Includes bibliographical references (leaves 47-50). / Comprehensive morphometric measurements and counts, external colouration, reproductive behaviour and diet of steentjies (Spondyliosoma emarginatum) in Langebaan Lagoon were recorded between September 2004 and February 2005. Sagittal otoliths were extracted and age estimates recorded from both whole and sectioned otoliths in order to estmate growth and mortality parameters. A total of 302 steentjies (139-312 mm fork length) was sampled of which 243 were females, 55 were males and four were individuals with hermaphroditic gonads. Macroscopic and microscopic examination of gonads and monthly gonadosomtic indices suggest a spring spawning season and that steentjies are protgynous hermaphrodites with males significantly than females (t = 64.75, df = 301, p < 0.001). During the reproductive season, males develop a blue colouration. All males with this colouration had testes classified as either active or ripe and had significantly larger gonad masses than those without mating colours (U = 3.5, Z = 6.48, p $lt; 0.001). The length-at-50%-maturity for females was 171 mm (FL) and for males was 204 mm (FL). The size at 50% sex change was estimted as 250 mm (FL). Analysis of foregut contents showed steentjies to be opportunitistic ominivores, changing from plant to animal food as they age. The average percent error and coefficient of variation values for whole otoliths were 4.59% and 6.00% respectively, and 10.89% and 14.53% for sectioned otoliths. Although the result of the whole otolith study provided more precise age estimates, there was a strong indication of bias so the sectioned otolith results were used for calculating the growth model. Growth (by sex) was modelled using a model developed for sex changing sparids that allows for accelerated growth after sex change. Steentjies appear to have shorter life spans than most sparids; the oldest individuals sampled attaining less than 7 years of age. Growth rates appear to be relatively high. The smallest, youngest fish were all female, with males only present in the larger size classes. The estimated mortality for steentjies was 0.7y[-1]. Such a high mortality rate implies that further increase in fishing pressure on this species may not be sustainable in Langebaan Lagoon.

Includes bibliographies. / Until recently not much of the physiology of the angulate tortoise had been investigated, and the work done focused on its thermoregulation (Craig 1973; Perrin & Campbell 1981). However, during the last few years research dealing with more of its thermoregulation and water relations was begun (S. Els pers. comm.). This study aims to examine the relationships between body size and the rates of resting metabolism, evaporative water loss and heat transfer. The relationship of the changes in certain physical properties to the corresponding changes in physiological properties, due to increases in body size, will be investigated. Aspects of a theory of biological similarity will be discussed in the light of the results. Other physiological processes, such as respiration and assimilation efficiency, will be looked at without reference to body size.

Includes bibliographical references. / This study investigates the colonisation of Praslin Island by C. vanzoi, three years after the translocation event. An examination of habitat use, lizard abundance, distribution and population genetics was conducted and population comparisons investigated changes in morphometrics or lizard condition since translocation. These investigations were cunducted during the wet season and the dry season. This information will help determine the value of translocation as a tool for the conservation of this species.

Samples of adult anchovy and plankton were collected on Sea Fisheries Research Institute R.S. Africana 1984-1985 anchovy spawner biomass estimation cruises. A total of 40 sets of anchovy samples of 30 fish each and 40 plankton samples were taken roughly at the same time within an area of intensive spawning over the Agulhas Bank, using an Engels 308 midwater trawl and a CalVET net respectively. Fish samples were frozen in a deep freeze at 20° C shortly after capture. Plankton samples were stored in formaldehyde solution. Fish were measured (total length), weighed (total and ovary mass) and de-stomached. Weight of the stomach contents were determined and anchovy eggs in the stomach were counted. Anchovy eggs in CalVET net samples were counted and staged. Frequency distributions of densities of eggs in the plankton on the Agulhas Bank and off the West Coast were plotted to compare egg density in the two areas. Approximately 53% of the eggs caught over Agulhas Bank occurred in only 25% of samples, indicating a patchy distribution. A frequency distribution was plotted of abundance of eggs in the stomachs of fish. Egg patchiness caused a skewed frequency distribution of egg abundance in anchovy stomachs. Feeding time was estimated from an examination of the relationship between weight of stomach contents versus time of day, taking into account time for gut evacuation. Based upon a developmental stage/temperature/age key, mortality rates of eggs in the sea were calculated, and it was estimated that 44% of anchovy eggs were lost daily. Taking into account estimated rates of egg mortality, egg production, gastric evacuation rate, number of eggs eaten and feeding time, cannibalism was estimated to account for about 62%-70% of the egg mortality. The rate of cannibalism was shown to be consistent with a density-dependent functional response.

The South African shark fishery began in World War 2, based on a demand for shark liver oil. The initial landings are estimated at 3750 tons per annum, but have since declined to approximately 820 tons of shark, principally Galeorhinus galeus, with an additional 800 tons of Callorhjncbus capensis. The fishery for the later species has largely only developed since 1980. The fisheries for both of the principal species are dependent for their commercial success on fishing schools of breeding aggregations and in both cases the catch is mainly taken in breeding and nursery areas. G. galeus is a largely ichthyophagous species, which forms schools of fish of similar size. After sexual maturity, which is reached at a minimum TL of 1 278 mm (8,5y) for males, and a minimum TL of 1 371 mm (9,9y) for females, these schools further assort to include only mature fish of one sex. They are highly mobile and may be migratory. Fecundity is low in newly sexually mature fish but gradually increases with increasing size, although a mathematical relationship for this could not be established. The maximum number of pups recorded in this study was 20.

Includes abstract. / Includes bibliographical references (leaves 131-144). / A study was initiated by the Environmental Section of the De Beers Marine Namibia group as part of a pre-feasibility study for a new dredger (a Trailing Suction Hopper Dredger) to be employed for offshore diamond mining along the Namibian south coast. The aims of this study were to assess (a) the potential mining-induced impacts on oxygen and turbidity, and (b) the behavioural responses of the rock lobster Jasus lalandii, to decreased oxygen, elevated hydrogen sulphide and turbidity. For the latter objective, laboratory-based experiments were designed in which juvenile lobsters were exposed to various levels of hypoxia and increased hydrogen sulphide and suspended sediment levels to determine the impacts these stressors may have on the behavioural responses of lobsters.

Neuropeptides are well known to govern numerous biological functions and are found in all phyla studied to date. Probably the best known neuroendocrine system is the hypophyseal-portal system found in vertebrates, and one of the functions of this system is to mediate reproduction. Mammalian reproduction is controlled by a hormonal cascade which begins in distinct brain regions, namely the hypothalamus and the pituitary gland. Gonadotropin-releasing hormone (GnRH) is a neuropeptide typically produced in the hypothalamus. It is the key neuropeptide for initiating this cascade, and without it, reproduction cannot occur. Naked mole rats (Heterocephalus glaber) have a rigid social hierarchy. The "queen" is the most dominant female and is the only female who breeds. All aspects of reproduction are suppressed in other females in the colony: these "subordinates" are in a prepuberty-like state as they do not ovulate or display breeding behaviours. They are, however, not infertile, and are capable of rising to the breeding position. Since GnRH is the "master hormone" of reproduction, this study investigates its role in the socially-induced suppression of reproduction in female H. glaber. Brains of breeding (n = 7) and non-breeding (n = 5) female naked mole rats were compared to determine any differences in brain size, particularly in regions related to GnRH production. Noteworthy morphological and physiological transformations accompany the change from subordinate to dominant social status, including a significant increase in body length (Mann Whitney U test; p = 0.005, U = 0.000), body mass (Mann Whitney U test; p = 0.009, 1.000) and pituitary width and length (Mann Whitney U test; p = 0.028, U = 0.500 and p = 0.018, U = 0.000, respectively). Since little is known about the GnRH system in H. glaber, this study used immunocytochemistry to identify the distribution and abundance of GnRH neurons in the brains of both breeding and non-breeding females. GnRH neurons were found in the median eminence of the hypothalamus and in the anterior pituitary of both queens and subordinates, however in the brain of queen (n = 7) naked mole rats, there is a significantly larger area of immunoreactivity in comparison to the subordinate (n = 5) brain tissue (Mann Whitney U = 4.000, p = 0.030). This suggests that, in , subordinates, GnRH is inhibited at the level of production. The amino acid structure of the form of GnRH found in the brain of the naked mole rat is currently unknown, therefore a pilot study was carried out, using synthetic mammalian GnRH (mGnRH), mouse brain tissue and naked mole rat pituitaries and hypothalami, to examine and modify (where necessary) the methodologies used for neuropeptide extraction, purification and identification. A limited number of naked mole rats were available as source tissue (n = 4), therefore this study also tested whether it is possible to extract and purify an unknown neuropeptide from only a few mammalian samples. Training for reverse-phase liquid chromatography (RP-HPLC) was achieved by practicing the necessary methods with crude extracts prepared from stick insect (Carausius morosus) corpora cardiaca, which also served to compare vertebrate and invertebrate neuroendocrine systems. Synthetic mGnRH was used to demonstrate repeatability of the protocol and to set up suitable conditions for elution of mGnRH: mGnRH elutes at â 12 min when a solvent gradient of 32 % - 47 % B is applied. Synthetic mGnRH was also used to establish the amount of peptide required for accurate identification of GnRH by antigenicity tests (ELISA) and mass spectrometry. However, when extracts of mouse or naked mole rat brain matter were applied to this system, purification of GnRH was unconvincing as either there was insufficient material, or, some endogenous factor was masking the GnRH. Future studies would benefit from using molecular techniques as they require smaller amounts of source tissue. Alternatively, a larger amount of source tissue would be necessary in order to proceed with biochemical studies. While the impaired production of GnRH in subordinate naked mole rats seems to be linked to their prepuberty like state, it is unlikely that GnRH is the sole factor involved. Many other hormones (such as kisspeptin, gonadotropin-inhibiting hormone and neuropeptide Y) may influence GnRH and sexual maturity, and future studies would benefit from a multi-layered approach to investigate suppression of reproduction in naked mole rats.

Bibliography: leaves 122-142. / The ecological effects of collecting the sandprawn Callianassa kraussi and the mudprawn Upogebia africana for bait were assessed at Langebaan Lagoon, South Africa, through surveys, observations and long-term experimental analyses. While only a small proportion of the prawns is removed from the lagoon annually, amounting to no more than 0.01 for the total lagoon, the physical disturbance inflicted by collecting is likely to be more detrimental and longer-lived than the actual removal of the prawns. Bait-collecting activities are concentrated on the centre sandbanks where some 400 000 prawns are removed annually - less than 5 of the total prawn population on these sandbanks. In terms of numbers removed, bait-collecting is thus no threat to the prawn populations at Langebaan Lagoon although there are signs that sandprawn densities and modal body sizes have been reduced in areas of intense bait-collecting activities. Massive quantities of sediment are disturbed through bait-collecting activities and this, inadvertently, results in the disturbance of other components of the biota. It is estimated that approximately 543 g of macrofauna are disturbed with each bag limit of prawns (50) removed, amounting to approximately 10 000 kg of macrofauna being disturbed annually. While the fate of this macrofauna is not completely determined, approximately 80 is subsequently preyed upon by scavenging gulls. The long-term effects of high-intensity disturbance on the biota of a sandflat were assessed through simulated bait-collecting activities, viz "digging" and "sucking" for C. kraussi and U. africana. The recovery of both C. kraussi and U. africana was far more protracted than predicted - probably being completed only 18 months after the initial disturbance. Meiofaunal and macrofaunal numbers decreased immediately following the initial disturbance.

Bibliography: leaves 90-95. / The strategy for analysing multivariate data presented by Field et. al. (1982) was tested for its sensitivity in detecting the effects of offshore marine mining on macrobenthic communities. The technique has proven to be particularly sensitive and robust in elucidating changes in the structure of marine communities following organic pollution events. The primary aim of this study was to investigate its applicability in discerning community changes in an area exposed to physical disturbance of the seabed. Statistical testing, using analysis of similarities, reveals a highly significant difference between mined and unmined samples. Statistical testing also detects natural spatial heterogeneity across the 6 study areas. Aggregation of the data to higher taxonomic levels did not result in the loss of information, and in fact, improved the resolution of the community patterns. Multivariate analyses were therefore performed using the community data aggregated to genus-level. Hierarchical agglomerative clustering reveals two major groups of samples, the mined and the unmined samples.

No description available.