A research program was developed to investigate basic and applied aspects of toxicity, both lethal and sublethal, of hemoxidants, particularly nitrite, on fish, non-fish aquatic vertebrates, and crayfish. The major objectives of this research were to determine A) acute and sublethal toxicity of nitrite to selected aquatic organisms: 1. aquatic salamander larvae, Ambystoma texanum, 2. swamp crayfish, Procambarus simulans, 3. bluegill, Lepomis macrochirus, 4. bullfrog, tadpoles, Rana catesbiana, 5. channel catfish, Ictalurus punctatus, B) the influence of environmental chloride on acute and sublethal exposures to hemoxidants: 1. on acute nitrite toxicity to salamander larvae, crayfish, and bluegill, 2. on nitrite-induced methemoglobinemia in bullfrog tadpoles, Rana catesbian, C) the effect of environmental hydrogen ion concentrations (pH) on acute nitrite toxicity 1. to the crayfish, Procambarus simulans, 2. to the bluegill, Lepomis macrochirus, D) the effect of temperature in sublethal exposures to nitrite 1. methemoglobin formation in channel catfish exposed at different acclimation temperatures, 2. recovery from methemoglobinemia at different acclimation temperatures, E) the effect of the fish anesthetic TMS-222 on nitrite-induced methemoglobinemia in channel catfish 1. supression of nitrite-induced methemoglobinemia, 2. dose-response curve for TMS-222 induced methemoglobinemia, and F) if a methemoglobin reductase system is present in channel catfish.

The decrease of biological diversity from low latitudes near the equator towards high latitudes is one of the most fundamental patterns noted in ecology. These trends have been the subject of numerous terrestrial and marine studies and have caused the proposal of several explanatory hypotheses. No single hypothesis has adequately accounted for these observed trends. Furthermore, large-scale patterns are frequently modified by processes acting at meso- to small-scales. It is imperative to understand the interaction of these processes to elucidate the mechanisms controlling the structure of intertidal assemblages. The main aim of this thesis was to test the influence of multiple processes at a range of spatial scales on biogenic engineers. Biogenic engineers, such as macroalgae, have been reported to be strongly influenced by processes such as grazing, biogeography and upwelling and subsequent effects are expected to be linked to their associated assemblages. I used infaunal assemblages associated with macroalgae as a model system to understand the interactive effects of meso-scale upwelling in conjunction with large-scale factors (regional and biogeographic). Three species of macroalgae, Hypnea spicifera, Champia lumbricalis and Bifurcaria brassicaeformis, inhabiting different but overlapping home ranges were used in this study. Smaller scale, physical attributes and the associated fauna of these algal species were compared. The effects of meso- to large-scale physical gradients on marine organisms have been investigated in many instances and resultant gradients in physical variables observed. This study was undertaken to investigate morphological trends exhibited by Hypnea spicifera and Champia lumbricalis to compare possible similarities between coastlines and to determine if inferred meso- to large-scale processes influence physical trends in a similar manner despite the coastlines being affected by different hydrodynamic processes. The macroalgae were located in two distinctly different upwelling regimes, Hypnea spicifera on the east and south coasts and Champia lumbricalis on the west coast. Upwelling (delivery of nutrients and primary production) was not found to be an important factor in determining the physical characteristics of macroalgae but there were clear patterns linked to latitude. Mean percentage cover of Hypnea spicifera was positively correlated with distance along the east and south coasts. Mean mass and mean surface area of H. spicifera were, however, negatively correlated with distance from Port St Johns. Mean frond length showed a negative trend but this relationship was not statistically significant. On the west coast, there was a significant increase in mean frond length from north to south for Champia lumbricalis. Mean surface area and mean mass of C. lumbricalis exhibited non-significant negative trends from north to south while there was a non-significant positive trend for mean percentage cover. Following the “Productivity Hypothesis”, these patterns from north to south along the coastlines of South Africa could have important implications for biodiversity associated with these algae. Many previous studies have focussed on the effects of upwelling on species (i.e. effects of nutrients and temperature) but have failed to separate this effect from large scale effects such as biogeography and latitudinal gradients. I tested the influence meso-scale upwelling, large-scale biogeographic processes and latitudinal gradients with two different species of macroalgae one on the east and south coasts of South Africa and the other on the west coast. Hypnea spicifera inhabits the east and south coasts of South Africa spanning two biogeographic provinces and is generally affected by relatively weak upwelling, whereas Champia lumbricalis inhabits the west coast spanning one biogeographic province which is subjected to intense persistent upwelling year round. Within the east and west coast biogeographic provinces there are, however, regions with both upwelling and non-upwelling. On the east and south coasts, entire assemblages differed significantly among the three Regions (St Lucia, Port Alfred and Knysna, while assemblages between upwelling/non-upwelling areas were not different. Assemblages on the different shores differed significantly from each other. There were no significant effects of region or upwelling for the number of individuals of Crustacea or Polychaeta, while Mollusca showed a significant effect of region. The number of individuals of other taxa showed a significant interaction of region and upwelling. Region had a significant effect on number of species of molluscs and other taxa, while there was no effect for either region or upwelling for the Crustacea or Polychaeta. In general different factors were shown to be important (region and shore) while upwelling was rarely important. On the west coast assemblages associated with C. lumbricalis were not influenced by region or upwelling but there were significant differences between shores. Region, upwelling ad shore did not affect number of individuals. There was a significant interaction of region and upwelling for the number of species of crustaceans, while numbers of species of Mollusca, Polychaeta and other taxa showed no effects. In general, regional factors strongly influenced most organisms, while upwelling played a minor role. Hypotheses about differences between upwelling and non-upwelling regions in terms of species abundances and composition due to the input of cold nutrient rich water were not supported. A link between the effects of larger-scale biogeographic factors and their influence on habitat forming taxa and the resultant effects on associated infauna would be a likely explanation for the patterns observed in this study. Diversity of assemblages is known to be influenced by the structural complexity of a habitat, increasing complexity increases the amount of available niches therefore potentially increasing the number of species found within that habitat. Three species of macroalgae, Hypnea spicifera, Champia lumbricalis and Bifurcaria brassicaeformis were selected on the basis of being structurally similar and having their distributional ranges overlap in order to elucidate the effects of structure and macroalgal species on associated assemblages. Hypnea spicifera is different in terms of surface area and biomass from Champia lumbricalis and Bifurcaria brassicaeformis. Assemblages of species found on the three rhodophytes showed both shore and the species of alga had significant effects on composition. Only numbers of individuals of molluscs were affected by the species of alga. Crustacea and Polychaeta showed an interaction of the species of alga with shore. In the case of this study, it is likely that the three species of macroalgae mitigate biological stressors such as predation and physical stressors such as wave exposure and desiccation. In general, meso-scale upwelling is suggested to be marginalised when considering the structuring of assemblages associated with macroalgae, while large scale biogeography has more of an influence. Within shores, assemblages were also strongly influenced by smaller-scale factors such as differences in the structure and species of alga. This study indicates the importance of investigating patterns across a range of spatial scales to gain a comprehensive understanding of factors influencing intertidal organisms.

M.Sc. / This study aims to assess the potential impacts of coal mine activity on exposed fish assemblages in the Mpumalanga Highveld area. A control-impact-assessment method was applied, with sites located upstream and downstream from mining activity. All sites were located in the upper Olifants catchment. Field assessments were done over a high flow and low flow period. It was ascertained that water quality variables contributing towards differences between upstream and downstream environmental conditions are salt concentrations, pH and oxygen saturation. Alterations on fish community structures indicated a categorical decline in abundances and species richness at sites situated downstream from mining activity, while Fish Response Assessment Index (FRAI) scores similarly reflected more impaired fish assemblages at downstream sites. Redundancy analysis (RDA), using fish assemblages and FRAI driver metrics explained 36.5 % of variance obtained and reflected particular alteration in Barbus neefi, B. paludinosus, Tilapia sparrmanii and Pseudocrenilabrus philander populations. Sites most severely impacted were identified, as well as the main drivers responsible for changes in fish assemblages. The Cover metric contributed significantly (p<0.05 to variance measured. Subsequent community threshold management targets are presented for problem sites, which aim to manage each site at an ecological category one class higher than the Present Ecological State (PES).

M.Sc. / South Africa can be classified as a semi-arid country and the management of its precious water resources is essential. Environmental pollution, caused by the development of industry, technology and informal settlements, threaten the health status of many freshwater ecosystems. The health of all living organisms living in aquatic ecosystems is therefore subsequently affected by this decrease in water quality. The health of an ecosystem is thus often reflected by the health of its fauna. Fish are relatively sensitive to changes in their surrounding environment, including an increase in pollution. Fish health may as a result reflect, and give a good indication of the health status of the aquatic ecosystem in which the fish occurs. The initial toxic effects of the pollution may, however, only be evident on cellular or tissue level before significant changes can be identified in fish behaviour or external appearance. Histological analysis appears to be a very sensitive parameter and is crucial in determining cellular changes that may occur in target organs, including the liver. A histological investigation may therefore prove to be a cost-effective tool to determine the health of fish populations, hence reflecting the health of an entire aquatic ecosystem within a bio-monitoring process. Within the South African context, limited research has been conducted regarding the histology and histopathology of endemic fish species, and its value and effectiveness in aquatic monitoring. The aim of this study was to investigate and validate whether fish liver histology can be used as a monitoring tool, to indicate the health status of an aquatic ecosystem. In order to accomplish this, a preliminary histological study was done on the liver of Clarias gariepinus, a freshwater fish species endemic to the southern African region. Fish were collected in the Rietvlei Dam, Marais Dam and the Palala River, chosen as the three aquatic ecosystems for this study, due to their dissimilar pollution levels. Fish liver samples were fixed in 10% neutrally buffered formalin and prepared for light microscopy analysis using standard techniques for Haematoxylin and Eosin (H & E) staining. From the results obtained during this study, it was concluded that the liver histology of C. gariepinus did appear to be a sensitive monitoring tool of aquatic health. The histological investigation of the fish livers obtained in the Palala River, chosen as an unpolluted site, appeared to be in an overall healthier condition than the liver histology of fish obtained from the two relatively polluted sites, the Rietvlei Dam and the Marais Dam. This conclusion was obtained by means of an objective, quantitative analysis. The histological results in this study were quantified in terms of a histological index. An index value representing the specific histological characteristics of the liver was assigned to each individual specimen indicating either a healthy histological structure (index value of 0-2) or a possible pathological condition (index value of 3-6). According to the quantitative assessment, an average histological index value of 2.5 (indicating a relatively healthy histological condition) was calculated for the unpolluted site while a higher average index value of 3.2 (indicating a pathological condition) were calculated for both the relatively polluted sites. The fish liver histology did therefore reflect the health status of the aquatic ecosystem in which the fish specimen occurred. It was therefore concluded that the liver histology of C. gariepinus seemed to be a sensitive biomarker of environmental pollution, and hence an effective monitoring tool for estimating the health status of an aquatic ecosystem. / Dr. G.M. Pieterse

M.Sc. (Zoology) / Please refer to full text to view abstract

MSc. (Zoology) / The Albasini Dam was used as a reference site outside the DDT (1, 1, 1-trichloro-2, 2-bis (p-chlorophenyl) ethane) - sprayed area in a previous survey conducted from 2006 to 2008. DDT, endocrine disrupting chemicals and inorganic chemicals were detected in the dam. A histological analysis of fish from the dam showed histological alterations in heart, liver, gills, and gonads. Therefore, it was necessary to follow up the health status of the dam by monitoring the levels of organic and inorganic chemicals and their effects on fish. The aim of the present study was to determine the health status of three freshwater fish species, Clarias gariepinus, Oreochromis mossambicus, and Coptodon (Tilapia) rendalli from the Albasini Dam, using fish histology as a biomonitoring tool. In total, 18 fish were sampled using gill nets; these included Clarias gariepinus (n=5); Coptodon (Tilapia) rendalli (n=4) and Oreochromis mossambicus (n=9). The histology-based fish health assessment included a standard fish necropsy; a calculation of blood parameters (haematocrit, leukocrit and total plasma protein); somatic indices and the condition factor and a qualitative and semi-quantitative histological assessment of five target organs: liver, heart, gills, kidney and gonads. Water, sediment and fish muscles samples were collected and analysed for inorganic chemicals and organic chemicals. Inorganic chemicals were analysed by means of Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) and Inductively Coupled Plasma-Optic Emission Spectrometry (ICP-OES). Organic chemicals in water and fish muscles were analysed using Gas Chromatography-Mass Spectrometry (GC-MS). Nutrients and physical parameters of water were also measured...

PhD. (Chemistry) / This thesis focuses on functionalising the silicone membrane by filling it with synthesised γ-Fe2O3 nanoparticles for the purpose of using it as a passive sampling device for monitoring microcystins and cylindrospermopsin in aquatic environments. The identity and purity of the synthesised nanoparticles was verified by X-ray diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscope (SEM), surface zeta potential analyser and Brunauer-Emmet-Teller (BET). There was no chemical uptake of microcystin compounds into the silicone membrane without a suspension of maghemite nanoparticles. But, a linear uptake of microcystin compounds was observed into the functionalised silicone membrane passive sampler throughout the exposure period. These findings suggest that the silicone membrane filled with a suspension of synthesised γ-Fe2O3 nanoparticles can be an efficient, green and less costly instrument useful in estimating biotoxin concentrations in South African waters and the world at large.

Three artificial stream designs, termed Large Artificial Stream Units (LASUs), Raceways, and Channels, at two major scales (1700 L, 12.5 L and 20 L recirculated volume) were developed at the Institute for Water Research, Rhodes University, Grahamstown, in order to explore the possibilities of using indigenous rheophilic macroinvertebrates in routine toxicity tests. This study compared these systems, using 96h-EC50 values from sodium sulphate toxicity tests as the experimental response. Two local Ephemeroptera (Leptophlebiidae: Adenophlebia auriculata Eaton, and Baetidae: Afroptilum sudafricanum Lestage) were evaluated for their suitability in routine toxicity tests; and the possible effects of elevated salinity levels in South African rivers on the test species were assessed. Two sets of experiments with each mayfly species were conducted, following an unreplicated regression design. Dechlorinated tap water was used as the water source. Experiments in the Channels were repeated to determine experimental variability. Results were compared statistically by testing for overlap of 95% confidence limits (95%Cls) of EC50 values. The differences between A. auriculata EC50 values in the different systems were statistically significant (no overlap of 95%CLs), but they were not more variable than has been considered normal for biological systems (Coefficient of variation 20.1 %; ratio of greatest EC50 / smallest EC50 1.63). The differences were not related to the scale or the average current velocity characteristic of each stream design (average current velocity LASUs - Raceways - Channels 0.090 - 0.083 - 0.038 m/s). The Channels proved to be most efficient with regard to practical performance as they are portable and easily transportable, user-friendly, reliable, splash-free, cost effective to construct, and can easily be adapted to specific requirements. These systems are therefore recommended for regular use. The suitability of the two mayfly species for routine toxicity testing was evaluated. A. auriculata EC50 values showed a significant negative correlation with the corresponding average body-size (range 1476 - 1610 μm, mean 1555 μm). The different average body-sizes probably reflected the abundance of a certain size range present in the Palmiet River at the time of collection. Both species reacted similarly to Na₂S0₄ (similar slopes of the toxicity curves), identifying this salt as a slow acting toxicant. A. sudafricanum populations were more sensitive to Na₂S0₄ (EC50 3.404 g/L) than A. auriculata (EC50 8.090 g/L), probably because of its smaller body-size (mean 709 μm) and a lack of extremely tolerant individuals. In comparison to other freshwater macro invertebrates, including the standard toxicity test organism Daphnia spp., both mayfly species seemed to be moderately tolerant of Na₂S0₄; therefore there was no particular advantage to using these indigenous taxa rather than Daphnia spp. An assessment of the effects of elevated salinity/TDS levels on the test taxa yielded preliminary insights. A NaCI-EC50 for A. sudafricanum could be extrapolated and suggested a higher sensitivity to Na₂S0₄ than to NaCl. When Na₂S0₄ EC50 values of both species were compared to selected TDS levels of South African rivers, 4. auriculata would mostly not be affected, but A. sudafricanum might occasionally suffer from sub-lethal effects, depending on the sulphate proportion of the TDS. The South African guideline for TDS seemed to protect both species sufficiently.

Cu is one of the most toxic elements that affect fish populations when the fish are exposed to concentrations exceeding their tolerance. To investigate the effects of elementary Cu on aspects of bioconcentration, histology and behaviour, O. mossambicus were exposed to 0 and 0.75 ± 0.20 mg/l of Cu for 96 hours (short-term study), and 0, 0.11 ± 0.02, 0.29 ± 0.02, and 0.47 ± 0.04 mg/l of Cu for 64 days (longterm study) under controlled conditions in the laboratory. For the long-term study fish were sampled for gills, liver, and kidney Cu accumulation analysis after 1, 32 and 64 days of exposure and after 1, 2, 4, 16, 32, and 64 days for gills, liver and spleen histology analysis. Cu accumulation was concentration-duration dependent with the highest accumulation capacity in the liver. A multifactor linear model was developed for the relationship between exposure dose, exposure duration and Cu accumulation in the organs with the liver model: Log L = 3.35 + 0.85W + 0.31T (r² = 0.892) giving a better fit than the gills: G = −35.09 + 10.58W + 17.58T (r² = 0.632). Where L = Cu accumulation values in the liver, G = Cu accumulation values in the gills (both in μg/g dry mass); W = exposure dose in water (mg/l); and T = exposure time (days). Using this model Cu accumulation in organs can be estimated when exposure concentration and duration is known. This model should be tested under different conditions to determine the potential of the model in monitoring Cu toxicity in the environment. Lesions were observed in the liver, gills and spleen in all Cu treatments at all exposure concentration and exposure durations. However, the incidence and the degree of alteration was related to the concentration of Cu and duration of exposure. The sequential appearance of lesions in the order of, hepatic vacuolar degeneration, fatty degeneration and necrosis indicated a gradual increase in liver damage with larger duration of exposure time and increasing Cu concentration. The initial lesions in the gills were manifested as hypertrophy and hyperplasia of the gill epithelium causing increase in the thickness of the secondary lamellae, mucous cell hypertrophy and proliferation, mucous hypersecretion, proliferation of eosinophilic granule cells and hyperplasia of interlamellar cells. With increase in exposure time, necrosis of the eosinophilic granule cells, lamellar oedema, epithelial desquamation and increase in severity of lamellar hyperplasia were observed. These lesions indicated an initial defence mechanism of the fish against Cu toxicity followed by advanced histological changes that were related to Cu concentration and duration of exposure. Changes in the spleen were haemosiderosis, increase in the white pulp and macrophage centres, reduction in the red pulp, and necrosis suggesting that fish exposed to environmentally relevant levels of Cu may be histopathologically altered leading to anaemia and immunosuppression. Regression analysis was used to quantify the relationship between the total activity of the fish, and duration of exposure. There was a gradual decline in fish activity related to Cu concentration and duration of exposure before introducing food into the tanks. There was a constant activity after introducing food in the tanks at the control and 0.11 ± 0.02 mg/l Cu exposure levels irrespective of exposure time. Analysis of covariance (ANCOVA) was used to test for the difference in slopes between treatments. There was no significant difference (p > 0.05) between slopes of the control and 0.11 ± 0.02 mg/l Cu, and between 0.29 ± 0.02 and 47 ± 0.04 mg/l Cu before and after introducing food in the tanks. The slopes of both the control and 0.11 ± 0.02 mg/l Cu were significantly different from those of 0.29 ± 0.02 and 0.47 ± 0.04 mg/l Cu (p < 0.05). There were significant differences in the mean opercular movements per minute between treatments (p < 0.05). There was hyperventilation at 0.11 ± 0.02 mg/l Cu i.e. 87 ± 18 opercular movements per minute (mean ± standard deviation) and hypoventilation at 0.29 ± 0.02 and 0.47 ± 0.04 mg/l Cu i.e. 37 ± 34 and 13 ± 6 opercular movements per minute compared to the control. Hypo- and hyperventilation were related to the lesser and greater gill damage, respectively. In conclusion Cu accumulation and effects on histology of the liver, gills and were related to the concentration of Cu in the water and duration of exposure showing a gradual increase in incidence and intensity with larger duration of exposure time and increasing Cu concentration. The fish were initially able to homeostatically regulate and detoxify Cu. However, as the exposure continued, the homeostatic mechanism appears to have failed to cope with the increasing metal burden causing advanced histological changes.

Studies were undertaken to determine the effect of Kraft mill effluent (KME) on a representative species of the aquatic environment. By transplanting a population of Pacific oysters (Crassostrea gigas) to the Port Mellon area (the site of a Kraft mill), the effect of varying concentrations (on a distance from the mill basis) of the pulp mill waste was monitored in terms of changes in shell dimension, body mass (meat) weight, visual observation of the oysters' physiological state and oyster mortality. A seasonal hydrographic survey was conducted at three regions within the study area to monitor changes in water quality imposed by the effluent. In the areas of effluent imposition (oyster stations 1, 2 and 3) the oysters decreased or showed little gain in shell dimension. The body mass of those oysters nearest the mill outfall began to decline (on a weight basis) shortly after placement. At Station 2 and 3 the deterioration in body mass due to changes in water quality began a short time after the decline at Station 1. Changes in the physiological state of the oyster expressed as a darkening of the gills and mantle edge and variations in body mass texture, can be correlated to an oyster's distance from the mill for each collection timei A mortality rate was calculated at each station for all collection times. The mortality rates at Station 1 (100% in 12 months), Station 2 (50% in 20 months), and Station 3 (20% in 24 months) were extensive and proportional to effluent levels. Several of the changes in water quality (increased temperature, reduced salinities, low oxygen contents, variable pH, dissolved and particulated organic matter and chemical additions) imposed by the effluent were individually tested as the major cause of oyster deterioration. Firstly, the critical oxygen tension (the P02 where V02 declines below the routine rate) was determined as 40 mmHg. Correlating this to the range of 02 levels at each station during a tidal cycle, the oxygen demand of the effluent was not considered as a major cause of oyster mortality. A range of filtered, neutralized (pH 7.0 at 22°C) and aerated percentage KME/volume (0-50%) test solutions were monitored in terms of their effect on the percentage time of shell closure. In these experiments percentages above 20 greatly increased the time of shell closure. The effect of shell closure on oysters was tested by continual (up to 28 days) periods of air exposure. In these experiments the P02, zC02 and pH of the pallial fluid was monitored (from time 0 to 28 days) to determine if anaerobic metabolism was undertaken and if it was, the time span of anaerobic life in juvenile oysters. Anaerobic metabolism was concluded to maintain life in juvenile oysters for 22 days. These results were consistent with the hypothesis that KME is deleterious to oyster populations. At high concentrations of effluent the duration of shell closure is extensive such that an anaerobic death results. At lower concentrations the effluent imposed changes in water quality are responsible for the gradual decline in oyster well-being. / Science, Faculty of / Zoology, Department of / Graduate