No description available.

A Dissertation submitted to the Faculty of Science, University of the Witwatersrand, in fulfilment of the requirements for the degree of Master of Science / Water hyacinth is one of the most invasive aquatic plants in the world. As such, there have been numerous attempts to model and predict its growth. Some of these models incorporate the influence of temperature or nutrients as the two most important determinants of water hyacinth growth. Other models include the effect of biological control on the growth of the plant, but only one model integrates environmental factors (temperature) with the effect of biological control. In this study, I attempt to incorporate temperature, and biological control effects on the growth of water hyacinth into a single model. Temperature-dependent water hyacinth and stage-structured Neochetina weevil population models were constructed in STELLA 9.1.4 and compared against an empirical dataset for two water hyacinth infested sites in South Africa for a two-year period (2004-2006). Although these models may not simulate field water hyacinth populations accurately, they suggest that Neochetina weevils can reduce water hyacinth populations, to below the assumed carrying capacity (70 kg/m2). It appears that the effects of Neochetina larvae are vital in reducing water hyacinth populations, and need to be further explored in order to simulate water hyacinth/weevil systems accurately.

No description available.

No description available.

The success of biological weed control programs has been limited by a lack of understanding of the stress physiology of insect damage and pathogen development. This case study with the perennial weed, Cirsium arvense, (L.) Scop. evaluated the stress of five natural enemies. Attack by a seed head predator, Orellia ruficauda (F.) caused about 21.5% predation and may reduce seed dispersal. The stress of stem gall formation (Urophora cardui (L.)) is greatest when the gall occurs on young plants and on the mainshoot and defoliation simulation (Cassida rubiginosa Muller) is most effective at high levels on young plants. In nature, however, the latter two natural enemies are not synchronized with these susceptible stages, thereby reducing their effectiveness. Although Cleonus piger Scop., a root crown inhabitant, can result in plant death, regeneration of damaged vascular tissue can occur. Plants which emerge systemically infected with Puccinia punctiformis (Str.) Rohl. (rust) rarely survive the season. A matrix model simulating the effects on Canada thistle population dynamics by the natural enemies was applied.

Colletotrichum coccodes (Wallr.) Hughes, a foliar pathogen of velvetleaf, is being developed as a bioherbicide. Formulation of living organisms for use as pest control products presents unique problems. This research has achieved the development of an adequate formulation of the pathogen by using kaolin clay or talcum powder (1:2.79 wt/wt) as the fillers to dry conidia. Formulated C. coccodes conidia stored at 4, 30C, or at room temperature in bags permeable to oxygen remained viable and able to infect velvetleaf plants at least six months in storage. Various reported germination stimulants increased germination of formulated conidia, although not significantly, whereas increasing concentrations of cutin resulted in subsequent decreases in germination and appressoria formation of fresh as well as formulated conidia. In controlled environment experiments, 14 day-old velvetleaf seedlings were severely diseased when stearic or oleic acids were added to conidia formulated in kaolin clay or talcum powder, respectively. Combinations of germination stimulants, cutinase and/or pectinase inducers did not significantly increase germination and appressoria formation of C. coccodes conidia. Germination of fresh and formulated conidia increased, although not significantly, with the addition of 1% sucrose.

Resistance responses of Abutilon theophrasti were investigated to determine defense mechanisms of the weed against Colletotrichum coccodes and to verify if some chemical suppression of the resistance mechanism could be exploited to enhance the virulence. Induced resistance in A. theophrasti has been confirmed in treatments with C. coccodes, benzothiadiazole, bentazon, and acifluorfen. Induction of peroxidase and phenylalanine ammonia lyase (PAL) activities in the leaves that did not contact with the inducing agents was observed after the localized stresses to the first leaf or the root of the plant with those agents. alpha-Amino-oxy acetic acid (AOA), 2-deoxy-D-glucose (DDG), mannose, oxalic acid, and analogues of oxalic acid and mannose were tested to enhance C. coccodes virulence. However, the compounds did not enhance C. coccodes virulence or affect A. theophrasti growth. Strong antifungal effects, poor inhibitory effects on plant defense mechanisms, or minor dependence of A. theophrasti on the defense mechanisms that the chemicals affected could be reasons. The efficacy of C. coccodes increased in the presence of 0.25 kg a.i. ha-1 bentazon more than when C. coccodes was applied alone, while the effect of glyphosate was minimal. Peroxidase activity was strongly induced by the treatment of C. coccodes and increased over time. PAL and activation of peroxidase was inhibited in the presence of bentazon, suggesting the synergy effect by bentazon is probably due to the suppression on the two defense-related enzymes. In conclusion, A. theophrasti exploits various biochemical and morphological types of defense mechanisms against C. coccodes infection. However, the activation of the defense responses can be suppressed or by-passed in an integrated weed management system.

Field and growth bench experiments were performed to assess the effect of a selective fungal pathogen of Abutilon theophrasti (velvetleaf) on various aspects of intra- and interspecific competition between this vigorous agricultural weed and soybean (Glycine max). In the absence of the foliar pathogen, Colletotrichum coccodes, A. theophrasti and soybean responded differently to the presence of conspecies or to individuals of the other species. In pure stand, the deleterious effects of intraspecific competition on reproductive output were substantially greater for A. theophrasti than for soybean, especially at lower monoculture densities. In mixtures, however, A. theophrasti reproductive performance was markedly higher than at equivalent monoculture densities, particularly at the lower mixture densities. Soybean reproduction at these lower mixture densities (10 to 20 plants m$ sp{-2}$) was severely curtailed compared with reproductive output at equivalent pure stand densities. A. theophrasti reproductive output was limited more by the presence of conspecies than by the presence of soybean, whereas the opposite trend was observed for soybean. In pure stand, application of C. coccodes had limited impact on either A. theophrasti or soybean yield. However, application of the fungal pathogen in A. theophrasti monocultures caused significant (30-44%) aboveground biomass reductions within five weeks of inoculation, in two of the three years in one field study. Eight weeks following C. coccodes inoculation, A. theophrasti biomass within inoculated monoculture plots did not differ significantly from biomass within uninoculated control plots, although height hierarchies were significantly more developed. In mixtures, C. coccodes applications caused reductions in A. theophrasti growth and reproduction when provided with an adequate dew period. Alternatively, soybean yield losses within inoculated mixture plots were generally lower than for uninoculated control plots, althoug

An anthracnose-inducing pathogen, Colletotrichum dematium, was studied as a bioherbicide for Epilobium angustifolium. A comparative study involving other C. dematium isolates suggests that the isolate from E. angustifolium is a forma specialis and should be designated as Colletotrichum dematium f.sp. epilobii. The most severe damage was achieved on seedlings using a conidial density of $1 times 10 sp9$ conidia m$ sp{-2}$. Virulence decreased with plant maturity. Satisfactory levels of control were limited to long dew duration ($>$18 h) and high temperature treatments. Of various adjuvants tested, significantly higher levels of control were achieved when inoculum was sprayed in a vegetable oil emulsion (25% v/v). An inoculum buffered to acidic pH levels (pH 3.0) similarly increased level of control compared with an unbuffered conidial suspension and the adjustment to more alkaline pH levels using the citrate-phosphate buffer (pH $>$ 4.0), inhibited disease expression. In host range studies, C. dematium f.sp. epilobii was restricted to the Onagraceae family. Both Camissonia bistorta and Clarkia pulchella were susceptible whereas the fungus was highly virulent on all but one E. angustifolium ecotype. Among nine commercial tree species tested for susceptibility yellow birch (Betula alleghaniensis) was susceptible to the fungus when conidia were applied in an oil emulsion. The application of inoculum in a tank mix combination with the oil emulsion and a low rate of glyphosate provided significant growth control of E. angustifolium seedlings (7-wk-old), whereas the effectiveness of this suspension significantly decreased with plant maturity. Under controlled conditions, post-emergence application of an oil-based formulation including the ground colonized substrate of another bioherbicide candidate of E. angustifolium, Alternaria sp., significantly reduced above-ground biomass when provided a 12-h dew and applied at a rate as low as $5 times 10 sp6$ conidia m$ sp{-2}$.

Development of microbial herbicides is constrained by unreliability in the field where conditions are often sub-optimal for infection. Crucially, sufficient moisture, often dew, is required to establish infection. Two model systems, Colletotrichum dematium (Pers. ex Fr. ) Grove on Chenopodium album L. and Mycocentrospora acerina (Hartig) Deighton on Viola arvensis Murr., have been investigated and formulation requirements for each system identified, principally to reduce the dew period necessary for infection. Effects of adjuvants (surfactants, stickers and humectants) on spore germination and appressorium formation were investigated in vitro. Few were toxic and then, principally, at high concentration. The surfactants Tween 40,60 and 80 were compatible with both pathogens. Similarly, the stickers acacia, ghatti, guar, karaya, locust bean and xanthan gums and low viscosity alginic acid were all non-toxic as was the humectant glycerol. Each pathogen reacted differently to the adjuvants and any potential microbial herbicide will need individual matching of adjuvants to give an effective formulation. A working formulation (rapeseed oil-in-water (1: 10 v/v) emulsion using 0.1 % v/v Tween 40 as the emulsifier) was found to reduce the dew period requirement of M. acerina from 36 to 18 hours. The formulation protected spores from desiccation for 24 hours after application, or for 16 hours following a sub-optimum dew period occurring immediately after application. Scanning electron microscopy showed that the applied spores, and the developing mycelium, were immersed in the oil deposit. Transmission electron microscopy of sections through formulation deposits on the leaf revealed that some inversion of the emulsion, to form a water-inoil deposit, had occurred, suggesting a mechanism of protection against desiccation. The oil phase infiltrated the cortical intercellular spaces only when the leaf was infected. This intercellular oil contained more water than that on the leaf surface. Emulsion-formulation applied to run-off with an 'air brush', consistently gave significantly better weed control under sub-optimal dew conditions than a formulation of surfactant only. When applied with a conventional hydraulic nozzle at 400 1 ha" the emulsion was only occasionally superior to the surfactant alone. Such interactions require further in-depth investigation. The importance of correct inoculum placement for maximum effectiveness, independent of formulation type, was highlighted. Unless all meristems are killed, survivors quickly grow, despite the death of neighbouring leaves and petioles, and the weed suffers merely a growth check. Formulation as emulsion improved diseasee stablishmenta nd diseasee xpressioni n the target weedo nly in somec ircumstancesF. urther researchin to spraya pplicationm ethodsa ndt heir interactions with formulation, host and environment is clearly necessary