Water hyacinth (Eichhornia crassipes) is an aquatic macrophyte originating from the Amazon basin. Due to its beautiful appearance it has been introduced into numerous countries across the world as an ornamental pond plant. It was introduced into South Africa in the early 1900s and has since reached pest proportions in many of the country’s fresh water bodies, causing significant economic and ecological losses. It is now considered to be the worst aquatic weed in South Africa. Efforts to control the spread of the weed began in the early 1970s and there have been some successes. Biological control has been used widely as an alternative to mechanical and chemical controls because it is cost-effective, self-sustaining and environmentally friendly. To date, six biological control agents have been introduced onto water hyacinth in South Africa. However, due to factors such as cold winter temperatures and interference from chemical control, the agent populations are occasionally knocked-down and thus the impact of biological control on the weed population is variable. In addition, many South African water systems are highly eutrophic, and in these systems the plant growth may be accelerated to such an extent that the negative impact of the agents’ herbivory is mitigated. One of the agents established on the weed is the galumnid mite Orthogalumna terebrantis, which originates from Uruguay. In South Africa, the mite was initially discovered on two water hyacinth infestations in the Mpumalanga Province in 1989 and it is now established at 17 sites across the country. Many biological control researchers believe that the mite is a good biological control agent but, prior to this thesis, little quantitative data existed to confirm the belief. Thus, this thesis is a post-release evaluation of O. terebrantis in which various aspects of the mite-plant relationship were investigated to determine the efficacy of the mite and thus better understand the role of the mite in the biological control programme of water hyacinth in South Africa. From laboratory experiments, in which mite densities were lower than densities occurring in the field, it was found that water hyacinth growth is largely unaffected by mite herbivory, except possibly at very high mite densities. When grown in high nutrient conditions the growth of the plant is so great that any affect the mite has is nullified. Plant growth is thus more affected by nutrients than by mite herbivory. However, mite feeding was also influenced by water nutrient levels and mite herbivory was greatest on plants grown in high nutrient conditions. The presence of the mite had a positive effect on the performance of the mirid Eccritotarsus catarinensis, such that the interactions of the two agents together had a greater negative impact on the plant’s growth than the individual agents had alone. Furthermore, water hyacinth physiological parameters, such as the plant’s photosynthetic ability, were negatively impacted by the mite, even at the very low mite densities used in the study. Plant growth rate is dependent on photosynthetic ability i.e. the rate of photosynthesis, and thus a decrease in the plant’s photosynthetic ability will eventually be translated into decreased plant growth rates which would ultimately result in the overall reduction of water hyacinth populations. In addition, temperature tolerance studies showed that the mite was tolerant of low temperatures. The mite already occurs at some of the coldest sites in South Africa. Therefore, the mite should be able to establish at all of the water hyacinth infestations in the country, but because it is a poor disperser it is unlikely to establish at new sites without human intervention. It is suggested that the mite be used as an additional biological control agent at sites where it does not yet occur, specifically at cold sites where some of the other, less cold-tolerant, agents have failed to establish. Finally, conditions of where, how many and how often the mite should be distributed to water hyacinth infestation in South Africa are discussed.

Potentially damaging submerged invasive freshwater macrophytes have been identified in South African freshwater systems, but have received less attention than their floating counterparts. To ascertain the changes and effects that these species may have on macrophyte ecology, an understanding of the drivers of macrophyte assemblages is essential. The aims of this thesis were to investigate select abiotic and biotic factors driving introduction, establishment and spread of submerged macrophytes in South Africa. Surveys on the status of submerged plant species in South Africa were conducted to find out the distribution and diversity of the species present, imported to, and traded in South Africa. Numerous submerged indigenous and invasive macrophyte locality records were collected during field surveys, of which many were first time records. Pet stores and aquarist trading activities were identified as potential vectors for the spread of submerged macrophytes through online surveys and personal interviews. These results highlighted the potential these species have for continuing to enter, and spread within South African water bodies. Maximum Entropy (MAXENT) is a general-purpose method used to predict or infer distributions from incomplete information, and was used here to predict areas suitable for the establishment of five of these invasive macrophytes. Many systems throughout South Africa, particularly those in the subtropical coastal regions, were found to be climatically suitable for the establishment of Elodea canadensis Michx., Egeria densa Planch., Hydrilla verticillata (L.f.) Royle (all Hydrocharitaceae), Myriophyllum spicatum L. (Haloragaceae), and Cabomba caroliniana Gray (Cabombaceae). Despite the high probability of invasion, facilitated by vectors and suitable climate, South Africa’s rich indigenous submerged aquatic flora may be preventing the establishment of these submerged invasive species. Studies on the competitive interactions between a common indigenous submerged macrophytes, Lagarosiphon major (Roxb.) (Hydrocharitaceae) and M.spicatum, an invasive native to Eurasia, were conducted to ascertain which conditions influence competitive superiority. High sediment nutrient conditions significantly increased the growth rate and competitive ability of both species, while clay sediments significantly increased the competitive ability of L. major over M. spicatum, but sandy sediments improved the competitive ability of M. spicatum. These results highlighted the dynamic changes in competition between submerged species driven by abiotic factors, but did not take into consideration the effect that herbivory, a biotic factor, could have on competition between the two species. The effect of herbivory by phytophagous insects of submerged plant species has been regarded as negligible. To find out what this effect is, multiple field surveys were undertaken throughout South Africa to find natural enemies of indigenous Lagarosiphon species with the aim of identifying such species, and quantifying their influence on plant growth dynamics. Several new phytophagous species were recorded for the first time. An ephydrid fly, Hydrellia lagarosiphon Deeming (Diptera: Ephydridae) was ascertained to be the most ubiquitous and abundant species associated with L. major in South Africa. The influence of herbivory by this fly on the competitive ability of L. major in the presence of M. spicatum was investigated using an inverse linear model, which showed that herbivory by H. lagarosiphon reduced the competitive ability of L. major by approximately five times in favour of M. spicatum. This study served to highlight the importance of herbivory as a driver of submerged aquatic plant dynamics. Current ecological theory emphasises the importance of investigating beyond plant-herbivore interactions, by including multitrophic interactions in community dynamics. Therefore, the potential of parasitism by a parasitoid wasp, Chaenusa luteostigma sp. n. Achterberg (Hymenoptera: Braconidae: Alysiinae) on H. lagarosiphon to shift the competitive interactions between the two plant species was also examined. The addition of the parasitoid reduced the effect of herbivory by the fly on L. major by half, thereby shifting the competitive balance in favour of L. major over M. spicatum. This study provides valuable insight into a selection of drivers of submerged macrophyte assemblages of South Africa. It highlights the precarious position of South African freshwater systems with regard to the potential invasion by damaging submerged invasive species. It also provides interesting insights into the effect of competition, herbivory and parasitism on the establishment and spread of species within submerged freshwater systems. Understanding the different influences could assist managers and policy makers to make validated decisions ensuring the integrity of South African freshwater systems.

The contamination of heavy metals in the environment is a looming concern worldwide. Egeria densa (Planch) (Submerged aquatic plant) from two ponds: Site A with co-ordinates (32º 48’22.04”S; 26°48’58.79” E) and Site B with co-ordinates (32°48’33.25”S; 26°48’33.25”S) in Alice (Eastern Cape) was evaluated for its ability to absorb heavy metals, phytochemical constituents, antimicrobial activity and ultra-structure using standard analytic procedures. Cadmium (Cd), copper (Cu), iron (Fe), lead (Pb), manganese (Mn), and zinc (Zn) were measured in water, sediments and plant. The concentrations of these metal elements were determined with use of Inductively Coupled Plasma- Optical Emission Spectrometry (ICP-OES). In sediments, the heavy metals (mg/kg) decreased in the order of their average concentration as follows: Fe (40.320) > Zn (1.259) > Pb (0.564) > Mn (0.186) > Cu (0.037) in Pond 1 whereas in Pond 2 Fe (61.527) > Cd (0.999) > Mn (0.648) > Pb (0.586) > Zn (0.156) > Cu (0.045). The highest concentration of Fe was detected in both sites and Cu being the least. The concentrations of the metals in the plants sample (from Pond 1) were found in order of Mn > Pb > Cu > Fe whereas cadmium and zinc were not detected, while the concentration in Pond 2 decreases in order of Zn > Mn > Pb > Cd > Fe > Cu. In the water samples, concentrations of heavy metals (mg/L) decreased in the order of their average concentrations as follows: Pb (35.36) > Fe (3.07) > Mn (0.238) > Cu (0.104), both cadmium and zinc were below the limit of detection in Pond 1, whereas in Pond 2 the concentrations decreased as follows: Pb (13.033) >Fe (1.69) > Cu (0.270) > Mn (0.248) > Cd (0.004) and Zinc was not detected. Phytochemical analyses of the plant extracts revealed the presence of phenols, flavonoids, proanthocyanidin, flavonols, saponins, alkaloid and tannins in all the extracts (water, acetone and n-hexane). Both acetone and water extracts, showed high concentration of proanthocyanidin, while tannin was the lowest in acetone extract. Antimicrobial evaluation using, Gram positive (Staphylococcus aureus, Bacillus pumilus, Bacillus cereus, Streptococcus pyogenes, Enterococcus faecalis) and Gram negative (Klebsiella pneumonia, Escherichia coli, Pseudomonas aeruginosa, Proteus vulgaris and Serratia marcescens) bacteria showed negative results for all the strain, except Streptococcus pyogenes which was inhibited at MIC of 0.1 mg/ml. Scanning electron microscopy (SEM) of ultra-structure of Egeria densa, showed that certain bacteria attached to the leaf, However more work has to be done on E. densa to verify the mechanism by which it accumulates heavy metals. The study shows that E. densa has a potential of accumulating heavy metals especial Manganese in plant.