4015; Dead shoot biomass (t/ha) 0.66** 0.70*** BR99NO/4132; Mulato 2) along with 2 parents Leaf biomass (t/ha) 0.94*** 0.89*** (B. decumbens CIAT 606 and B. brizantha Stem biomass (t/ha) 0.77*** 0.83*** CIAT 6294). The trial was planted as a Leaf N content (%) -0.63** -0.71*** randomized block in split-plot arrangement with Leaf P content (%) 0.05 -0.07 two levels of initial fertilizer application (low: kg/ Leaf TNC content (mg g-1) -0.08 -0.22 ha of 20P, 20K, 33Ca, 14 Mg, 10S; and high: 80N, Leaf ash content (%) 0.15 -0.11 50P, 100K, 66Ca, 28Mg, 20S and micronutrients) Stem N content (%) -0.37 -0.58** as main plots and genotypes as sub-plots with 3 Stem P content (%) -0.61 -0.19 replications. The plot size was 5 x 2 m. A number Stem TNC content (mg g-1) 0.21 -0.2 of plant attributes including forage yield, dry Stem ash content (%) 0.40 0.53 matter distribution, nutrient (N, P, K, Ca and Mg) Shoot N uptake (kg/ha) 0.95*** 0.88*** uptake, leaf and stem total nonstructural Shoot P uptake (kg/ha) 0.93*** 0.89*** carbohydrate (TNC) content and leaf and stem Shoot K uptake (kg/ha) 0.61** 0.91*** ash (mineral) content were measured at 34 Shoot Ca uptake (kg/ha) 0.86*** 0.85*** months after establishment (March 2004). Shoot Mg uptake (kg/ha) 0.86*** 0.89*** *, **, *** Significant at the 0.05, 0.01 and 0.001 probability levels, respectively. 88 Among the 4 hybrids tested, 4624 was results reported last year from the same outstanding in its adaptation to low initial fertilizer experiment. Correlation coefficients between live application. It is important to note that both the forage yield and other plant attributes indicated hybrid 4624 and CIAT 6294 had greater amount that greater nutrient acquisition with low initial of dead biomass and stem biomass under low fertilizer application contributed to superior fertilizer application (Figure 40). performance (Table 32). No significant correlations were found between live forage yield As observed last year, results on shoot nutrient and leaf and stem TNC or ash contents. uptake, particularly Ca and Mg, indicated that the hybrid, 4624 was superior to CIAT 606 under low The performance of the 4 hybrids in comparison fertilizer application (Figure 41). Nutrient with two parents with maintenance fertilizer acquisition by the hybrid 4624 was also greater application will be monitored for next year in than the rest of the hybrids with high initial terms of green forage yield and nutrient fertilization. These results are consistent with the acquisition. 5000 Dead; LSD0.05 = 590 686 4000 Stems ; LSD0.05 = 200 NS Leaves ; LSD0.05 = 236 187 3000 HIGH FERTILIZER LOW FERTILIZER 2000 1000 0 60 6 51 15 322 24 94 06 51 5 2 4 4 1 40 41 46 62 6 12 40 1 41 3 46 2 62 9 Genotype Figure 40. Genotypic variation as influenced by fertilizer application in shoot biomass production (forage yield) of two parents (CIAT 606, 6294) and four genetic recombinants (1251, 4015, 4132, 4624) of Brachiaria grown in a sandy loam oxisol at Matazul, Colombia. Plant attributes were measured at 30 months after establishment (November 2003). LSD values are at the 0.05 probability level. NS = not significant. 60 N ; LSD0.05 = 3.43 3.54 P ; LSD0.05 = 1.05 0.53 50 K ; LSD0.05 = 6.04 4.59 Ca ; LSD0.05 = 1.51 1.22 Mg ; LSD0.05 = NS NS 40 HIGH FERTILIZER LOW FERTILIZER 30 20 10 0 06 1 6 25 01 5 32 4 4 6 1 5 2 4 4 1 4 41 46 2 29 60 25 016 1 4 41 3 62 294 6 Genotypes Figure 41. Genotypic variation as influenced by fertilizer application in nutrient uptake (N, P, K, Ca and Mg) of two parents (CIAT 606, 6294) and four genetic recombinants (1251, 4015, 4132, 4624) of Brachiaria grown in a sandy loam oxisol at Matazul, Colombia. Plant attributes were measured at 30 months after establishment (November 2003). LSD values are at the 0.05 probability level. NS = not significant. 89 Shoot nutrient uptakes (kg ha-1) Shoot biomass (Kg ha-1) 3.3 Grasses with adaptation to poorly drained soils Highlights • Plant survival and yield of Brachiaria hybrids and accessions were affected by waterlogged conditions. However, Brachiaria hybrid CIAT 36087 (Mulato 2) showed less plant mortality than other hybrids under these conditions. • Paspalum atratum cv. Pojuca showed no plant mortality and increased yield under waterlogged soil condition. 3.3.1 Field evaluation of Brachiaria and Paspalum genotypes in poorly and well drained sites in Costa Rica Contributors: Pedro J. Argel and Guillermo Pérez (CIAT) Rationale Twelve plants were established in each strata as described in 2003, and in September of this year Poorly drained sites are frequently found in many three dikes were built along the lower part of the areas of the low land tropics were cattle is an plots to create variable gradients of soil humidity: important economic activity. However, improved (a) waterlogged, (b) moderately waterlogged and, forage options are limited for permanent or (c) well drained condition. Plant mortality, vigor periodically waterlogged conditions, and as a and plant yield were measured during the wet result native or naturalized grasses of medium to period of 2004 along the three humidity strata that poor feeding qualities predominate in these areas. were created. For this reason field tests are necessary to characterize the adaptation of promising forage Results and Discussion germplasm to poorly drained soil conditions. The dikes built created the expected waterlogged Material and Methods conditions. At the lower part of the plots a permanent water table of 5 to 10 cm depth As described in the IP-5 Annual Report 2003, remained and covered partially the grass plants seedlings of the Paspalum atratum cv. Pojuca during the evaluation period. At the middle of the (CIAT 26986), B. brizantha CIAT 26124, CIAT plots, moderately waterlogged conditions were 26318, CIAT 26990, a line of this species called also created, and the water table remained Mixe, and the Brachiaria hybrids CIAT 36061 around 20 cm. The well drained conditions had a (cv. Mulato), CIAT 36087, CIAT 4015 and CIAT water table that ranged from 30 to 50 cm. 36062, were transplanted for evaluation in a site with variable slope, sufficient to create three Plant vigor and plant mortality of all Brachiaria different moisture conditions. The soil is a heavy species were affected during the evaluation clay (45-55% clay) with the following period by the soil moisture conditions created as characteristics: pH 5.6, 0.4 meq/100 ml of Al shown in Table 33. With the exception of content, high content of Ca (26.9), Mg (10.4) and P. atratum cv. Pojuca, the Brachiaria species medium content of K (0.44). Phosphorous (4 lost vigor as the soil humidity increased, indicating ppm) and Zinc (2.5 ppm) contents are low, poor adaptation to waterlogged soils. At the whereas Mn (27.5), Cu (16.3) and Fe (39.7) waterlogged site, plant mortality was relatively contents are medium. high (3 plants out of 12 plants) with Brachiaria 90 Table 33. Plant vigor and plant mortality of grass species established along a humidity gradient formed by (1) well drained conditions, (2) moderately drained, and (3) poorly drained (waterlogged) conditions in a heavy soil of Atenas, Costa Rica. Species CIAT Vigor* Plant mortality No. (No.) (1) (2) (3) (1) (2) (3) Brachiaria hybrid (cv. Mulato) 36061 3.8 4.0 2.0 0 0 3 B. hybrid (Mulato 2) 36087 4.2 4.3 3.5 0 0 1 B. hybrid 36062 4.8 2.8 2.3 0 1 2 B. hybrid 4015 3.2 3.3 1.8 0 0 3 B. brizantha 26124 4.3 2.8 1.7 0 0 3 B. brizantha 26318 2.5 2.2 1.5 0 0 2 B. brizantha 26990 2.5 2.0 1.2 1 2 2 B. brizantha (Mixe) 3.7 3.2 1.8 0 0 3 Paspalum atratum cv. Pojuca 26986 3.5 4.0 5.0 0 0 0 * Vigor rated: 1.0 = poor vigor; 5.0 = highly vigorous plant. hybrids cv. Mulato and CIAT 4015, for Brachiaria species. In contrast with P. atratum B.brizantha CIAT 26124 and Mixe, and cv. Pojuca there was higher yields under water- moderately (2 plants dead out of 12 plants) for logged conditions (Table 34). Brachiaria hybrid CIAT 36062, B. brizantha CIAT 26318 and CIAT 26990. The Brachiaria It was interesting to observe that the yields hybrid CIAT 36087 (Mulato 2) had only 1 dead recorded for the Brachiaria hybrid cv. Mulato plant at the flooded site, while P. atratum cv. and other Brachiaria species, were similar at the Pojuca did not show any sign of plant mortality three humidity treatments, indicating that for the under these conditions, indicating the good degree of plant mortality recorded, the survival adaptation of this species to waterlogged soils. plants compensated in growth for the lost ones, with the exception of B. hybrid CIAT 36087 As the soil moisture increased, there was a (Mulato 2) and Mixe that produced significant tendency for reduced plant yields in all more plant yield at well drained sites. Table 34. Dry matter yields (DMY) of grass species established along a humidity gradient formed by (1) well drained conditions, (2) moderately drained, and (3) poorly drained (waterlogged) conditions in a heavy soil of Atenas, Costa Rica (means of 4 cuts of 5 weeks re-growths during the wet period 2004). DMY (g/plant) Species CIAT Well Moderately Poorly Probability No. drained drained drained Brachiaria hybrid (cv. Mulato) 36061 86.7 a* 76.1 a 72.5 a ns Brachiaria hybrid (Mulato 2) 36087 99.4 a 75.0 b 75.2 b p=0.02 Brachiaria hybrid 36062 71.3 a 59.3 a 56.7 a ns Brachiaria hybrid 4015 55.4 a 51.7 a 43.8 a ns B. brizantha 26124 67.7 a 62.3 a 61.2 a ns B. brizantha 26318 56.3 a 48.8 a 44.6 a ns B. brizantha 26990 37.0 a 30.6 a 30.0 a ns B. brizantha (Mixe) 64.8 a 57.3 ab 47.2 b p=0.10 Paspalum atratum cv. Pojuca 26986 101.5 a 104.3 ab 237.4 c p=0.0001 *Within the same line means followed by the same letter are not statistically significant 91 3.4 Nitrification inhibition in tropical grasses Highlights • Root exudates from B. humidicola are effective, persistent and stable at inhibiting nitrification up to at least 75 days. • Presence of NH4-N stimulates the synthesis and release of NI activity in exudates produced by the roots of B. humidicola. • Genetic variability in capacity to inhibit nitrification was found among accessions of B. humidicola held by CIAT, which opens up the possibility for breeding for this trait 3.4.1 Bioassay – Further improvements and refinements to the methodology: Expression of NI activity in AT (equivalent to allylthiourea inhibition) units Contributors: G.V. Subbarao, O. Ito, T. Ishikawa, and K. Nakahara (JIRCAS, Japan) Rationale it possible for the evaluation and comparative analysis of crop and forage germplasm We have further improved the bioassay accessions and breeding lines for the NI methodology and developed ways to express (nitrification inhibitory) activity in root exudates. inhibitory effect (on nitrification from root 100 exudates) in equivalent standard inhibitor, allylthiourea (AT) units. The transgenic 80 Nitrosomonas responds linearly to the AT concentration in the bioassay medium (Figure 42). 60 Using this relationship, the inhibitory effect from 40 root exudates (that is determined using bioassay) is expressed in AT units, which can be subjected to 20 statistical analysis. One AT unit of NI is defined as 0 the inhibitory activity caused by the presence of 0.22 µM of AT in the bioassay medium. These 0.00 0.05 0.10 0.15 0.20 0.25 improvements in the bioassay methodology will AT concentration (micro M) in bioassay medium make it now possible to characterize the nitrification inhibition phenomenon in root exudates Figure 42. Transgenic Nitrosomonas europaea response of plants. Also, the bioassay methodology will make to synthetic nitrification inhibitor, allylthiourea in the bioassay medium. 3.4.2 Stability, persistence and effectiveness of Brachiaria humidicola root exudates in inhibiting nitrification in soil Contributors: G.V. Subbarao, H. Wong, T. Ishikawa and O. Ito (JIRCAS, Japan); M. Rondon and I.M. Rao (CIAT) Rationale exudates to determine the inhibitory effect on nitrification in soil (IP-5 Annual Report, 2003). This year, we have improved further the We have tested the stability, persistence and protocols in processing and testing of root effectiveness of the inhibitory effect from root 92 Nitrification inhibition (%) exudates of B. humidicola on nitrification in soil. NI activity of 10 AT units g-1 soil (Soil from 80 Tsukuba, Japan) was added to the soil with 182 ppm of N as (NH ) SO and incubated at 20 οC 60 4 2 4 and 95% RH. The experiment was replicated Root exudates 40 three times. Sequential sampling was done at 25 Nitrapyrin d intervals and the incubation was continued for 20 100 days. NI activity of 10 AT units g-1 soil was very effective in inhibiting nitrate formation in soil 0 (about 70% inhibition) and remained effective in 0 25 50 75 100 inhibiting nitrification (about 50%) until 75 days. A Incubation tim e (d) substantial portion of the inhibitory effect from NI activity was lost between 75 and 100 period of Figure 43. Inhibitory effect from root exudates (10 AT incubation in soil. units NI activity g-1 soil) and nitrapyrin (4.5 ppm) on nitrate formation in soil during 100 d incubation period  (Note: In control, nearly 90% of the added NH -N was The synthetic nitrification inhibitor, Nitrapyrin 4 nitrified by 75 days). did not inhibit nitrification effectively (only about 20% inhibition on nitrate formation) at 4.5 ppm Our results also indicate that the NI activity under these conditions and lost its effectiveness release rates mentioned above can be maintained after 30 days of incubation (Figure 43). for long periods of time (we have tested up to 15 days and that the release rates were maintained). Our results demonstrate that root exudates from B.humidicola are effective, persistent and stable This is the first time that we have demonstrated in inhibiting nitrification in soil (up to 75 days at the effectiveness, stability and persistence of root least). Our results indicate that two exudates (from B. humidicola) inhibitory effect B. humidicola plants of 60 to 70 d old can on nitrification in soil. release up to 100 AT units of NI activity (in 24 h period) under optimum conditions. 3.4.3 Influence of NH -N on expression/regulation and release of NI activity in root exudates 4 of B. humidicola Contributors: G.V. Subbarao, H. Wong, T. Ishikawa, O. Ito and K. Nakahara (JIRCAS, Japan); M Rondon and I.M. Rao (CIAT) Rationale or 1 mM KNO for 24 h. NI activity of root 3 exudates was determined with the NI bioassay. We have tested the hypothesis that nitrogen Root exudates of NH -N grown plants showed 4 forms (NH -N vs NO -N) can influence the NI activity, whereas NI activity was completely 4 3 release of NI activity from roots in absent in the root exudates of NO -N grown 3 B.humidicola. Plants of B. humidicola were plants (data not shown for NO -N grown plants 3 grown hydroponically with two sources of as there was no NI activity detected in root nitrogen – 1 mM N as (NH ) SO or KNO for exudates). 4 2 4 3 70 days. The experiment was replicated three times. Root exudates were collected by keeping Presence of NH -N in the root exudates 4 intact plant roots in distilled water, 1 mM NH Cl collection solutions further stimulated the release 4 of NI activity in NH -N grown plants (Figure 44). 4 93 Inhibition on nitrate formation in soil (%) The NI activity released in the presence of NH 18 4-N was several-fold higher than in the absence of NH4-N (i.e. when root exudates are 16 collected using distilled water). 14 12 Our results support the hypothesis that presence 10 of NH4-N stimulates the synthesis and release of 8 NI activity from root exudates (data not 6 presented on the root tissue NI levels). The 4 release of NI activity from roots appears to be a 2 highly regulated phenomenon and NH4-N in the 0 rhizosphere is certainly one of the important RE-water RE-NH4-N regulating factors for the release of NI activity. Root exudates collected using distilled water or 1 mM NH4Cl Also, regulatory role of NH4-N in the rhizosphere Figure 44. Influence of NH4-N in the root exudates for the release of NI activity from roots further collection medium on the release of NI activity into root indicates the functional significance of NI activity exudates from B. humidicola roots (Specific NI activity = -1 in protecting NH4-N in soil from nitrification. NI activity g root dry weight). 3.4.4 Screening for genetic variability in the ability to inhibit nitrification in accessions of B. humidicola Contributors: M. Rondón, I.M. Rao, C.E. Lascano, J.A. Ramírez, M.P. Hurtado, J. Ricaurte (CIAT); G.V. Subbarao, T. Ishikawa, K. Nakahara, and O. Ito (JIRCAS, Japan) Rationale collected from plants grown under greenhouse conditions using infertile acid soil. Also we intend Ongoing collaborative research with JIRCAS, to test the relationship between nitrification Japan, has shown that B. humidicola CIAT 679 inhibition and root production in terms of biomass inhibits nitrification of ammonium and reduces the and length. emission of nitrous oxide into the atmosphere. Given these findings with the commercial cultivar Materials and Methods of B. humidicola CIAT 679, and the fact that a range of inhibition of nitrification was observed A sandy loam Oxisol from the Llanos (Matazul) among different tropical grasses, there is a need of Colombia was used to grow the plants (4 kg of to determine the extent of genetic variation soil/pot) under greenhouse conditions. A basal among the 69 accessions of B. humidicola that level of nutrients were applied before planting are part of CIAT germplasm bank. This (kg/ha): 40 N, 50 P, 100 K, 66 Ca, 28.5 Mg, 20 S information will be extremely useful to develop and micronutrients at 2 Zn, 2 Cu, 0.1 B and 0.1 screening methods to select genetic recombinants Mo. A total of ten accessions were used of Brachiaria grasses that not only are resistant (accessions CIAT 679, 6133, 6369, 6707, 16866, to major biotic and abiotic stress factors but also 16867, 16886, 16888, 26149, 26159). A control can protect the environment. Given the vast without plants was also included. The experiment areas under B. humidicola in the tropics, was arranged as a completely randomized block reductions in net emissions of N2O could also design with four replications. Each pot contained have important environmental implications. four plants. After sowing, plants were allowed to grow for 15 weeks and were cut to 10 cm height The main objective was to quantify differences to simulate grazing effects under field conditions. among 10 accessions of B. humidicola regarding Plant tissue was dried and saved. the nitrification inhibition activity of root exudates 94 Specific NI activity (AT units) of root exudates Plants were allowed to re-grow during 5 weeks Table 35. Dry matter partitioning differences among ten more to promote a well developed root system accessions of B. humidicola grown in pots under greenhouse conditions. Plants were harvested at six months after planting. and then ammonium sulfate was applied in solution at a rate of 38.5 mg N-NH /kg soil 4 CIAT Dry matter (g/pot) (equivalent to 100 kg N-NH per hectare). Five Accession Root biomass Shoot biomass Total biomass 4 Number weeks later plants were harvested (at 25 weeks CIAT 679 4.29 (1.19) a 14.76 (3.76) d 19.05 (3.68) f after sowing). At the end of the experiment, CIAT 6133 4.14 (1.65) a 15.06 (1.90) d 19.20 (3.49) f plants were carefully removed from soil CIAT 6369 4.77 (1.58) b 14.35 (1.59) d 19.12 (2.52) f CIAT 6707 4.92 (0.72) a 17.84 (2.75) d 22.75 (2.61) f minimizing mechanical damage to the roots. Soil CIAT 16866 3.52 (0.89) a 13.45 (0.96) e 16.97 (0.95) g adhered to the fine toots was removed and the CIAT 16867 3.50 (0.38) a 14.70 (1.65) e 18.20 (1.56) g CIAT 16886 4.48 (1.09) b 15.53 (4.56) d 20.01 (5.12) f roots were rinsed with deionized water. Once CIAT 16888 3.26 (0.72) a 16.97 (1.40) d 20.22 (1.17) f clean, the roots were fully immersed in 1 liter of CIAT 26149 2.39 (0.30) c 17.31 (3.20) d 19.70 (3.09) f CIAT 26159 2.96 (1.43) c 16.15 (2.09) d 19.10 (2.20) f deionized water and were allowed to produce Numbers in parenthesis indicate standard deviation. In a given column, root exudates during 24 hours. Collected root data followed by the same letter indicate non- significant differences exudates were kept in the refrigerator and were (LSD, p<0.05). reduced in volume to approximately 100 ml using a freeze drier. two fold greater than the value for the lowest in the group, the accession 26149. Harvested plants were separated into shoot and roots. Root length was measured using a root Results from the bioassay indicated substantial length scanner. Dry matter content and N status level of NI (nitrification inhibitory) activity in the of both shoot and root biomass was determined. root exudates of most of the accessions tested At harvest time, soil samples were extracted with (Table 36). However a range in NI activity was KCL and analyzed for nitrate and ammonium found among the tested accessions. Accessions levels. The concentrated root exudates were could be grouped in 3 classes in relation to their further concentrated using a rotovapor using specific NI activity. Group 1 with the accession protocols that were developed for this purpose. CIAT 16867 showed no NI effects, behaving The final concentrate was tested for its nitrification inhibitory activity using a specific Table 36. Nitrification inhibitory activity (total NI activity pot-1 and specific activity .g root dry weight-1) of the root bioassay developed at JIRCAS. exudates from ten accessions of B. humidicola grown under glasshouse conditions. Plants were grown for six months before the collection of root exudates. Results and Discussion CIAT NI activity Specific NI activity Results on dry matter partitioning among shoot Accession (in AT units pot-1) (in AT units g root dwt-1) Number and root biomass from the comparative evaluation of the ten accessions are presented in CIAT 679 68.84 (24.1) cd 17.48 (8.4) c CIAT 6133 51.58 (16.9) cd 12.24 (2.83) c Table 35. No significant differences were found CIAT 6369 86.94 (14.3) c 20.72 (4.2) c in total biomass production among most of the CIAT 6707 69.68 (5.5) cd 14.86 (1.2) c CIAT accessions except for the accessions of CIAT 16866 41.48 (6.9) d 11.26 (2.9) c CIAT 16867 -48.55 (18.1) e* -13.42 (3.35) d 16866 and 16867, which were lower than the rest CIAT 16886 128.05 (15.3) ab 27.95 (5.8) bc of the accessions. However, significant CIAT 16888 160.95 (6.08) a 53.76 (17.45) a CIAT 26149 33.5 (39.8) d 15.22 (18.15) c differences among accessions were found in root CIAT 26159 126.17 (19.9) b 46.33 (19.0) ab biomass production. The commercial cultivar, Note: Numbers in parenthesis indicate standard deviation. In a given column, data followed by the same letter indicate non- CIAT 679, which has been used in most of the significant differences (LSD, p<0.05). NI activity is expressed previous work, seems to have root biomass as AT units; One AT unit is defined as the inhibitory activity around the average value for the group tested. caused by the addition of 0.22 µM of allylthiourea (AT) in the bioassay medium. Thus, the inhibitory activity of the test The accession 6707 produced the highest root samples of root exudates is converted into AT units for the ease biomass among the tested accessions. Values of of expression in numerical form. *Negative activity indicates that nitrification was stimulated by root biomass of this accession were more than the root exudates. 95 similarly to other grasses such as Panicum genetic recombinants together with other maximum, which also lack the NI activity. Group desirable agronomic traits. 2 that included accessions CIAT 6133, 6707, 16866, 26149, 6369, and 6707 showed similar The presence of substantially higher levels of NI levels of NI that was observed with the activity in the root exudates of the two CIAT commercial cultivar CIAT 679. Group 3 that accessions (16888 and 26159) draws attention to included the accessions 16886, 16888, and 26159 the need to study these accessions in more detail. showed significantly higher levels of NI than the The immediate task is to continue the screening accession 679. The accession 16888 was of other accessions of B. humidicola from the outstanding in its NI activity with a value of more gene bank and to initiate screening of other than three times to that of the value of CIAT 679. commercially important grasses and crops for their ability to inhibit nitrification. As a Results on NI activity indicate that wide genetic continuation of this work, this year we have variability exists among accessions of initiated the screening of another 11 accessions of B. humidicola in relation to the effectiveness of B. humidicola including all materials that are root exudates to inhibit nitrification in soils. This classified as putatively sexual. An additional genetic variability for NI activity could be experiment will be conducted to obtain and test exploited in a breeding program to select for NI activity of root exudates from maize, rice, genotypes with different levels of NI activity. sorghum, soybean, cowpea and common bean. Once all the accessions in the gene bank are Results from this study will be reported next year. tested, accessions with superior NI activity could Further research work is needed to determine the be used as parents to regulate NI activity in the relative importance of total NI activity vs. specific NI activity in influencing the nitrification process (i.e. inhibition) in a soil environment. 3.4.5 Field validation of the phenomenon of nitrification inhibition from Brachiaria humidicola Contributors: M. Rondón, I.M. Rao, C.E. Lascano, M. P. Hurtado, J. Ricaurte (CIAT); G.V. Subbarao, T. Ishikawa and O. Ito (JIRCAS, Japan) Rationale the nitrification process, while those of soybean seem to stimulate the nitrification process in soils. Research conducted at JIRCAS and CIAT for Soybean (usually in rotation with maize) is the past three years using B. humidicola has becoming increasingly important as a crop not shown that root exudates from this tropical grass only in Latin America but also in many tropical have the capability to inhibit/suppress the and temperate regions. Other grasses such as nitrifying populations in the soil. Factors such as Panicum maximum lack the NI activity, while the presence of NH4-N in the soil seem to have a Brachiaria hybrid cv. Mulato was found to have stimulating effect on the expression of a moderate level of NI activity. The use of this nitrification inhibition (NI) activity in the root hybrid is expanding rapidly in Latin America due exudates of B. humidicola. Differences have to its high productivity and forage quality. been found among accessions of B. humidicola with regard to their NI activity. Also, our recent All these above studies were conducted either studies involving soils incubated with root using hydroponic systems or soil in pots under exudates of B. humidicola and soybean have greenhouse conditions to test and verify the shown that root exudates from B. humidicola concept of the biological phenomenon of have suppressed the N O emissions and inhibited nitrification inhibition. There is a clear need to 2 96 validate some of these findings under field Soybean lacks NI ability (indeed accelerate conditions. This year a collaborative (CIAT- nitrification), while maize shows some degree of JIRCAS) long-term experiment was initiated to inhibitory capability. As first crop of the rotation validate the phenomenon of NI under field we used maize variety (ICA V109). A plot conditions and to monitor whether the NI activity without plants where emerging weeds are is a cumulative process in the soil removed manually is used as an absolute control. Given the vast areas currently grown in the Plot size for each treatment was 10m x 10m. tropics on tropical grasses, an understanding of Irrigation will be provided if necessary. Maize the NI process and the possibility of managing it was planted from seeds and the tropical forage to improve fertilizer N use efficiency, reduce grasses were propagated from vegetative nitrate pollution of surface and ground waters as cuttings. Fertilizer will be applied (broadcast) for well as reduce net impact on the atmosphere every crop cycle, consisting of (kg/ha) 96 N (as through reduced emissions of nitrous oxide, could urea), 48 K, 16 P, 0.4 Zn, 0.4 B and 8 S. The have potential global implications for sustainable fertilizer is split into two equal applications: one at agricultural development and environmental 20 days after sowing of each crop (either maize protection. or soybean) and the other at flowering time at approximately 60 days after sowing. Materials and Methods A number of soil and plant parameters will be The field experiment was established on 31 measured at every four months. These include August 2004 at CIAT-Palmira on a Mollisol nitrate and ammonium availability in the soil, (Typic Pellustert) as a randomized complete dynamics of nitrifier organisms in soil, plant block (RCB) design with six treatments and 3 nitrogen uptake and nitrous oxide (N2O) replications. Annual rainfall at this site is about emissions. The NI activity of soil water extracts 1000 mm with a mean temperature of 25 οC. Soil will be measured using the bioassay. Soils is fertile with a pH of 6.9. Two accessions of samples will be periodically collected and sent to B. humidicola were included: the reference JIRCAS to assess changes in inhibitory material (CIAT 679) that has been used for most compounds in the soil. Gas samples for of our previous studies, and the high NI activity measuring N2O fluxes will be collected every germplasm accession (CIAT 16888). The Hybrid month. Once a year, soil incubation studies will Mulato was included as a moderate NI and be conducted using rhizosphere soil, to monitor Panicum maximum var. common was included nitrogen dynamics and fluxes of N2O. Currently as a negative non-inhibiting control. A crop plants are growing well and the initial sampling is rotation (maize-soybean) was included to assess expected in January 2005. Results from this field under field condition the recent finding that study will be reported next year. 3.5 Legumes (herbaceous and woody) with adaptation to acid soils and drought Highlights • Selected accessions of Desmodium velutinum with superior productivity, forage quality and that have persisted over 2 years under cutting. • Selected accessions of Canavalia brasiliensis with superior productivity and drought tolerance for seed multiplication. 97 3.5.1 Evaluation of new collections of the multipurpose shrub legumes Flemingia macrophylla, Flemingia spp. and Cratylia argentea Contributors: M. Peters, L.H. Franco (CIAT), R. Schultze-Kraft (University of Hohenheim), B. Hincapié, P. Ávila and G. Ramírez (CIAT) Rationale 2) To optimize the use and management, including conservation, of the collections. For The work of CIAT on shrub legumes emphasizes this, different approaches to identify core the development of materials to be utilized as collections for each species were tested and feed supplement during extended dry seasons. compared based on: (a) genetic diversity Tropical shrub legumes of high quality for better assessment by agronomic characterization/ soils are readily available, but germplasm with evaluation and (b) germplasm origin similar characteristics adapted to acid, infertile information. soils is scarce. Flemingia macrophylla and Cratylia argentea have shown promising results Material and Methods in such environments and hence work on these genera is part of the overall germplasm In the Annual Report of 2003 we presented development strategy of the CIAT Forages team. results on the evaluations of large collections of both Flemingia macrophylla and Cratylia C. argentea is increasingly adopted and utilized, argentea. To complement this work, we acquired particularly in the seasonally dry hillsides of additional germplasm of C. argentea and Central America, and more recently, the Llanos F. macrophylla. In view of the interest of Orientales de Colombia. However, most research Flemingia as a genus; we are also evaluating a and development is based on only few accessions new collection of different Flemingia species. and hence activities to acquire and test novel germplasm of C. argentea are of high priority. In 2002, 8 accessions of Cratylia argentea - CIAT 22377, 22395, 22388, 22389, 22395, 22401, F. macrophylla also is a highly promising shrub 22402, 22403 and Yacapani previously evaluated legume with excellent adaptation to infertile soils. in Costa Rica were sown at CIATs research In contrast to C. argentea, whose adaptation is station in Santander de Quilichao (Photo 16). In limited to an altitude below 1200 masl, contrast to other accessions of C. argentea, F. macrophylla can successfully be grown up to Yacapani is of prostrate growth habit. Accessions altitudes of 2000 masl. However, the potential CIAT 18516 and 18668 were included as utilization of F. macrophylla is so far limited by standards. the poor quality and acceptability of the few evaluated accessions. The project aims to investigate the genetic diversity within collections of F. macrophylla, Flemingia spp. and C. argentea with two main objectives: 1) To identify new, superior forage genotypes based on conventional germplasm characterization/evaluation procedures (morphological and agronomic traits, forage quality parameters, including IVDMD and tannin contents). Photo 16. Cratylia argentea at Quilichao, 2002. 98 Plants were sown in the greenhouse in jiffys and 18516, 22377, 22395, 22389, 22403 and 18668 transplanted to the field 6 weeks later. We were higher than reported for the larger employed the same evaluation methodology as collection in the Annual Report 2003, most likely for the larger collections of C. argentea and as a result of a more favorable growing F. macrophylla described in the Annual Report conditions in 2003/2004. No significant (P>0.05) 2003 (for more details on the methodology please yield differences among accessions were refer to the Annual Report 2003). recorded in either the wet or dry season. Mean dry season yields were superior to those recorded Results and Discussion in the wet season, and better regrowth was also found under dry conditions. Agronomic evaluation: Three cuts were carried out in each the dry and wet season and Highest DM yields were obtained for CIAT results on dry matter yield are presented for 18516, which had a higher yield and regrowth Cratylia argentea and Flemingia macrophylla ability than accession CIAT 18668. The mixture in Tables 37 and 39. of these accessions was released in Costa Rica as cv. Veraniega and in Colombia as cv Cratylia argentea (2002): Except for the Veranera. Lowest DM yields in both seasons distinct growth habit of Yacapani, no were recorded for accession CIAT 22401 morphological differences between accessions (Table 37). In line with results obtained for DM were observed. Though Yacapani has normally a yield, in-vitro dry matter digestibility (IVDMD) prostrate growth habit, erect plants were also was higher in the dry than in the wet period. The encountered; it is not clear if this variation is to be accessions 18516 and 18668 (mixture of these attributed to contamination of the seed, two accessions formed cv. Veranera) had lower outcrossing or diversity within the accession. IVDMD and CP as compared to accessions Additional plants of the accession were sown to CIAT 22402, 22389, 22041 and 22388. However, study the growth habit of this particular accession with the exception of CIAT 22389, accessions of in more detail. No disease and pest problems higher quality had relatively low DM yields were observed. (Table 38). In general, yields were much higher than Flemingia macrophylla (2002): In contrast to previously obtained in Costa Rica for the same the large diversity in growth types encountered in accessions. DM yields of the accessions CIAT the larger collection of F. macrophylla (see Table 37. Agronomic evaluation of a collection (2002) of Cratylia argentea in Quilichao. Data of six evaluation cuts (three in the dry season and three in the wet season). Underlaid in grey: Accessions CIAT 18516/18668. Regrowing Mean Regrowing Mean Height Diameter points DM Height Diameter points DM Accession yields yields No. CIAT Wet Dry (cm) (No.) (g/pl) (cm) (No.) (g/pl) 18516 169 149 28 580 151 126 31 683 22395 142 137 20 436 145 123 25 542 22389 152 121 17 389 156 111 22 515 22377 145 110 18 370 144 111 21 437 22403 148 124 18 351 144 113 23 437 18668 140 120 19 325 136 106 23 425 22388 146 118 14 321 150 101 18 411 22402 144 106 14 310 150 87 17 378 Yacapani 135 131 17 284 122 111 27 400 22401 121 107 17 219 129 93 17 247 Mean 144 122 18 351 143 108 22 439 LSD (P<0.05) 246.9 344.6 99 Table 38. Forage quality of accessions of Cratylia argentea evaluated in Quilichao, 2004. Grey underlaid: Accessions CIAT 18516/18668. Accession Wet Dry No. CIAT IVDMD C P ADF IVDMD C P ADF (%) 22402 60.2 20.2 24.1 70.6 22.8 22.8 22389 65.0 20.1 22.4 70.6 22.1 20.7 22401 60.1 21.7 24.9 68.4 22.4 24.9 22388 63.2 21.8 23.5 67.6 22.0 22.1 22395 60.6 21.1 25.1 65.7 23.5 25.1 22403 57.6 20.5 25.9 65.5 20.6 27.8 18668 58.6 19.3 25.2 65.1 20.8 25.4 18516 60.6 20.1 23.3 65.0 23.2 22.3 22377 55.9 21.2 27.0 64.3 22.8 26.4 Yacapani 56.8 21.5 23.5 64.1 22.0 22.7 Mean 59.8 20.8 24.5 66.7 22.2 24.1 LSD (P< 0.05) 6.9 2.8 3.7 3.7 2.1 2.9 Annual Report 2003), materials evaluated this year were more homogenous and all of erect growth habit (Photo 17). Differences in yield between the wet and dry seasons were not significant. However, significant (P<0.05) differences in DM yield were found among accessions when averaged across seasons. The highest DM yields and the best regrowth were recorded for accessions CIAT 659 and 906 (Table 39). Photo 17. Collection 2002 of Flemingia macrophylla in Quilichao. Table 39. Agronomic evaluation of a collection (2002) of Flemingia macrophylla in Quilichao. Data of six evaluation cuts (tree in the dry season and tree in the wet season). Grey underlaid: Accessions with digestibility >44% and dry matter yield >250 g/plant. Accession Height Diameter Regrowing Mean DM Regrowing Mean DM Number points yields Height Diameter points yields Wet Dry (cm) (No.) (g/pl) (cm) (No.) (g/pl) 659 127 117 37 416 119 116 46 419 906 134 108 30 409 134 102 40 409 870 126 107 37 378 124 115 45 350 591 135 114 34 374 129 102 38 315 914 134 105 31 370 137 114 40 395 816 119 108 31 347 119 109 41 304 595 133 108 19 335 128 101 26 300 615 123 107 28 328 104 97 34 280 780 128 108 30 321 130 110 35 298 857 126 99 25 314 118 97 32 324 632 124 97 29 310 120 101 39 273 753 109 93 25 297 109 94 31 246 601 126 108 26 297 134 108 34 353 629 110 94 25 291 105 101 42 274 821 110 104 33 274 112 99 31 266 804 117 103 25 274 119 101 30 313 576 130 96 20 267 129 93 25 303 542 122 96 18 206 123 80 22 186 Mean 124 104 28 323 122 102 35 312 LSD(P< 0.05) 166.62 180.62 100 In terms of quality parameters significant values were lower than for the accessions with (P<0.01) differences among accessions were highest quality selected from the larger collection measured for IVDMD, ADF and soluble tannin of Flemingia macrophylla previously evaluated content (Table 40). While some accessions had a (see Annual report 2003). higher digestibility than CIAT 17403 (control), Table 40. Forage quality of accessions of Flemingia macrophylla evaluated in Quilichao, 2004.* Accession Wet Dry Number IVDMD C P ADF Tan Sol IVDMD C P ADF Tan Sol (%) 870 45.4 21.7 23.0 4.8 44.5 20.4 21.4 1.6 816 41.3 21.1 23.4 7.4 41.1 21.1 21.6 5.5 601 41.4 20.6 22.1 3.3 40.1 20.4 21.8 5.9 780 44.6 21.5 22.2 6.0 40.1 21.4 22.2 3.9 595 41.6 21.7 21.4 2.4 39.3 20.6 22.0 6.2 857 44.8 20.0 22.3 6.5 38.6 19.8 22.7 3.4 804 37.0 22.0 22.3 6.9 38.4 21.2 23.3 6.2 542 40.4 21.3 22.8 5.3 38.2 20.6 22.7 5.3 914 38.5 19.0 23.5 5.8 37.8 19.8 24.4 1.8 576 40.1 20.9 20.5 5.9 36.9 19.9 21.3 7.4 906 42.7 20.4 22.3 5.0 36.8 20.5 23.0 2.4 591 39.6 21.7 20.9 2.1 36.7 21.1 20.9 8.0 659 38.2 20.6 20.5 7.6 35.7 20.6 21.2 4.1 753 37.8 21.9 23.8 6.8 35.4 20.4 23.6 6.5 615 39.3 21.1 24.6 3.9 35.4 20.0 24.6 8.1 821 38.0 20.1 21.7 4.7 34.8 20.3 21.1 3.7 629 32.6 20.2 23.4 3.0 32.9 19.3 20.3 8.0 632 37.1 20.1 23.2 4.7 30.9 18.7 21.4 7.3 Mean 40.1 20.9 22.3 5.1 37.4 20.3 22.2 5.3 LSD(P< 0.05) 5.8 2.8 2.3 3.9 7.0 2.5 1.93 2.7 * CIAT 17403 (IVDMD = 43.68 - 41.83; CP = 20.52 - 20.06; ADF = 20.36 - 24.09; T Sol= 4.32 - 9.62) 3.5.2 Genetic diversity in the multipurpose shrub legume Desmodium velutinum Contributors: R. Schultze-Kraft (University of Hohenheim), F. Parra (Corpoica), N.Vivas (Universidad Nacional de Colombia, Palmira), M. Peters, L.H. Franco, B. Hincapié, and G. Ramírez (CIAT) Rationale We are currently exploring the genetic diversity in a collection of D. velutinum held by CIAT in For acid, low-fertility soils in drought-prone terms of morphology, yield and quality environments there are few options in terms of parameters. From this work we expect to derive shrub legumes. Species such as Desmodium a core collection based on agronomic and velutinum may offer an option in such morphological parameters, origin information, environments (where they would complement (using GIS tools), and characterization with Cratylia argentea). There are very few studies on molecular markers, will be identified for more D. velutinum and the ones available concentrates detailed regional evaluation. on one or two accessions. However, the available information indicates that this legume produces Materials and Methods forage of high quality and has the potential to adapt to drought and (acid) low-fertility soils. A total of 137 accessions of Desmodium velutinum, mostly originating from Asia and to a 101 lesser extent from Africa, were planted at week regrowth, in both seasons (Table 41). A Quilichao (Photo 18). Plants were sown in jiffy higher number of regrowing points was recorded pots and transplanted 6 weeks later into the field in the dry season, indicating that D. velutinum in single-row plots, with 4 replications. Dry does not only survive dry periods but that it matter yield, drought tolerance and forage quality remains productive (Table 41). are the main parameters being measured. Among the semi-erect groups DM yield differences among accessions were significant (P< 0.05), with slightly higher yields in the wet than the dry season. In this group only 2 accessions, CIAT 13218 and 23983 had DM yields above 200 g DM/plant (Table 42). As observed in the erect group, there were more regrowing points in the dry than in the wet season. Quality of the semi-erect types was similar to the erect types, with a range of 59 to 76% and 17 to 25 % IVDMD and CP, respectively. A larger number of accessions in this group had digestibilities above 70%, with Photo 18. Regrowth at 8 weeks of Desmodium velutinum accessions CIAT 23992, 23923, 23922, 33387, at Quilichao 23986, 23995 and 23975 having values above 73% though at a low yield level. Results and Discussion In the prostrate group significant (P<0.01) yield Five months after transplanting plants were well differences among accessions were measured established, had a good vigor and were free of (Table 43). However yields were lower than for pests and diseases. Accessions were classified the other groups, with only CIAT 13212 having into three groups according to their growth habit: DM yields above 200g/plant (Table 43). As e = erect (54 accessions), se = semi-erect (66), digestibilities were also relatively low, this group r = prostrate (17 accessions). is probably of the least agronomic interest. DM yields, averaged over two cuts each in the wet and dry periods are presented in Tables 41, Forage quality parameters of some promising 42 and 43 for the 3 groups of D. velutinum that Desmodiun velutinun accessions in the Erect were formed based on growth habit. Each Group are presented in Table 44. The IVDMD growth type was analyzed separately as these and CP content ranged between 59% and 75%, are likely to occupy different niches. Significant and 19% and 26%, respectively. (P< 0.01) differences among accessions were recorded in each group. Accessions CIAT 23988, 23079, 23272, 33138 and 13948 had digestibilities above 70%. Highest In general, results indicate that for the group stability for digestibility and yield across seasons classified as erect, DM yields were slightly higher was measured for accessions CIAT 33443, in the wet than in the dry season, with accessions 23985, 33352, 23994 and 33138 (Table 44). CIAT 33443, 13953, 23985, 23994 and 33352 The level of CP was high in all accessions producing more than 200g DM/plant for an 8 evaluated. 102 Table 41. Erect Group: Agronomic evaluation of a collection of Desmodium velutinum in Quilichao. Data of four evaluation cuts (two in the dry season and two in the wet season). Grey underlaid: Accessions with digestibility > 67 % and dry matter yield >200 g/plant. Mean Mean Regrowing DM Regrowing DM Accession Height Diameter points yields Height Diameter points yields No. CIAT Wet Dry (cm) (No.) (g/pl) (cm) (No.) (g/pl) 33443 102 178 54 300 85 163 73 237 23985 109 151 41 230 105 143 76 215 33352 97 143 51 195 88 132 78 207 13953 99 152 61 301 88 140 60 204 23994 92 152 65 201 84 144 64 204 33138 108 134 48 185 106 107 69 194 23081 105 142 68 244 84 130 74 177 23086 90 143 62 184 78 136 72 170 33247 89 153 51 177 83 131 74 170 23136 112 143 43 151 103 123 52 158 23079 104 127 45 172 95 117 57 155 23989 112 134 37 165 101 115 46 153 D2430 80 133 53 145 81 129 65 141 23083 93 124 53 136 85 118 78 140 23132 98 131 38 129 90 107 56 134 23988 95 157 50 168 80 138 71 133 D3456 93 138 33 145 80 134 71 126 13947 108 119 42 159 93 112 59 125 13391 93 124 48 131 80 111 62 120 23133 118 131 35 148 98 108 47 116 13948 90 120 35 119 85 115 56 113 D81995 85 141 44 185 73 119 57 105 13222 90 119 37 132 78 120 56 105 23929 92 122 40 113 79 112 53 103 33250 94 125 43 121 81 101 62 103 23135 102 129 34 109 82 100 56 101 23930 77 135 42 92 73 133 56 100 14314 102 121 42 138 83 96 52 98 33254 85 134 36 117 74 112 53 93 23084 82 112 42 85 79 110 61 92 23158 89 135 39 91 84 132 53 89 23667 94 122 36 86 87 103 42 81 23272 80 128 48 82 78 124 53 79 23987 93 125 48 100 81 112 65 77 13945 105 102 29 102 94 97 47 74 23320 82 125 45 75 86 113 48 69 23669 86 103 41 67 83 101 49 69 23157 103 114 37 102 94 100 42 68 13954 88 121 34 104 70 111 46 68 23160 85 126 36 84 78 119 48 68 23324 85 130 43 70 85 120 50 65 D6 75 132 33 95 71 87 33 64 23322 77 132 43 64 78 119 62 64 23274 76 124 48 91 71 111 52 59 23326 89 123 46 66 80 107 48 55 23248 74 133 48 81 67 107 63 53 13221 95 119 32 57 84 106 32 52 23134 107 103 25 71 85 75 30 49 23321 79 112 33 41 79 114 46 46 23319 63 103 37 51 55 95 46 37 33255 74 83 17 32 70 68 23 33 D7NAPRI 67 110 24 77 56 86 24 30 23323 68 100 26 27 73 104 34 20 23082 71 118 34 44 67 92 42 19 Mean 91 128 42 123 82 114 55 105 LSD(P< 0.05) 112.099 124.81 103 Table 42. Semi-Erect Group: Agronomic evaluation of a collection of Desmodium velutinum in Quilichao. Data of four evaluation cuts (two in the dry season and two in the wet season). Grey underlaid: Accessions with digestibility >66 % and dry matter yield >190 g/plant. Accession Height Diameter Regrowing Mean DM Height Diameter Regrowing Mean DM No. CIAT point yields points yields Wet Dry (cm) (No.) (g/pl) (cm) (No.) (g/pl) 13218 78 163 61 218 73 148 61 225 23983 81 179 37 223 71 169 76 200 23982 90 166 51 225 79 148 72 198 23981 86 173 59 281 77 155 69 194 23276 70 171 51 224 60 172 81 194 33463 79 164 63 241 74 149 73 193 33003 98 162 53 209 87 152 79 177 33396 70 179 52 207 62 146 65 177 23928 87 173 55 217 76 159 70 171 23996 90 173 51 325 72 153 71 169 33459 88 154 59 172 88 144 65 168 33451 69 177 41 230 64 167 70 164 23920 69 144 52 210 63 151 78 161 33428 82 155 48 184 79 151 81 159 13216 75 162 32 185 73 150 81 159 23991 68 166 46 181 62 155 71 155 23923 82 156 54 176 79 140 55 154 23977 82 169 45 195 77 156 68 151 23325 86 160 61 228 74 135 62 146 13691 84 165 54 227 73 139 68 146 23980 88 174 59 197 74 147 73 142 23922 76 150 45 133 66 138 69 129 13220 100 148 41 176 69 123 58 127 23279 96 164 59 181 97 160 63 123 33242 84 142 41 123 73 126 60 122 23993 68 163 51 154 55 151 68 121 13219 67 161 51 141 59 144 48 115 23973 56 154 41 153 59 144 62 112 23927 86 136 42 122 80 129 60 111 23275 73 138 48 176 64 120 61 110 23921 76 159 56 224 67 148 52 106 13417 78 150 38 116 74 142 48 105 13952 78 134 51 141 75 129 37 105 33356 79 146 47 125 65 139 63 103 23975 86 145 52 179 67 117 74 103 13692 53 158 54 154 56 130 55 102 23986 77 153 62 160 63 129 73 99 13227 78 152 44 121 66 133 63 98 33249 71 145 44 115 58 111 52 90 23979 87 162 53 145 78 128 67 88 23278 92 161 53 116 83 142 56 87 23080 69 149 40 122 68 137 55 86 23668 82 131 38 87 65 123 58 84 13526 69 149 34 126 66 133 46 83 23974 63 143 50 142 54 124 61 82 23995 68 151 46 96 62 133 64 79 13676 94 146 51 95 77 128 62 76 23282 73 125 47 76 72 119 59 72 33401 77 142 46 112 68 115 52 72 13204 64 148 52 179 56 118 74 68 33464 80 130 40 100 71 116 50 67 23271 80 141 48 80 70 117 48 66 23277 90 135 44 91 75 111 58 64 23280 59 95 26 48 48 87 35 59 23992 75 140 45 104 61 110 59 58 33387 71 149 52 89 58 116 53 57 23327 74 129 41 88 70 117 59 56 23976 75 143 46 92 52 120 48 54 13690 73 137 54 125 58 120 69 49 23915 61 113 50 71 54 97 53 44 23273 74 113 37 59 53 88 46 41 23926 61 111 49 70 56 113 60 41 23925 69 111 41 69 53 106 45 37 13207 64 107 33 46 56 94 44 29 23924 49 83 33 30 47 84 44 19 23281 51 76 29 14 47 73 35 15 Mean 76 147 47 147 67 131 61 109 LSD(P<0.05) 113.1 110.6 104 Table 43. Prostrate Group: Agronomic evaluation of a collection of Desmodium velutinum in Quilichao. Data of four evaluation cuts (two in the dry season and two in the wet season). Accession Regrowing Mean DM Regrowing Mean DM No. CIAT Height Diameter points yields Height Diameter points yields Wet Dry (cm) (No.) (g/pl) (cm) (No.) (g/pl) 33520 44 171 41 179 42 154 59 143 13694 50 162 49 124 56 156 68 132 33481 42 174 45 158 45 154 59 121 13213 51 152 45 142 58 134 61 118 13215 53 167 50 148 48 151 68 94 33484 48 161 42 119 47 152 62 92 13693 41 160 51 127 38 142 58 92 13697 36 173 52 199 52 150 64 92 13212 52 163 54 206 45 141 55 91 33471 61 169 56 164 57 151 59 91 23990 56 168 43 126 50 158 54 82 13214 28 152 48 110 32 141 58 77 13695 46 162 50 145 37 151 53 75 13687 38 147 37 115 41 132 49 74 13688 53 139 42 81 34 132 52 70 13211 44 155 39 82 42 129 51 68 13217 37 147 48 171 38 144 45 58 Mean 46 160 47 141 45 145 57 92 LSD(P<0.05) 92.3 64.5 Table 44. Forage quality of selected accessions (erect group) of Desmodium velutinum evaluated in Quilichao, 2003-2004. Accession IVDMD CP NDF ADF % 23988 73.2 22.5 29.8 21.4 33451 70.6 22.7 30.8 21.6 23986 70.1 24.2 32.2 20.6 23994 69.7 22.0 32.8 23.9 23982 69.7 23.3 31.5 21.9 23981 69.2 21.7 34.4 23.8 23985 68.6 21.8 33.0 23.1 33463 68.3 23.4 36.0 25.8 13953 68.0 19.9 31.5 23.3 23921 67.9 21.9 33.8 24.0 23996 66.8 20.9 38.7 26.5 33443 66.8 20.5 37.1 28.2 23325 66.5 22.9 32.9 23.9 13218 66.1 20.9 36.3 25.1 23275 66.1 21.9 33.2 23.2 13691 65.8 21.6 35.4 25.2 23928 65.8 21.7 34.6 24.8 23081 64.9 20.9 36.5 26.6 23086 64.1 20.7 38.2 26.8 13952 63.8 19.3 35.2 25.6 Mean 67.8 21.6 34.2 24.3 105 3.5.3 Evaluation of a core collection of Canavalia brasiliensis for multipurpose uses in Santander de Quilichao, Colombia, 2004 Contributors: M. Peters, R. Schultze-Kraft (University of Hohenheim), Luis H. Franco, B. Hincapié, and G. Ramírez (CIAT) Rationale Canavalia brasiliensis Mart. ex Benth. (“Brazilian jackbean”) is a weakly perennial, prostrate to twining herbaceous legume with a wide natural distribution in the New World tropics and subtropics. In comparison with C. ensiformis (“jackbean”), research reports on C. brasiliensis are scattered and restricted to studies done in Latin America. The species develops a dense and extensive, deep-reaching root system and subsequently tolerates a 5 month dry period. Based on studies that generally were done with only one genotype, it is adapted to a wide range of soils, including very acid, low-fertility soils. Its main use is as green manure, for fallow improvement and erosion control. Due to medium biomass decomposition, nutrient release of C. brasiliensis green manure has the potential to synchronize well with the nutrient demand of the succeeding crop and leads to high N recovery rates. In Central America, the legume is being used to improve the value of stubble grazing in the dry Photo 19. Canavalia brasiliensis sown at Quilichao, season. Antinutritive substances such as toxic 2004 amino acids (e.g., canavanin), lectins (e.g., concanavalin Br) and trypsin inhibitors have been Results and Discussion reported in seeds of C. brasiliensis. However, there is little information on the nutritive value of Plants established quickly, and incidence of pest the herbage of this species for ruminants. and diseases was low, in particular when compared to a collection of Canavalia sp. planted at the Materials and Methods same time in Quilichao. 12 weeks alter transplanting the majority of accessions had soil A total of 53 accessions of Canavalia covers above 70%, with accessions CIAT 808, brasiliensis, mostly from Latin America, were 18515, 7319, 7648, 7970, 8557, 2009517008,17009 sown into jiffy pots and transplanted 4 weeks and 20096 showing the highest values (Table 45). later to the field at CIAT’s research station in Santander de Quilichao. A total of 6 plants per Forage yields 16 weeks after transplanting varied plot were sown, in a Randomized Complete significantly (P< 0.01) among accessions, with Block Design with 4 replications. DM yield, yields above 4 t DM /ha recorded for accessions drought tolerance and forage quality parameters CIAT 808, 17009, 8557, 17012, 20098, 21824, are measured (Photo 19). 18515, 20303,17973, 7178, 20306, 7648 and 7319. 106 Table 45. Vigor, soil cover (%) and dry matter yields of Canavalia brasiliensis in Quilichao, 2004. 6 Weeks 12 Weeks 16 Weeks Accession Vigor Soil Vigor Soil Vigor Soil Mean dry matter cover cover cover yields 1-5 % 1-5 % 1-5 % kg/ha 808 4 50 5 92 5 100 6320 17009 4 43 4 85 5 100 5333 8557 4 47 5 87 5 97 4693 17012 3 35 3 78 4 90 4560 21824 3 40 3 73 3 78 4440 20098 3 33 3 75 4 80 4440 18515 3 45 4 88 5 100 4333 20303 3 35 3 77 4 80 4333 17973 3 32 3 68 4 75 4293 7178 4 40 3 73 4 82 4200 20306 3 30 3 72 4 85 4187 7648 4 52 4 87 5 100 4107 7319 4 52 4 87 4 90 4067 20095 3 47 4 87 4 93 3907 7969 3 35 4 80 5 93 3880 22132 3 40 3 73 4 77 3840 7321 3 35 3 78 4 87 3787 7973 3 40 4 77 4 80 3760 18501 3 38 3 70 4 75 3720 17008 4 42 4 85 4 80 3680 8770 3 38 3 75 5 100 3680 19034 3 43 4 83 4 85 3653 17462 3 40 4 80 5 97 3627 905 3 43 3 63 3 67 3613 7647 3 37 4 77 4 87 3507 7971 3 30 3 72 4 82 3400 7175 2 27 3 63 3 72 3373 19029 4 43 4 83 4 83 3320 20518 3 42 4 83 4 80 3147 20096 3 37 4 85 4 83 3120 20516 3 35 3 73 4 83 3053 21825 3 33 3 73 4 82 3040 17010 3 35 3 70 4 80 2933 19035 3 35 3 72 3 63 2893 20514 4 52 3 83 3 80 2840 19361 4 43 3 78 4 88 2813 17011 3 40 3 80 3 85 2813 7972 3 32 4 80 4 90 2787 20513 4 50 4 75 3 82 2773 20301 3 27 3 78 3 83 2720 8768 3 32 3 63 3 72 2653 7970 3 43 4 87 4 90 2640 20295 3 35 3 77 3 77 2627 19359 3 37 3 73 3 73 2600 9146 3 37 3 75 3 80 2560 20304 2 28 3 63 3 75 2560 19632 3 27 3 65 3 67 2440 20296 2 23 3 60 2 67 2360 20090 2 22 3 70 3 70 2333 5033 3 38 3 70 3 80 2227 7174 3 32 3 67 3 70 2227 17828 3 28 4 68 3 63 2160 7894 3 28 2 55 3 72 1813 Mean 3.1 37 3.4 76 3.8 82 3400 Range 2-5 20-70 2-5 30-100 1-5 30-100 1000-8600 LSD (P< 0.05) 2855.21 107 Accession CIAT 17009 is a line previously tolerance. Despite a severe dry period all selected by farmers in Central America for use accessions remained healthy, with soil covers as green manure and to improve fallows and crop above 80%; in a few materials some leaf loss residues, due to its high yield and drought under drought conditions was observed. 3.5.4 Evaluation of a core collection of Canavalia sp. for multipurpose uses in Santander de Quilichao, Colombia, 2004 Contributors: M. Peters, Luis H. Franco, R. Schultze-Kraft (University of Hohenheim), B. Hincapié, G. Ramírez, (CIAT) Rationale In view of the promising results obtained with Canavalia brasiliensis, there is an interest to define the potential of other species of Canavalia for use mainly as green manure and for fallow improvement in low fertility, drought prone environments. Materials and Methods A total of 47 accessions of Canavalia sp, originating from Latin America, China and Thailand were sown into jiffy pots and transplanted to the field in Santander de Quilichao (Photo 20). The design and variables are the same as described for C. brasiliensis. Establishment of most accessions was slow, with soil covers below 53% after 12 weeks of transplanting. On the other hand, 16 weeks after transplanting, only 11 materials had soil covers above 80%. Several materials appear not to be well adapted to the acid soils in Quilichao and were severely affected by pests and diseases. Accessions with the best adaptation during establishment phase were CIAT 21012, 21014, 19038, 21209, 7317, 7383, 8719, 21013, 21211 and Photo 20. Canavalia sp. sown at Quilichao, 2004. 18587, all of which are showing good drought tolerance (Table 46). 108 Table 46. Vigor and soil cover (%) of Canavalia sp. in Quilichao, 2004. Number 6 Weeks 12 Weeks 16 Weeks Accession Vigor Soil cover Vigor Soil cover Vigor Soil cover 1-5 % 1-5 % 1-5 % 19038 5 55 5 90 4 100 21209 5 52 5 80 4 97 7317 5 50 4 83 4 93 21012 5 42 4 60 5 92 21014 4 40 4 57 5 90 8719 5 57 5 83 4 87 18271 5 53 4 73 3 87 7383 4 40 4 67 4 87 7318 5 58 4 80 3 83 21013 4 37 4 63 4 83 21211 4 33 4 63 4 82 7322 5 58 4 72 3 78 19033 5 45 4 70 3 78 20803 4 42 4 63 2 75 19031 4 42 4 60 3 73 20307 4 40 4 73 3 73 18587 4 33 4 50 4 73 18272 5 52 4 73 3 72 20145 4 37 3 47 3 72 18580 3 32 3 47 3 72 22031 4 35 3 50 3 70 20691 4 32 3 50 2 70 20748 3 28 4 57 2 70 8769 4 40 4 57 2 68 19032 4 37 4 63 3 68 20305 4 32 3 50 2 68 20298 4 42 4 70 3 67 19357 4 35 4 67 2 67 18270 4 45 4 60 3 65 21210 4 45 4 57 3 65 17929 4 45 4 70 2 63 17451 4 38 4 57 2 63 8771 3 30 3 43 2 63 21212 4 32 3 40 2 60 8185 4 45 3 43 2 57 20113 4 40 3 47 2 57 21487 3 28 3 47 2 57 19052 2 17 3 33 3 57 18268 2 11 2 20 2 37 19356 2 13 2 20 2 33 18258 2 18 2 13 2 28 18261 2 8 2 13 1 28 18263 2 8 2 13 1 25 20300 4 37 3 40 1 23 20093 3 32 2 20 1 10 20297 2 7 2 10 1 10 20299 2 6 2 10 1 5 Mean 3.7 36 3.5 53 2.6 64 Range 1-5 1-70 1-5 10-95 1-5 5-100 3.6 Annual legumes for multipurpose use in different agroecosystems and production systems Highlights • Selected an accession (IT95K-52-34) of cowpea with superior grain yield as compared to local checks in acid infertile and fertile soils. 109 • Showed that cowpea as green manure can substitute the N applied (80 kg) to maize by farmers in hillside of Nicaragua. • Selected an accession (CPI-67639-early flowering) of Lablab in hillsides of Nicaragua based on rapid establishment, high cover and high seed yield. 3.6.1 Evaluation of new Vigna unguiculata accessions in Quilichao and Palmira, Colombia Contributors: M. Peters, Luis H. Franco, B. Hincapié, G. Ramírez, (CIAT), R. Schultze-Kraft (University of Hohenheim) and B.B. Singh (IITA, Nigeria) Rationale remains to be seen if cultural traditions allow for the inclusion of cowpeas in the daily menu of Cowpea (Vigna unguiculata) is utilized in the people in Central America. subhumid/semi-arid tropics of West Africa and India as a source of food and feed for livestock, Materials and Methods but the utilization of cowpea in Latin America is so far limited. We visualize that, cowpea could be A new collection of cowpea obtained from IITA an alternative crop for the second planting season was sown in Santander de Quilichao and Palmira in the central hillsides region of Nicaragua and in 2004 in order to select accession with both high Honduras where the legume could provide not forage and grain yields and good adaptation to only higher grain yields as compared to common contrasting soils. (Photo 21). Our previous beans, but could also allow for a third crop in selection criteria for cowpea had been mainly November/December in order to provide hay as forage yield in alkaline and acid soils. The same animal feed in the dry season or contribute to soil evaluation methodology as presented in previous fertility enhancement for the following maize annual reports was utilized. The main variables crop. Cowpea could also be used for hay, silage measured are forage production and quality, grain and feed meal production, which in turn could be yield and effect as green manure on a subsequent an option for income generation by smallholder maize crop. A particular emphasis is given to livestock and non-livestock owners. material adapted to a wide range of soils. Good adaptation to climatic and edaphic Results and Discussion conditions, especially to water stress, are prerequisites for a successful development of The collection established was highly diverse in cowpea as an option for the traditional maize- terms of flowering response, with very early to bean cropping systems in Central America. It very late accessions present. The differential flowering pattern will be taken into account for the planning of subsequent trials. In Palmira it was necessary to replant one replication because of negative effects of waterlogging and consumption of seed by birds. Soil cover was more rapid in Palmira than in Quilichao, with a mean of 80% and 64% covered respectively 10 weeks after planting. Pest and diseases were present in both sites but did not limit the development and productivity Photo 21. Cowpea (Vigna unguiculata) at Quilichao. of plants. Grain was harvested 12 110 weeks after planting when pods where dry. Mean achieved with local checks (CIDICCO3, yields in Quilichao were above 2 t/ha and CIDICCO4), and with the new accession, CIAT significant (P<0.05) differences among 9611, IT95K-52-34 (Table 47). accessions were measured. In terms of forage quality, significant (P<0.01) The accessions IT97K-1069-6, IT95K-52-34 and differences among accessions were measured IT98K-412-8, had yields above 3 t/ha. Mean for IVDMD, P and K, but not for CP. In vitro grain yields in Palmira were double (4 t/ha) than dry matter digestibility was above 83% while CP those obtained in Quilichao. Differences among contents ranged between 19.7% and 24.2 %, accessions were significant (P<0.01) and the confirming the high quality of the forage from highest grains yields with more than 5 t/ha were cowpea (Table 48). Table 47. Vigor, soil cover (%) and yield (kg/ha) of Vigna unguiculata grain in Quilichao and Palmira, 2004. Quilichao (acid infertile soils) Palmira (fertile soils) Accession Vigor Cover (%) Grain Vigor Cover (%) Grain (kg/ha) (kg/ha)))) 1 - 5 10 weeks 12 weeks 1 - 5 10 weeks 12 weeks IT97K-1069-6 5 85 3327 4 87 4360 IT95K-52-34 5 92 3313 5 100 5513 IT98K-412-8 4 72 3180 4 83 4900 IT98K-131-2 4 67 2873 3 77 4873 IT97K-819-118 4 68 2867 3 73 2460 IT98K-406-2 4 72 2607 3 77 3847 CIDICCO 3 (local check) 5 82 2560 5 100 6013 IT97K-1069-2 4 73 2533 4 90 4087 IT97K-818-35 4 72 2460 3 77 3020 CIDICCO 2 (local check) 5 85 2420 5 93 4653 IT99K-7-14 3 63 2360 3 63 3267 IT99K-409-8 4 77 2347 4 87 3440 IT99K-429-2 3 53 2327 3 67 3120 IT99K-1060 2 47 2280 3 62 3760 IT97K-825-3 4 75 2267 3 67 2853 IT98K-476-8 4 73 2260 3 73 3553 CIAT 9611 (local check) 4 60 2167 5 93 5433 IT98K-391-2 4 72 2153 5 93 4553 IT98D-1399 3 53 2087 3 60 3440 IT98K-412-13 4 67 1993 3 73 4533 IT96D-610 4 65 1987 4 83 4327 CIDICCO 1 (local check) 4 75 1980 5 93 4613 IT98K-428-3 4 70 1953 4 77 3333 CIDICCO 4 (local check) 5 70 1927 5 100 5533 IT97K-570-18 4 77 1820 5 93 5533 IT98K-506-1 3 55 1733 3 77 3913 IT97K-356-1 3 20 1673 3 70 3913 IT98K-205-8 3 53 1660 3 67 3800 IT97K-499-38 4 60 1627 4 78 2980 IT98K-390-2 4 63 1533 5 100 4233 IT99K-7-21-2-2 2 37 1520 3 67 2833 IT89KD-2881 2 43 1413 3 77 3480 IT97K-494-3 2 47 1380 3 70 2527 IT99K-216-24-2 2 53 1367 3 53 4153 IT99K-1122 2 50 1327 4 90 4233 IT97K-461-4 3 50 1260 5 87 4293 FHIA (local check) 2 43 1093 4 83 4487 IT98K-311-8-2 4 80 1027 4 90 3880 Mean 3.8 64 2070 3.8 80 4020 MSD (P<0.05) 1810.2 3353.4 1 Accession selected from the first core collection from IITA evaluated in Palmira and Quilichao. 111 Table 48. Quality of the forage in accessions of Vigna unguiculata grown in Quilichao, 2004. Number Forage Accession IVDMD Protein P K % IT98K-131-2 89.8 21 0.16 1.46 IT97K-825-3 89.2 20.6 0.16 1.85 IT96D-610 88.8 19.8 0.13 1.21 CIAT 9611 (local check) 88.8 20.2 0.16 1.44 FHIA (local check) 88.6 19.4 0.13 1.57 IT98K-311-8-2 88.4 19.4 0.15 1.44 IT98K-476-8 88.4 19.9 0.12 1.77 IT99K-7-14 88.3 20.5 0.14 1.29 IT89KD-288 87.9 20.4 0.12 1.45 IT99K-216-24-2 87.8 21.5 0.16 1.51 IT98K-205-8 87.5 20.4 0.14 1.44 IT95K-52-34 87.5 19.7 0.14 1.44 IT97K-819-118 86.8 20.8 0.16 1.64 IT97K-570-18 86.5 20.0 0.13 1.49 CIDICO 2 (local check) 86.8 20.7 0.14 1.59 IT97K-356-1 86.7 20.3 0.15 1.54 IT97K-818-35 86.7 20.9 0.15 1.44 IT97K-499-38 86.6 22.2 0.15 1.78 IT99K-429-2 86.6 20.3 0.14 1.43 IT99K-1122 86.5 23.1 0.20 1.95 IT98K-506-1 86.4 21.1 0.16 1.69 IT97K-1069-2 86.3 20.9 0.17 1.37 IT98K-412-13 86.2 22.1 0.16 1.77 IT97K-1069-6 85.7 23.5 0.18 1.73 IT97K-461-4 85.5 22.4 0.16 1.69 IT97K-494-3 85.3 22.7 0.17 1.86 CIDICO 4 (local check) 85.1 22.1 0.15 1.80 IT98K-412-8 85.1 22.6 0.16 1.93 IT98K-428-3 85.1 22.5 0.16 1.5 IT98D-1399 85.0 20.5 0.13 1.77 IT99K-7-21-2-2 84.7 22.6 0.16 1.95 CIDICO 1 (local check) 84.7 21.1 0.14 1.77 IT98K-390-2 84.6 21.7 0.15 1.69 IT98K-391-2 84.6 20.6 0.10 1.78 CIDICO 3 (local check) 84.5 23.6 0.14 1.81 IT99K-1060 83.7 21.6 0.15 1.5 IT99K-409-8 83.7 22.5 0.15 1.7 YT98K-406-2 83.0 24.2 0.21 1.9 Mean 86.4 21.3 0.15 1.64 LSD 4.79 5.74 0.08 0.71 (P< 0.01) (P< 0.07) (P< 0.01) (P< 0.01) 3.6.2 Evaluation of Vigna unguiculata germplasm in Nicaragua Collaborators: A. Schmidt, C. Davies, E. López, M. Peters, L.H. Franco, and G. Ramirez (CIAT) Rationale fertility soils in hillsides of Nicaragua and Honduras. This legume could allow for a third We visualize that cowpea could be an alternative crop in November/December the maize/bean crop for a) the first planting season (“primera”) systems in order to provide grain, hay for animal as a soil improving starter crop for maize, and b) feeding in the dry season or contribute to soil for the second planting season (“postrera”) in low fertility enhancement for the following maize 112 crop. Cowpea could also be used for hay, silage Results and Discussion and feed meal, which in turn could be an option for income generation by smallholder livestock In Table 49 maize plant height, DM and grain and non-livestock owners. Adaptation to climatic yields are presented. No significant differences and edaphic conditions, especially to water stress, (P>0.05) were detected among accessions or are prerequisites for a successful development of between treatments (mulch vs. grain harvested). a cowpea option for the traditional maize-bean Mean plant height, DM and grain yield are in-line cropping systems in Central America. with farmer maize crops in San Dionisio. Farmers, depending on their economic Materials and Methods possibilities, apply up to 80 kg/ha N to their maize crops. Results of our experiment showed that a During 2002 a small core collection of Vigna cowpea crop can easily replace the application of unguiculata from IITA comprising 14 these amounts of nitrogen, even as a preceding accessions, which were selected in Quilichao and dry season crop. This corroborates our findings Palmira for good adaptation to soils (acid and from on-farm experiments, reported in AR 2003 alkaline) was introduced into Nicaragua. In (p. 145), where we argued that with a legume October 2002 the core collection, complemented crop planted at the end of the rainy season, with 5 Central American accessions, was planted traditional nitrogen fertilizer applications for the out in small plots at the SOL seco site in San following maize crop can be substituted in the dry Dionisio, Matagalpa, Nicaragua (Lat N 12° 45' hillsides of Nicaragua and Honduras. 05.8", Long. W 85° 53' 16.5", Alt. 537 masl, rainfall 990 mm/a, mean temp. 26°C). Standard Our results this year with cowpea indicate that evaluation procedures were applied as in previous farmers can not only reduce production cost of years. At flowering 50% of each plot was cut maize, but also have an additional legume grain and biomass kept on plot surface as mulch; from harvest for human consumption or animal the remaining 50% grain was harvested in feeding. Since no significant differences between December 2002. Crop residues were equally kept plots with mulched or grain-harvested cowpeas on plot surface for the rest of the dry season. were found, farmers can choose their cowpea The respective results were reported in AR 2003, accession based on their utilization preference. p.116). Upon the outset of the rainy season 2003 a maize crop was established in each plot. Total In Table 50 we present the results of dry matter and grain production were recorded in participatory evaluations of the cowpea collection November 2003 in order to detect possible in Nicaragua. Selection criteria employed by residual fertility effects from the preceding Vigna farmers were: grain yield, leafiness and leaf size, unguiculata accessions. Additional plots were plant vigor, pod size, and plant height. Of the established where 45, 80, 140, 200 kg/ha N, accessions selected, the local accessions Rojo, respectively, were applied. A non-fertilized (N 0) INTA and Negro are leafy types for animal treatment was also included. feeding, while the introduced IT90K-284/2 accession is a dual purpose type as shown before Early 2003, due to increasing interest in Vigna (see AR 2003, p. 117). This accession is widely unguiculata by farmers in San Dionisio (who adapted across different environments in demanded the re-establishment of our collection Colombia and Central America. for further participatory evaluations and for seed increase for on-farm evaluation) the accessions Seed production began in the first planting season were re-established in 6m x 6m plots without 2004 with the production of 50 kg of accession replicates. After 2 field events with participatory Rojo and IT90K-284/2. In the second planting evaluations (n=21 persons) plots were harvested season seed production efforts will be increased. and seed distributed to 35 farmers. Evaluation of new accessions, both local and introduced materials, will continue throughout 113 2004-2005. With regard to soil fertility effects TSBF/Soils Group in Central America (see also resulting from cowpea as a green manure, we 4.1.7 in this report). plan to intensify our collaborative work with the Table 49. Plant height, DM and grain yields of maize planted after Vigna unguiculata (cowpea) at San Dionisio, Nicaragua, 2003. Maize yields - Cowpea mulch Maize yields - Cowpea grain harvested Accession Plant DM total Grain yield Plant height DM total Grain yield height (kg/ha) (kg/ha) (cm) (kg/ha) (kg/ha) (cm) IT95K-1088/2 194 7999 3385 198 6512 3255 IT95K-1088/4 191 8279 3149 184 6408 2041 IT90K-277/2 198 8278 3284 189 6873 2785 IT90K-284/2 188 5763 2844 172 4967 2213 IT89KD-288 202 10206 3528 190 8930 4209 IT89KD-391 200 6858 2699 201 8448 4048 IT93K-503/1 181 6408 3224 176 7910 3448 IT93K-573/5 191 6185 2792 199 7611 3006 IT93K-637/1 194 7029 3050 209 9095 4516 IT86D-715 204 7937 3416 191 5643 3008 IT86D-716 198 6772 2015 200 7534 3294 IT86D-719 201 6537 4014 192 7735 3422 IT6D-733 187 5607 2824 197 8367 3804 IT96D-740 197 6188 2683 193 7628 3094 Café 212 8163 4082 208 8399 3671 INTA 199 6955 2873 196 9115 3253 Negro 209 8129 3845 185 8124 3905 Rojo 190 8667 3432 210 8688 3770 SF Libre 201 6461 2777 189 7015 2886 Mean Acc. 197 7285 3153 194 7632 3349 LSD (P<0.05) 46.9 6873.9 3044.9 42.1 4619.7 2349 Nitrogen treatments N 0 180 4324 2239 N 45 179 5104 2621 N 80 177 6668 3085 N 140 164 7169 4060 N 200 184 7612 4368 Table 50. Farmers ranking of Vigna unguiculata accessions and their potential uses expressed by farmers in San Dionisio, Nicaragua, 2003. Framers ranking Potential use expressed by farmers Good Regular Animal feed Grain production Rojo IT86D-715 Rojo Rojo Café IT86D-719 INTA IT90K-284/2 IT96D-740 IT93K-637/1 Negro IT89KD-391 IT6D-733 INTA IT95K-1088/2 Negro IT93K-573/5 IT90K-284/2 114 3.6.3 Evaluation of Lablab purpureus germplasm in Nicaragua Contributors: A. Schmidt, C. Davies, E. Lopez, M. Peters, L.H. Franco, G. Ramirez (CIAT) Rationale vigour, flowering patterns, incidence of pests and diseases, and seed production. No fertilizer was A major problem facing livestock producers in applied throughout the experiment. Dry matter yields Central America is inadequate animal nutrition prior to flowering were not recorded due to the small during the dry season when pastures, sorghum and initial amount of seed introduced. Priority was given maize stover are limiting in quality. Problems such to genotype characterization and seed increase. as sickness and weight loss due to a poor nutrition are frequent. One way for improving the utilisation Result and Discussion of such crop residues is by adequate supplementation with leguminous forages of high Accessions of the core collection established well at quality. the experimental site with an average of 79% of all seeds emerging (Table 51). Plants reached an The legume Lablab purpureus is recognized not average height of 49 cm and covered 12 weeks after only as drought resistant, but also for its adaptability planting on average 74% of the plots. Plants showed to a wide range of environmental conditions. good vigour and no incidence of pests or diseases Though the legume is widely known in Central throughout the experiment, with the exception of America under a number of names (e.g. dolichos, accession CPI-36903. caballero) and has the capability of being an outstanding resource for crop-livestock systems in Accession CPI-67639 (early flowering) performed this region, (e.g. the legume can be used as cover outstanding in this experiment in terms of crop, grazed in a pasture setting or as a companion establishment, soil cover and seed yield. It remains to crop to maize, cut as hay, or mixed with corn be seen if this seed high yield is correlated with low silage), it is not being used to its full potential. So far biomass production. Biomass production and on soil only two commercial lines have been available to fertility effects on subsequent maize crops will be farmers in Nicaragua, but seed availability remains obtained on larger plots in the postrera 2004 and a major limitation. Thus to select for more primera season 2005. Enough seeds was harvested in productive and better-adapted germplasm for the the present experiment to include an additional dry hillside regions of Nicaragua, a Lablab experimental site in 2004-2005, which will be under purpureus core collection from ILRI/CSIRO is the responsibility of our national partner INTA. currently under evaluation. We conclude from this first experiment with Lablab Material and Methods purpureus that the accessions evaluated adapted well to the dry conditions of the central region of During 2003 a core collection from ILRI/CSIRO Nicaragua where small farmer crop-livestock comprising 12 accessions was introduced into systems are predominant. Our results showed Nicaragua. In October 2003, accessions were variability within the collection with regard to seed planted out in 2m x 2.5 m small plots in three yield and flowering patterns. Nevertheless, biomass replicates in a randomized complete block design at production data and the results from participatory the SOL seco site in San Dionisio, Matagalpa, evaluation by farmers will determine which accession Nicaragua (Lat N 12° 45' 05.8", Long. W 85° 53' will be multiplied for forage production, soil 16.5", Alt. 537 masl, rainfall 990 mm/a, mean improvement or grain production. The selected temperature 26°C). After an initial evaluation of Lablab accessions will be an additional annual plant emergence, accessions were evaluated in a legume alternative to farmers in the drier regions of two-weeks interval for plant height, soil cover, plant Central America. 115 Table 51. Plant emergence, plant height, soil cover, plant vigour, flowering behaviour and seed yield of a Lablab purpureus core collection at San Dionisio, Nicaragua, 2003-2004 Accessions CIAT Plant Plant Soil Plant Vigour Flowering Seed Yield No. Emergence height (cm) cover (%) (1-5) (Early/Late) (g/m2) (%) CPI-67639 17197 100 41 97 4 E 127 CPI-34777 22598 67 59 62 4 L 113 21603 21603 77 44 70 5 L 92 I-14442 22768 88 50 87 5 L 83 CQ-2975 22735 90 47 83 5 E 82 L-987 22660 97 60 91 5 E 74 CPI-52535 22604 63 42 63 3 L 70 cv. Highworth 22660 85 56 83 4 E 70 I-11632 22764 90 50 72 4 L 43 I-6533 22770 85 51 85 5 L 37 CPI-36903 22653 33 32 30 3 L 34 CPI-106471 22663 77 56 62 4 E 28 Mean 79 49 74 4.3 71 LSD (P<0.05) 78.8 43.5 72.4 81.9 3.6.4 Effect of Lablab purpureus accessions as a green manure in Quilichao and Palmira Contributors: M. Peters, L. H. Franco, B. Hincapié, and G. Ramírez (CIAT) Rationale Materials and Methods Lablab purpureus is a free seeding, fast growing, A multilocational trial to evaluate of Lablab short-term perennial legume, with widespread use purpureus selected from previous work in through the tropics as a fodder plant. In Africa the Colombia was initiated in contrasting sites – soil, use of Lablab for human consumption is also climate and altitude – in Colombia (6 sites), Costa common. The origin of the Lablab germplasm Rica and Nicaragua. In this section we report currently utilized is mainly Eastern/Southern Africa results on effects of Lablab on a succeeding and Asia. In addition, it is well documented that maize crop in Quilichao (acid low fertility soils) Lablab purpureus is best adapted to areas with and Palmira (alkaline high fertility soils) rainfall regimes of 750–2000 mm/year. This species (Photo 22). grows in a variety of soils, but the ideal pH for growing Lablab is reported to be between 5.0 and 7.5. Results and Discussion In order to evaluate the potential of Lablab in Due to the short rainfall cycles in the bimodal tropical America, we obtained a collection available rainfall system prevalent in Palmira and at ILRI/CSIRO. Our main objective with the Quilichao, it is not possible to plant a crop directly collection is to select accessions with broad after lablab as a green manure. Hence an adaptation to different soils and climate conditions alternative strategy was employed, with the in tropical America. However, of immediate lablab green manure being incorporated at the interest is the evaluation of the Lablab collection in end of the dry season, followed by the crop sown acid and neutral soils to define niches for this at the beginning of the next wet season. Maize species as green manure and fodder (especially for DM and grain yields after incorporating lablab as hay and silage or deferred feed), with emphasis on green manure are presented in Table 52. Central America where soils are highly variable in pH. Maize yields in Palmira were higher than in 116 Quilichao. In Palmira there was no positive effect neither of the green manure or N-fertilization, (A) probably due to the inherent high fertility of soils. In contrast, in Quilichao, maize yields without fertilization were only 55% of the yields obtained with 120 kg/ha N. Following incorporation of lablab accessions CIAT 22663 and 21663, maize grain yields were similar to those recorded with the 3 levels of N applied. Table 52. Effect of Lablab purpureus as green manure on biomass and grain yields of a succeeding maize crop (Palmira and Quilichao, 2004). Palmira Quilichao Accessions (fertile soil) (acid infertile soil) CIAT No. Yield (kg/ha) DM Grain DM Grain (B) Plant Maize Plant Maize 22653 5648 6014 4548 4336 22768 3366 5319 4665 4554 22766 4411 5282 4381 4231 22762 4026 5162 4047 3268 22660 4385 4627 4242 3983 17197 4156 4870 4375 4080 22770 3591 4796 4175 3886 22652 5171 4657 3963 3886 N0 5824 4596 3240 2689 22598 4513 4545 5010 4426 22764 3738 4471 4431 3963 N120 5574 4439 4320 4826 21603 4012 4360 4821 4682 22663 5152 4179 5016 4826 22604 3897 4128 4336 4587 N80 4124 3929 4136 4292 N160 4694 3850 3674 4470 22735 2863 3627 4387 3607 N40 4138 3077 4119 3958 Photo 22. (A) Lablab purpureus as green manure and Fallow 3064 2906 Mean 4398 4558 4266 4087 (B) Maize after Lablab in Palmira. LSD (P<0.05) 3401 3134 3.6.5 Forage and green manure potential of a collection of Mucuna spp. Contributors: M. Peters, L. H. Franco, B. Hincapié, and G. Ramírez (CIAT) Rationale content (see Annual report 2002) were planted in Quilichao (see Photo 23) to be used as a green Mucuna is a legume utilized by farmers in Central manure for a succeeding maize crop. A America (particularly in the humid tropics) as a Randomized Complete Block Design with 3 green manure and rarely as forage. We were replications was employed. interested in determining the variability among Mucuna accessions when used as green manure. Results and Discussion Materials and Methods The establishment of Mucuna was rapid, with the exception of Mucuna sp cv. Rayada-61. Mean Eight accessions of Mucuna sp., obtained from soil cover was 90% 12 weeks after sowing. At CIEPCA, and previously evaluated for L-Dopa 16 weeks soil cover declined to 81% due to some 117 leaf loss. However, soil cover of Mucuna. sp cv. However, highest maize grain yields (6.1 t/ha) were Preta-82 and M. pruriens CIAT 9349 remained achieved after incorporation of M. pruriens cv. stable with 100% soil cover even under drought Utilis-109. Interestingly the lower biomass yielding conditions. accession (cv. Rayada-61) had a better effect as green manure as compared to the higher biomass Biomass yields were above 5.8 t/ha both at 12 accession (cv. Jaspeada –106), probably related to and 16 weeks, with M. pruriens cv. Jaspeada- immobilization of nutrients in the soil. 106, M. pruriens CIAT 9349. These accessions had significantly (Pd<0.01) higher DM yields Digestibilities of Mucuna accessions under after 16 weeks, of growth that other accessions evaluation ranged between 52% and 62%, and (Table 53). digestibility values were below the values obtained for lablab included as a control. The CP contents ranged between17% and 21% (Table 54). Table 53. Soil cover (%) and biomass yield (kg/ha) of Mucuna sp and maize grain yield in Quilichao, 2004. Accession Biomass Mucuna Maize Soil cover (%) DM yield (kg/ha) DM yield (kg/ha) 12 Weeks 16 Weeks 12 Weeks 16 Weeks Plant Grain M. pruriens cv. Cochinchinensis 100 85 7987 5720 11768 5773 M. pruriens cv. Jaspeada-106 100 80 6940 10033 11206 5099 M. sp cv. Ghana-4 70 32 6433 3020 9413 3991 M. sp cv. Rayada-61 53 17 6213 1687 12291 5728 M. pruriens CIAT 9349 100 100 5427 6293 11723 5160 M. pruriens cv. IRZ-99 100 85 5347 7867 11367 5422 L. purpureus CIAT I-14442 77 68 5053 5507 11846 5544 (control)) M. pruriens cv. Utilis-109 100 97 4420 6580 12748 6140 M. sp cv. Preta-82 100 100 4400 6007 10449 4787 Fallow 5720 8676 4259 Mean 90 81 5802 5857 11234 5222 LSD (P<0.05) 3295.41 4788.27 4657.72 2355.44 Table 54. Forage quality of a collection of Mucuna sp evaluated in Quilichao, 2003-2004. Accession PC DIVMS % % L. purpureus (control) 18.1 72.7 M. pruriens cv. Cochinchinensis 20.2 61.6 M. sp cv. Ghana-4 15.1 60.4 M. pruriens cv. IRZ-99 20.8 58.1 M. pruriens cv. Jaspeada-106 16.9 57.9 M. pruriens CIAT 9349 21.8 57.5 M. sp cv. Preta-82 20.0 57.2 M. pruriens cv. Utilis-109 21.4 55.3 M. sp cv. Rayada-61 17.0 51.9 Mean 19.0 59.2 LSD (P<0.05) 6.1 6.6 Photo 23. Mucuna sp. at Quilichao 118 3.6.6 Evaluation of legumes as covers for plantations in the Llanos of Colombia Contributors: C. Plazas, M. Peters, L.H. Franco, B. Hincapie (CIAT) and Oil Palm and Rubber Growers of the Colombian Llanos Rationale Results and Discussion In plantations of the Llanos of Colombia there is a need to find sustainable ways to reduce weed The legume covers were evaluated at the infestation, to maintain and improve soil fertility, beginning of the wet season of 2004. Now 5 to control erosion and increase soil fauna years after planting, several accessions sown biomass. There is currently a trend to promote under high shade conditions in established plantation systems in the Llanos. In the rubber plantations disappeared. However, legume covers plantation the target group for this promotion are established in young plantations with moderate small to medium size farmers who want to shade continue to be vigorous and cover the soil diversify there farming operations. In the oil well. Several A. pintoi accessions have covers palm plantations plots of up to 5 ha are rented above 70%, while D. heterocarpon cv. out to landless farmers to manage the oil palms Maquenque (CIAT 13651) covered 67% of the for the oil palm industry. soil and D. heterocarpon (CIAT 23672) covered 80% of the soil. The control P. phaseoloides In 1999 a range of legume species (Arachis almost disappeared under low shade (Table 55). pintoi, Desmodium heterocarpon subsp. ovalifolium and Pueraria phaseoloides) were Results confirm the utility of D. heterocarpon sown under shade and no-shade conditions in the subsp. ovalifolium and A. pintoi as plantation Meta department of Colombia. covers, provided moderate light in available. These legume when used as covers should be Materials and Methods established in the early phase of the plantation and/or utilized to cover more open spaces In plots of 80 m2 we established legumes between rows of trees, where weeds are a major covers in a commercial rubber (young and old) problem. and oil palm plantations in the savannas and Piedmont areas of the Llanos. The following legumes were sown in a Randomized Block Table 55. Soil cover of different forage legumes under rubber 5 years after sowing, Design with three replications: Arachis pintoi: under high shade and low shade conditions, Llanos of Colombia. 17434, 18744, 18748, 22159, 22160 (seed rate Accession High shade Low shade 10 kg/ha); Desmodium heterocarpon subsp. CIAT No. Soil Vigor Soil Vigor cover cover ovalifolium (D. ovalifolium): 350, 13105, (%) (%) 13110, 13651, 23762 (0.5 kg/ha); Pueraria A. pintoi 17434 17 3 73 4 18744 5 3 73 4 phaseoloides: 8042, 9900 (3 kg/ha). 18748 12 3 60 3 Additionally a mixture of Arachis pintoi CIAT 22159 5 2 72 4 22160 10 2 73 3 18744 and Desmodium ovalifolium CIAT D. heterocarpon 350 0 0 57 3 13105 0 0 52 3 13651 was sown. 13110 2 1 52 3 13651 (cv. Maquenque) 2 1 67 3 23762 0 0 80 3 These plots have been monitored through P. phaseoloides 8042 (control) 0 0 12 2 9900 2 1 2 1 visual observation in regular intervals, to Asoc. A. pintoi – D. heterocarpon 5 2 45 3 assess long-term persistence of legumes sown as plantation covers. 119