Initiating biological control for Nymphaea mexicana zuccarini (Nymphaeaceae) in South Africa

Nymphaea mexicana Zuccarini (Nymphaeaceae) is an aquatic plant originating from south-eastern USA that is becoming increasingly invasive in South Africa as other invasive aquatic plants are being managed successfully through biological control. Mechanical and chemical control of aquatic weeds is expensive, damaging to the environment, and only effective in the short term, so biological control is more desirable as a management strategy for N. mexicana. The biological control of invasive alien plants requires that agents are host specific so that non-target risks are mitigated. For success to be achieved, it is important to ensure that the genetic structure of invasive populations is clarified so that agents can be collected from populations in the native range that match genetically to populations in the invasive range. This is especially important in cases where the morphology of invasive alien plants does not reflect genetic differences between populations. A previous study of the genetic structure of the invasive populations of N. mexicana in South Africa suggests the presence of hybrid forms of the plant in South Africa, with only one of these populations matching with samples from the native range. However, the study only used samples from two sites in the native range using amplified fragment length polymorphisms (AFLPs), so it was necessary to conduct further genetic analyses using samples from more sites in the native range. Hence, the first aim of this study was to develop a better understanding of the genetic structure of N. mexicana populations in the native and invaded range. Genetic samples were collected from sites in the native range during field surveys for potential biological control agents, and inter-simple sequence repeats (ISSRs) were used to compare the genetic structure of invasive and native populations of N. mexicana in South Africa. The results from these analyses suggest that seven of the 14 investigated invasive populations of N. mexicana in South Africa are genetically similar to populations in the native range, while the remaining seven populations are likely to be hybrid forms of the plant. This knowledge will be useful to target populations for biological control and highlights the need for further genetic analyses to determine the parentage of these hybrids so that biological control efforts are more likely to be successful. The initiation of a biological control programme requires that a series of steps are taken in order to maximise the likelihood that this form of intervention will be successful. The first few steps include: identification of the target weed and its genetic structure; exploration in the native range for potential biological control agents; and prioritisation of these agents based on factors such as climatic and genetic compatibility, feeding damage, abundance, and likely host range. Hence, the second aim of this study was to conduct surveys for potential biological control agents in the native range of N. mexicana, and to prioritise these agents. Field surveys were conducted between August and October in 2018 at 17 sites in Florida, Louisiana, and Texas, USA. Sites were selected based on climatic similarity of native sites compared to invasive sites by use of MaxEnt modelling. Native N. mexicana plants were searched for natural enemies, and these were prioritised based on feeding damage, abundance, incidence, and observations of field host range. Two species were prioritised: Bagous americanus LeConte (Coleoptera: Curculionidae) and Megamelus toddi Beamer (Hemiptera: Delphacidae). These species will be imported into quarantine facilities at Rhodes University for host specificity tests to be conducted. Understanding the factors that contribute to the successful establishment of biological control agents is important to improve the efficiency and reduce the costs incurred during the initiation of biological control programmes. Acquiring knowledge of the factors that predict the efficacy of biological control agents is similarly important, and these factors are discussed in the last chapter of this study. The challenges of the biological control of hybrids are also considered, and recommendations are made for the control of N. mexicana and other plants in South Africa