Simulating the evolution of glyphosate resistance in grains farming in northern Australia.

Background and Aims: The evolution of resistance to herbicides is a substantial problem in contemporary agriculture. Solutions to this problem generally consist of the use of practices to control the resistant population once it evolves, and/or to institute preventative measures before populations become resistant. Herbicide resistance evolves in populations over years or decades, so predicting the effectiveness of preventative strategies in particular relies on computational modelling approaches. While models of herbicide resistance already exist, none deals with the complex regional variability in the northern Australian sub-tropical grains farming region. For this reason, a new computer model was developed. Methods: The model consists of an age- and stage-structured population model of weeds, with an existing crop model used to simulate plant growth and competition, and extensions to the crop model added to simulate seed bank ecology and population genetics factors. Using awnless barnyard grass (Echinochloa colona) as a test case, the model was used to investigate the likely rate of evolution under conditions expected to produce high selection pressure. Key Results: Simulating continuous summer fallows with glyphosate used as the only means of weed control resulted in predicted resistant weed populations after approx. 15 years. Validation of the model against the paddock history for the first real-world glyphosate-resistant awnless barnyard grass population shows that the model predicted resistance evolution to within a few years of the real situation. Conclusions: This validation work shows that empirical validation of herbicide resistance models is problematic. However, the model simulates the complexities of sub-tropical grains farming in Australia well, and can be used to investigate, generate and improve glyphosate resistance prevention strategies.

Demography of three perennial grasses in a central Queensland eucalypt woodland.

The population dynamics of the palatable, perennial grasses Bothriochloa ewartiana (Domin) C.E.Hubb. (desert Mitchell grass), Chrysopogon fallax S.T.Blake (golden beard grass) and Heteropogon contortus (L.) P.Beauv. ex Roem. & Schult. (black speargrass), were studied in an extensive grazing study conducted in a eucalypt woodland within the Aristida-Bothriochloa pasture community in central Queensland between 1994 and 2000. Treatments were three grazing pressures based on light, medium and heavy utilisation of forage available at the end of summer and two timber treatments (trees intact and trees killed). Seasonal rainfall throughout this study was generally favourable for plant growth with no severe drought periods. Grazing pressure had a greater overall impact on plant dynamics than timber treatment, which had minimal impact. Grazing pressure had a large impact on H. contortus dynamics, an intermediate impact on B. ewartiana and no impact on C. fallax. Fluctuations in plant density of both B. ewartiana and C. fallax were small because both species were long lived with low levels of seedling recruitment and plant death, whereas fluctuations in H. contortus density were relatively high because of its relatively short life span and higher levels of both recruitment and death. Heavy grazing pressure increased the recruitment of B. ewartiana and H. contortus in some years but had no impact on that of C. fallax. Heavy grazing pressure reduced the survival of the original plants of both B. ewartiana and H.contortus but not of C. fallax. For H. contortus, the size of the original plants was larger where trees were killed than where trees were left intact and plants of the 1995 seedling cohort were larger in 1998 at heavy compared with those at light and medium grazing pressure. Grazing had a minor negative impact on the soil seed bank of H. contortus. Populations of all three species remained stable throughout this study, although the favourable seasonal rainfall experienced and the short duration of this study relative to the life span of these species may have masked longer term, deleterious impacts of heavy grazing pressure.

Adaptation and management of Australian buffalograss cultivars for shade and water conservation

Soft-leaf buffalo grass is increasing in popularity as an amenity turfgrass in Australia. This project was instigated to assess the adaptation of and establish management guidelines for its use in Australias vast array of growing environments. There is an extensive selection of soft-leaf buffalo grass cultivars throughout Australia and with the countrys changing climates from temperate in the south to tropical in the north not all cultivars are going to be adapted to all regions. The project evaluated 19 buffalo grass cultivars along with other warm-season grasses including green couch, kikuyu and sweet smother grass. The soft-leaf buffalo grasses were evaluated for their growth and adaptation in a number of regions throughout Australia including Western Australia, Victoria, ACT, NSW and Queensland. The growth habit of the individual cultivars was examined along with their level of shade tolerance, water use, herbicide tolerance, resistance to wear, response to nitrogen applications and growth potential in highly alkaline (pH) soils. The growth habit of the various cultivars currently commercially available in Australia differs considerably from the more robust type that spreads quicker and is thicker in appearance (Sir Walter, Kings Pride, Ned Kelly and Jabiru) to the dwarf types that are shorter and thinner in appearance (AusTine and AusDwarf). Soft-leaf buffalo grass types tested do not differ in water use when compared to old-style common buffalo grass. Thus, soft-leaf buffalo grasses, like other warm-season turfgrass species, are efficient in water use. These grasses also recover after periods of low water availability. Individual cultivar differences were not discernible. In high pH soils (i.e. on alkaline-side) some elements essential for plant growth (e.g. iron and manganese) may be deficient causing turfgrass to appear pale green, and visually unacceptable. When 14 soft-leaf buffalo grass genotypes were grown on a highly alkaline soil (pH 7.5-7.9), cultivars differed in leaf iron, but not in leaf manganese, concentrations. Nitrogen is critical to the production of quality turf. The methods for applying this essential element can be manipulated to minimise the maintenance inputs (mowing) during the peak growing period (summer). By applying the greatest proportion of the turfs total nitrogen requirements in early spring, peak summer growth can be reduced resulting in a corresponding reduction in mowing requirements. Soft-leaf buffalo grass cultivars are more shade and wear tolerant than other warm-season turfgrasses being used by homeowners. There are differences between the individual buffalo grass varieties however. The majority of types currently available would be classified as having moderate levels of shade tolerance and wear reasonably well with good recovery rates. The impact of wear in a shaded environment was not tested and there is a need to investigate this as this is a typical growing environment for many homeowners. The use of herbicides is required to maintain quality soft-leaf buffalo grass turf. The development of softer herbicides for other turfgrasses has seen an increase in their popularity. The buffalo grass cultivars currently available have shown varying levels of susceptibility to the chemicals tested. The majority of the cultivars evaluated have demonstrated low levels of phytotoxicity to the herbicides chlorsulfuron (Glean) and fluroxypyr (Starane and Comet). In general, soft leaf buffalo grasses are varied in their makeup and have demonstrated varying levels of tolerance/susceptibility/adaptation to the conditions they are grown under. Consequently, there is a need to choose the cultivar most suited to the environment it is expected to perform in and the management style it will be exposed to. Future work is required to assess how the structure of the different cultivars impacts on their capacity to tolerate wear, varying shade levels, water use and herbicide tolerance. The development of a growth model may provide the solution.

Rice husk biochar and crop residue amendment in subtropical cropping soils: Effect on biomass production, nitrogen use efficiency and greenhouse gas emissions

We investigated the effect of maize residues and rice husk biochar on biomass production, fertiliser nitrogen recovery (FNR) and nitrous oxide (N2O) emissions for three different subtropical cropping soils. Maize residues at two rates (0 and 10 t ha−1) combined with three rates (0, 15 and 30 t ha-1) of rice husk biochar were added to three soil types in a pot trial with maize plants. Soil N2O emissions were monitored with static chambers for 91 days. Isotopic 15N-labelled urea was applied to the treatments without added crop residues to measure the FNR. Crop residue incorporation significantly reduced N uptake in all treatments but did not affect overall FNR. Rice husk biochar amendment had no effect on plant growth and N uptake but significantly reduced N2O and carbon dioxide (CO2) emissions in two of the three soils. The incorporation of crop residues had a contrasting effect on soil N2O emissions depending on the mineral N status of the soil. The study shows that effects of crop residues depend on soil properties at the time of application. Adding crop residues with a high C/N ratio to soil can immobilise N in the soil profile and hence reduce N uptake and/or total biomass production. Crop residue incorporation can either stimulate or reduce N2O emissions depending on the mineral N content of the soil. Crop residues pyrolysed to biochar can potentially stabilise native soil C (negative priming) and reduce N2O emissions from cropping soils thus providing climate change mitigation potential beyond the biochar C storage in soils. Incorporation of crop residues as an approach to recycle organic materials and reduce synthetic N fertiliser use in agricultural production requires a thorough evaluation, both in terms of biomass production and greenhouse gas emissions.

Lantana camara L. (Verbenaceae) invasion effects on soil physicochemical properties

Lantana camara is a recognized weed of worldwide significance due to its extensive distribution and its impacts on primary industries and nature conservation. However, quantitative data on the impact of the weed on soil ecosystem properties are scanty, especially in SE Australia, despite the pervasive presence of the weed along its coastal and inland regions. Consequently, mineral soils for physicochemical analyses were collected beneath and away from L. camara infestations in four sites west of Brisbane, SE Australia. These sites (hoop pine plantation, cattle farm, and two eucalyptus forests with occasional grazing and a fire regime, respectively) vary in landscape and land-use types. Significant site effect was more frequently observed than effect due to invasion status. Nonetheless, after controlling for site differences, ~50% of the 23 soil traits examined differed significantly between infested and non-infested soils. Moisture, pH, Ca, total and organic C, and total N (but not exchangeable N in form of NO3-) were significantly elevated, while sodium, chloride, copper, iron, sulfur, and manganese, many of which can be toxic to plant growth if present in excess levels, were present at lower levels in soils supporting L. camara compared to soils lacking the weed. These results indicate that L. camara can improve soil fertility and influence nutrient cycling, making the substratum ideal for its own growth and might explain the ability of the weed to outcompete other species, especially native ones.

Environment characterization as an aid to wheat improvement: interpreting genotype-environment interactions by modelling water-deficit patterns in North-Eastern Australia.

Genotype-environment interactions (GEI) limit genetic gain for complex traits such as tolerance to drought. Characterization of the crop environment is an important step in understanding GEI. A modelling approach is proposed here to characterize broadly (large geographic area, long-term period) and locally (field experiment) drought-related environmental stresses, which enables breeders to analyse their experimental trials with regard to the broad population of environments that they target. Water-deficit patterns experienced by wheat crops were determined for drought-prone north-eastern Australia, using the APSIM crop model to account for the interactions of crops with their environment (e.g. feedback of plant growth on water depletion). Simulations based on more than 100 years of historical climate data were conducted for representative locations, soils, and management systems, for a check cultivar, Hartog. The three main environment types identified differed in their patterns of simulated water stress around flowering and during grain-filling. Over the entire region, the terminal drought-stress pattern was most common (50% of production environments) followed by a flowering stress (24%), although the frequencies of occurrence of the three types varied greatly across regions, years, and management. This environment classification was applied to 16 trials relevant to late stages testing of a breeding programme. The incorporation of the independently-determined environment types in a statistical analysis assisted interpretation of the GEI for yield among the 18 representative genotypes by reducing the relative effect of GEI compared with genotypic variance, and helped to identify opportunities to improve breeding and germplasm-testing strategies for this region.

Modulation of arginine decarboxylase activity in cucumber (Cucumis sativus) cotyledons in short-term organ culture

Among the various amines administered to excisedCucumis sativus cotyledons in short-term organ culture, agmatine (AGM) inhibited arginine decarboxylase (ADC) activity to around 50%, and putrescine was the most potent entity in this regard. Homoarginine (HARG) dramatically stimulated (3- to 4-fold) the enzyme activity. Both AGM inhibition and HARG stimulation of ADC were transient, the maximum response being elicited at 12 h of culture. Mixing experiments ruled out involvement of a macromolecular effector in the observed modulation of ADC. HARG-stimulated ADC activity was completely abolished by cycloheximide, whereas AGM-mediated inhibition was unaffected. Half-life of the enzyme did not alter on treatment with either HARG or AGM. The observed alterations in ADC activity are accompanied by change in Km of the enzyme. HARG-stimulated ADC activity is additive to that induced by benzyladenine (BA) whereas in presence of KCl, HARG failed to enhance ADC activity, thus demonstrating the overriding influence of K+ on amine metabolism.

Development of sustainable management for Pimelea and pimelea poisoning

Toxic Pimelea species (desert riceflower) are naturally occurring species found throughout beef cattle regions of Queensland, New South Wales, South Australia and the Northern Territory. Three species of Pimelea (simplex, elongata, and trichostachya) are poisonous to livestock and potentially fatal to cattle, with serious economic consequences through the loss of production, stock deaths and the costs of agistment. A better understanding of the ecology of the plant/disease is required to develop best practice to manage Pimelea in cattle-producing areas. Development of a chemical assay for the toxin (simplexin) is a key component of the current research project enabling toxin levels to be related to stage of plant growth, environmental and climatic factors.

Evaluating the effects of using a Silverwater technologies disinfestation system

This project aims to determine plant growth response to water treated with the Silverwater disinfestation system against plants that have been irrigated with non treated water. The trial will monitor and measure plant growth rates and health to identify any variations that can be attributed to the Silverwater Technologies disinfestation unit.

Controlling Pythium and associated pests in ginger

This project is to identify treatments that ginger growers can use to control two serious soil-borne pathogens that have emerged and threaten the viability of the ginger industry. Pythium myriotylum, responsible for a severe rhizome rot, is the more serious of the two. It was first identified by ginger growers in the 2007/08 growing season, with some producers reporting total crop losses in some blocks. Symphylids are wingless soil-inhabiting arthropods that feed on the ginger plant's root tips and impair the plants´ ability to absorb nutrients, seriously restricting plant growth and development. Damage caused by symphylids to ginger roots is also expected to facilitate entry of Pythium into the plant.

Biological control of Mikania micrantha in Papua New Guinea and Fiji using the rust fungus Puccinia spegazzinii

Mikania micrantha Kunth (Asteraceae), commonly known as ‘mile-a-minute’, is a neotropical plant species now found in 17 Pacific island countries and territories, invading small cropping areas and plantations, thereby reducing productivity and food security. In 2006, a biocontrol project on M. micrantha commenced in Fiji and Papua New Guinea (PNG). The distribution of M. micrantha as well as baseline data such as plant growth rates and socio-economic impacts were determined before the importation of any biocontrol agents. Mikania micrantha was recorded in all 15 lowland provinces in PNG and on all major islands in Fiji. Plants grow about 3.2cm/day in PNG and about 1.9cm/day in Fiji. A socio-economic survey, involving over 370 respondents in over 220 villages from 15 provinces in PNG, found that 78% of respondents considered M. micrantha a serious weed and about 44% had M. micrantha, which they needed to weed at least fortnightly, in over a third of their land. Over 80% of respondents used slashing and/or handpulling as the preferred method of weed control. About 40% of respondents considered that M. micrantha reduced crop yield by more than 30%. In Fiji, 52 respondents from four islands participated in the survey. Over 60% of respondents in Fiji considered M. micrantha a serious weed and 23% had about 30% of their farm lands infested with the weed. Only 15% of respondents needed to weed at least fortnightly, with 56% using slashing and/or hand-pulling as the preferred means of control. Over 65% of respondents estimated that they lost at least 30% of potential crop yield to M. micrantha. Nearly 90% of respondents used M. micrantha as a medicinal plant to treat cuts and wounds. The life history of the rust Puccinia spegazzinii de Toni (Pucciniales: Pucciniaceae), originating from Ecuador, and imported into PNG and Fiji in 2008, was studied. P. spegazzinii is a microcyclic and autoecious rust and has a life cycle of 18-22 days. An efficient culturing and field release method was developed. Since 2008, the rust has been released at over 450 sites in 15 provinces in PNG, establishing at nearly 70 sites in four provinces. From some sites, the rust has spread over 7 km in 12 months. In Fiji, the rust has been released at over 80 sites, on four of the main islands, namely Viti Levu, Vanua Levu, Taveuni and Ovalau, and has established at 20 sites on Viti Levu and Vanua Levu. Plant growth studies and field monitoring in PNG showed that P. spegazzinii can significantly reduce the growth and density of M. micrantha and offers great potential for the control of this weed.

Classifying genotype by environment interactions for targeted germplasm deployment with a focus on Eucalyptus

A novel methodology for describing genotype by environment interactions estimated from multi-environment field trials is described and an empirical example using an extensive trial network of eucalypts is presented. The network of experiments containing 65 eucalypts was established in 38 replicated field trials across the tropics and subtropics of eastern Australia, with a selection of well-tested species used to provide a more detailed examination of productivity differentials across environmental gradients. By focusing on changes in species’ productivity across environmental gradients, the results are applicable for all species established across the range of environments evaluated in the trial network and simultaneously classify species and environments so that results may be applied across the landscape. The methodology developed was able to explain most (93 %) of the variation in the selected species relative changes in productivity across the various environmental variables examined. Responses were primarily regulated by changes in variables related to water availability and secondarily by temperature related variables. Clustering and ordination can identify groups of species with similar physiological responses to environment and may also guide the parameterisation and calibration of process based models of plant growth. Ordination was particularly useful in the identification of species with distinct environmental response patterns that would be useful as probes for extracting more information from future trials.

Molecular Biomonitoring during Rhizoremediation of Oil-Contaminated Soil

Rhizoremediation is the use of microbial populations present in the rhizosphere of plants for environmental cleanup. The idea of this work was that bacteria living in the rhizosphere of a nitrogen-fixing leguminous plant, goat's rue (Galega orientalis), could take part in the degradation of harmful monoaromatic hydrocarbons, such as benzene, toluene and xylene (BTEX), from oil-contaminated soils. In addition to chemical (e.g. pollutant concentration) and physical (e.g. soil structure) information, the knowledge of biological aspects (e.g. bacteria and their catabolic genes) is essential when developing the rhizoremediation into controlled and effective bioremediation practice. Therefore, the need for reliable biomonitoring methods is obvious. The main aims of this thesis were to evaluate the symbiotic G. orientalis - Rhizobium galegae system for rhizoremediation of oil-contaminated soils, to develop molecular methods for biomonitoring, and to apply these methods for studying the microbiology of rhizoremediation. In vitro, Galega plants and rhizobia remained viable in m-toluate concentrations up to 3000 mg/l. Plant growth and nodulation were inhibited in 500 mg/l m-toluate, but were restored when plants were transferred to clean medium. In the greenhouse, Galega showed good growth, nodulation and nitrogen fixation, and developed a strong rhizosphere in soils contaminated with oil or spiked with 2000 mg/l m-toluate. The high aromatic tolerance of R. galegae and the viability of Galega plants in oil-polluted soils proved this legume system to be a promising method for the rhizoremediation of oil-contaminated soils. Molecular biomonitoring methods were designed and/or developed further for bacteria and their degradation genes. A combination of genomic fingerprinting ((GTG)5-PCR), taxonomic ribotyping of 16S rRNA genes and partial 16S rRNA gene sequencing were chosen for molecular grouping of culturable, heterogeneous rhizosphere bacteria. PCR primers specific for the xylE gene were designed for TOL plasmid detection. Amplified enzyme-coding DNA restriction analysis (AEDRA) with AluI was used to profile both TOL plasmids (xylE primers) and, in general, aromatics-degrading plasmids (C230 primers). The sensitivity of the direct monitoring of TOL plasmids in soil was enhanced by nested C23O-xylE-PCR. Rhizosphere bacteria were isolated from the greenhouse and field lysimeter experiments. High genetic diversity was observed among the 50 isolated, m-toluate tolerating rhizosphere bacteria in the form of five major lineages of the Bacteria domain. Gram-positive Rhodococcus, Bacillus and Arthrobacter and gram-negative Pseudomonas were the most abundant genera. The inoculum Pseudomonas putida PaW85/pWW0 was not found in the rhizosphere samples. Even if there were no ecological niches available for the bioaugmentation bacterium itself, its conjugative catabolic plasmid might have had some additional value for other bacterial species and thus, for rhizoremediation. Only 10 to 20% of the isolated, m-toluate tolerating bacterial strains were also able to degrade m-toluate. TOL plasmids were a major group of catabolic plasmids among these bacteria. The ability to degrade m-toluate by using enzymes encoded by a TOL plasmid was detected only in species of the genus Pseudomonas, and the best m-toluate degraders were these Pseudomonas species. Strain-specific differences in degradation abilities were found for P.oryzihabitans and P. migulae: some of these strains harbored a TOL plasmid - a new finding observed in this work, indicating putative horizontal plasmid transfer in the rhizosphere. One P. oryzihabitans strain harbored the pWW0 plasmid that had probably conjugated from the bioaugmentation Pseudomonas. Some P. migulae and P. oryzihabitans strains seemed to harbor both the pWW0- and the pDK1-type TOL plasmid. Alternatively, they might have harbored a TOL plasmid with both the pWW0- and the pDK1-type xylE gene. The breakdown of m-toluate by gram-negative bacteria was not restricted to the TOL pathway. Also some gram-positive Rhodococcus erythropolis and Arthrobacter aurescens strains were able to degrade m-toluate in the absence of a TOL plasmid. Three aspects of the rhizosphere effect of G. orientalis were manifested in oil-contaminated soil in the field: 1) G. orientalis and Pseudomonas bioaugmentation increased the amount of rhizosphere bacteria. G. orientalis especially together with Pseudomonas bioaugmentation increased the numbers of m-toluate utilizing and catechol positive bacteria indicating an increase in degradation potential. 2) Also the bacterial diversity, when measured as the amount of ribotypes, was increased in the Galega rhizosphere with or without Pseudomonas bioaugmentation. However, the diversity of m-toluate utilizing bacteria did not significantly increase. At the community level, by using the 16S rRNA gene PCR-DGGE method, the highest diversity of species was also observed in vegetated soils compared with non-vegetated soils. Diversified communities may best guarantee the overall success in rhizoremediation by offering various genetic machineries for catabolic processes. 3) At the end of the experiment, no TOL plasmid could be detected by direct DNA analysis in soil treated with both G. orientalis and Pseudomonas. The detection limit for TOL plasmids was encountered indicating decreased amount of degradation plasmids and thus, the success of rhizoremediation. The use of G. orientalis for rhizoremediation is unique. In this thesis new information was obtained about the rhizosphere effect of Galega orientalis in BTEX contaminated soils. The molecular biomonitoring methods can be applied for several purposes within environmental biotechnology, such as for evaluating the intrinsic biodegradation potential, monitoring the enhanced bioremediation, and estimating the success of bioremediation. Environmental protection by using nature's own resources and thus, acting according to the principle of sustainable development, would be both economically and environmentally beneficial for society. Keywords: molecular biomonitoring, genetic fingerprinting, soil bacteria, bacterial diversity, TOL plasmid, catabolic genes, horizontal gene transfer, rhizoremediation, rhizosphere effect, Galega orientalis, aerobic biodegradation, petroleum hydrocarbons, BTEX

Field inoculation with arbuscular-mycorrhizal fungi overcomes phosphorus and zinc deficiencies of linseed (Linum usitatissimum) in a vertisol subject to long-fallow disorder

Long-fallow disorder is expressed as exacerbated deficiencies of phosphorus (P) and/or zinc (Zn) in field crops growing after long periods of weed-free fallow. The hypothesis that arbuscular-mycorrhizal fungi (AMF) improve the P and Zn nutrition, and thereby biomass production and seed yield of linseed (Linum usitatissimum) was tested in a field experiment. A factorial combination of treatments consisting of +/- fumigation, +/- AMF inoculation with Glomus spp., +/- P and +/- Zn fertilisers was used on a long-fallowed vertisol. The use of such methods allowed an absolute comparison of plants growing with and without AMF in the field for the first time in a soil disposed to long-fallow disorder. Plant biomass, height, P and Zn concentrations and contents, boll number and final seed yield were (a) least in fumigated soil with negligible AMF colonisation of the roots, (b) low initially in long-fallow soil but increased with time as AMF colonisation of the roots developed, and (c) greatest in soil inoculated with AMF cultures. The results showed for the first time in the field that inflows of both P and Zn into linseed roots were highly dependent on %AMF-colonisation (R-2 = 0.95 for P and 0.85 for Zn, P < 0.001) in a soil disposed to long-fallow disorder. Relative field mycorrhizal dependencies without and with P+Zn fertiliser were 85 % and 86 % for biomass and 68 % and 52 % for seed yield respectively. This research showed in the field that AMF greatly improved the P and Zn nutrition, biomass production and seed yield of linseed growing in a soil disposed to long-fallow disorder. The level of mycorrhizal colonisation of plants suffering from long-fallow disorder can increase during the growing season resulting in improved plant growth and residual AMF inoculum in the soil, and thus it is important for growers to recognise the cause and not terminate a poor crop prematurely in order to sow another. Other positive management options to reduce long fallows and foster AMF include adoption of conservation tillage and opportunity cropping.

Ground cover management alters development of Fusarium wilt symptoms in Ducasse bananas

Many banana producing regions around the world experience climate variability as a result of seasonal rainfall and temperature conditions, which result in sub-optimal conditions for banana production. This can create periods of plant stress which impact on plant growth, development and yields. Furthermore, diseases such as Fusarium wilt caused by Fusarium oxysporum f. sp. cubense, can become more predominant following periods of environmental stress, particularly for many culturally significant cultivars such as Ducasse (synonym Pisang Awak) (Musa ABB). The aim of this experiment was to determine if expression of symptoms of Fusarium wilt of bananas in a susceptible cultivar could be explained by environmental conditions, and if soil management could reduce the impact of the disease and increase production. An experiment was established in an abandoned commercial field of Ducasse bananas with a high incidence of Fusarium wilt. Vegetated ground cover was maintained around the base of banana plants and compared with plants grown in bare soil for changes in growth, production and disease symptoms. Expression of Fusarium wilt was found to be a function of water stress potential and the heat unit requirement for bananas. The inclusion of vegetative ground cover around the base of the banana plants significantly reduced the severity and incidence of Fusarium wilt by 20 % and altered the periods of symptom development. The growth of bananas and development of the bunch followed the accumulated heat units, with a greater number of bunched plants evident during warmer periods of the year. The weight of bunches harvested in a second crop cycle was increased when banana plants were grown in areas with vegetative ground cover, with fewer losses of plants due to Fusarium wilt.