Weeds are responsible for about 30 percent of all crop losses worldwide, but all weed research, including biological control, receives only about 10% of crop pest control support. The impact of weeds is insidious, diffuse, and complex Increased use of biological control may help to reduce the massive quantity of energy and time now expended for weed control worldwide. Batra (1981) "The vitality of biocontrol of weeds requires two types of evaluation before a project is completed. To maintain financial support it is necessary to show that biocontrol is an economical method of solving certain types of weed problems and to improve the effectiveness of biocontrol in the future, it is necessary to have scientific feed-back. Essentially, the first type of evaluation is concerned with what has been achieved and the second with why the result, either success or failure, has been achieved." Harris (1980a) The biological control of weeds using insects is a young science. The first intentional use of an insect to control a weed was in 1863 in India where the cochineal insect, Dactylopius ceylonicus (Green), was dispersed to control Opuntia vulgaris Miller some 68 years after the accidental introduction of this insect to that country (Goeden, 1978; Moran & Zimmermann, 1984). However, the first full scale attempt at a classical weed biocontrol programme was against Lantana camara L. in Hawaii in 1902 (Goeden, 1978; Harley, 1985a). By 1984 there had been 499 releases of exotic invertebrates and fungi, 488 of which were insects, for the control of 101 weed species in 70 countries (Julien et al., 1984). Despite this marked increase in the number of biocontrol programmes in later years, attempts which failed have rarely been studied or documented and even successes are seldom adequately quantified. This has been implied by Harley (1985a), Julien, (1982), Julien et al. (1984) and Maw (1984). The pressure to produce results and the difficulty of determining the reasons for failures are prohibitive (Dennill et al., 1987; Appendix 6; Goeden & Louda, 1976). The advancement of this science thus relies heavily on the study of its successes. This thesis is an evaluation of the gall wasp Trichilogaster acaciaelongifoliae as a biocontrol agent for the weed Acacia longifolia in South Africa. The wasp has been successful, and, in accordance with Harris (1980a), my aims were twofold: to quantify that success and to provide reasons for it. This information not only produces guidelines ii for the future selection of biocontrol agents, sensu Harris (1980a), but also serves as a means of enhancing our understanding of insect-plant relationships. In addition, this thesis provides an indication of the potential of gall forming insects in biological control of weeds. Gall formers have seldom been used to control weeds (Julien, 1982; Maw, 1984). Since their effects are indirect compared with those of insects attacking vegetative plant parts, their potential for weed biocontrol has apparently been underestimated in the past by some authors (Goeden, 1983; Harris, 1973; Hokkanen, 1985a). The present attempt is the first ever in which a gall forming hymenopteran has been used to control a weed, and shows beyond doubt that certain gall formers can have strong potential in this field. Chapter 1 concerns the establishment of the wasp, its population increases, dispersal, host-seeking, and the reduction of reproductive potential of A. longifolia populations. The development of new techniques that were necessary for the determination of these parameters is included. Emphasis is placed on the reasons for the successful establishment and rapid population increases observed. The potential of the insect for suppressing both reproduction and growth of the weed is great, and is examined in detail in Chapter 2. The phenological information obtained during the course of this study showed important inadequacies in the knowledge of the phenology of A. longifolia. This is rectified in Chapter 3 in which the implications of an incomplete understanding of weed phenology for biological control are emphasised. In particular, the resource allocation between reproduction and growth of the plant provides a crucial setting for the next chapter. In Chapter 4 the nature of the galling by T. acaciaelongifoliae and its relation to the phenology of A. longifolia are examined in order to explain how the wasp is able to exploit its host so effectively under South African conditions. Chapter 5 deals with the release of the wasp throughout the South African range of the weed and the identification of regions in which its performance appears inadequate. The climates of South Africa and Australia are compared to determine to what degree the establishment of T. acaciaelongifoliae in various regions in South Africa is related to the climate of the regions in which the wasp was collected in Australia. This chapter includes maps illustrating weed distribution, co-ordinates for all release points, and documents the method whereby the wasps were released. Chapter 6 is an essay dealing with the implications of the study, both practical and theoretical. Recommendations regarding the future selection of biocontrol agents are made, with emphasis on perspectives emerging from this thesis, and the future role of the wasp in the control of A. longifolia in South Africa is discussed.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:uct/oai:localhost:11427/6250 |
| Date | January 1987 |
| Creators | Dennill, George Bentley |
| Contributors | Moran, V C |
| Publisher | University of Cape Town, Faculty of Science, Department of Biological Sciences |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Doctoral Thesis, Doctoral, PhD |
| Format | application/pdf |
Page generated in 0.0018 seconds