From Introduction: Rainbow trout (Salmo gairdneri) were introduced into Europe from North America in the latter half of the last century. They can tolerate higher water temperatures and lower oxygen concentrations than the native brown trout (Salmo trutta). Rainbows grow faster than brown trout under similar conditions and are thus particularly attractive for artificial rearing methods. In Denmark there is a thriving rainbow trout farming industry producing about 9,000 metric tons annually which is largely exported for table use (Mills 1971). In Britain production of rainbow trout for food is not on such a large scale but they form the basis of a considerable sport fishery. In Scotland and Northern England although rainbow trout reach sexual maturity they do not generally breed so the population is entirely dependent on restocking with hatchery reared fish, thus although living more or less wild in many British waters , this species is essentially an artificially managed resource upon which man can impose genetic selection (Donaldson and Olson 1957) as well as normal fishery controls. A detailed understanding of the biology and physiology of this species is hence of particular importance.

This study reports on investigation of ways of improving the breeding programme for growth-related traits in common carp in Vietnam. The base population was synthesized following a single pair mating scheme from six carp stocks: (1) 2nd generation of family selection; (2) Hungarian 6th generation of mass selection; (3) Hungarian scaled carp; (4) Indonesian yellow 6th generation of mass selection; (5) Indonesian yellow carp; and (6) Vietnamese 6th generation of mass selection. The next two selected generations were produced using a partial factorial mating scheme, with each family being split and reared using communal early rearing (CER) or separate early rearing (SER) methods. The second generation (G2) was produced from selected fish from the CER G1 group. The total number of selection, control and reference families was 135 in the G1 and 101 in the G2 respectively. The control and reference (Hungarian P33 line) families were produced by single pair mating (reference families with the G2 only). Seven microsatellite loci were used for parentage assignment in the CER groups: 96.8% of the offspring (1284 individuals) and 96.2% offspring (1341 individuals) were unambiguously assigned to 113 families (selection, control) in the G1 and 99 families (selection, control and reference) in the G2 generations, respectively. Restricted maximum likelihood in the individual model was used to estimate phenotypic and genetic parameters. In CER, the estimated heritability values of common carp were from 0.20 ± 0.04 to 0.29 ± 0.05 for both weight and length at final harvest, indicating substantial additive genetic variation for selection on growth-related traits. The overall obtained maternal and common environmental effects were consistently close to zero. The average of direct response to selection for body weight was 15.0% per generation. In SER, the number of families in the G1 and G2 were 135 (selection and control) and 101 (selection, control and reference), respectively. The heritability estimates were from 0.20 ± 0.07 to 0.31 ± 0.08 at final measurement. Common environmental (full-sib family) effect were all lower at tagging and slightly higher at last measurement, ranging from 0.05 to 0.22. The response in each generation of selection as the difference between the selection and control lines was 8.1% on average for weight at final harvest, lower than under CER. The high genetic correlations of growth-related traits between the third (one year old, mature) and second (7 months old) measurements could allow selection to be based on the earlier assessment, reducing handling stress close to spawning. The benefits of using microsatellite markers to ascertain parentage, achieve greater growth rate (close to farming systems), shorten time to maturity and selection, and the overall relative merits of using CER v’s SER in this genetic improvement programme are discussed.

Research and development in aquaculture has supported the knowledge-based development of the sector over the last decades. In particular, species diversification is playing an important role to ensure sustainability of the industry and helping to reduce pressure on wild stocks of those aquatic species for which farming technology is still at the early stages. Due to the increasing pressures on more traditional carnivorous marine finfish species (aquafeed reliance on fishmeal and fish oil, environmental impact, market price) low trophic organisms are receiving more attention to provide sustainable alternatives and integrate production activities with the aim of reducing environmental impacts and to provide secondary high value crops. Integrated Multi-Trophic Aquaculture (IMTA) systems are therefore at the forefront of innovation in the industry. Several invertebrate species have been investigated and tested as integral part of IMTA (mussels, oysters, abalone and macroalgae) and echinoderms have also been considered as good candidates for the future development of this technology. In order to allow for a more widespread uptake of integrated aquaculture, several technical and biological challenges need to be overcome, including a reliable supply of juveniles. In recent years, this has prompted investigation on Echiniculture as a whole and on hatchery technologies in particular. This PhD investigated key constraints in edible sea urchin (Paracentrotus lividus) juvenile production with the aim to improve commercial sea urchin hatchery outputs. The research firstly focused on larval nutrition (Chapter 3 and 4) and specifically tested the hypothesis that larvae required higher dietary inputs of long chain fatty acids than those provided by Dunaliella tertiolecta, a microalgae species widely used in echinoderm larval rearing. Fatty acid composition of P. lividus eggs, investigated in Chapter 3, supported this hypothesis, which was further confirmed by the results obtained in Chapter 4 where microalgae (Cricosphaera elongata, Pleurochrisis carterae and Tetraselmis suecica) with a more balanced fatty acid profile, in particular richer in long chain fatty acids, were employed. This resulted in a significantly improved larval development and survival. Results also indicated that these alternative microalgae species could be successfully grown without modification of the microalgae production protocols in the hatchery where the experimentation had taken place. The third experimental chapter compared static and flow through systems which provides more stable water quality through constant water exchange and reduces larval handling and associated stress. Results indicated that larval survival was significantly improved by the flow-through system and the need for tank cleaning was reduced (three versus seven times per larval cycle when using flow-through and static rearing systems respectively). However, water quality, based on the parameters assessed (NH4, PO4-3, NO2 and NO3), did not show any significant differences between systems. Reduced handling could have therefore played the most important role in promoting larval survival. Both these trials resulted in a significant 5 to 20 % increased survival. A follow-up study, combining flow-through with more suitable microalgae, should be carried out and could result in even further enhanced survival. Then, chapters 6 and 7 focused on broodstock nutrition and subsequent improvement of gamete quantity and quality. These two trials aimed to explore and describe the biological effects that some important nutrients, such as proteins, lipids, fatty acids and carotenoids, have on urchins’ somatic and gonadal growth, gonad biochemical composition during gametogenesis, fecundity and maternal provisioning to developing embryos. Results from the experiment described in Chapter 6 indicated that higher protein content can improve somatic growth in P. lividus adults and that more expensive, protein-, lipid- and energy-rich diets do not significantly enhance fecundity or offspring performance. Results, moreover, highlighted the need for a specifically formulated broodstock diet and gave some insights into what its composition should be, especially in relation to carotenoids. In Chapter 7, fatty acid profiles of P. lividus gonads throughout gametogenesis were studied for the first time. It was observed that, among Long Chain Polyunsaturated Fatty Acids (LC-PUFAs), Eicosapentaenoic acid (EPA) and Docosahexaenoic acid (DHA) are primarily accumulated during gametogenesis, whilst Arachidonic acid (ARA) appears to be independent of dietary input. In addition, it was clearly shown that ARA is the only LC-PUFA accumulated in the eggs along with Non Methylene Interrupted Fatty Acids (NMI FAs). As well as looking at the biological effects of different diets on fatty acid profiles of gonadal and larval tissues, the work also expanded on a more fundamental level to explore the metabolic pathway through which precursors could be used by sea urchins for the endogenous production of long chain fatty acids (Chapter 8). Three Expressed Sequence Tags (ESTs) for putative fatty acyl desaturases, one of which was closely related to Octopus vulgaris ∆5-like fatty acyl desaturase, were identified. The newly cloned putative desaturase of P. lividus possessed all typical features of other fatty acyl desaturases. However, because of time constraints, functional characterisation, originally planned, of the new protein could not be performed and further research effort is needed to investigate this important aspect of sea urchin physiology. Overall, the aim of this research project has been achieved as it provided a set of exploitable results and protocols to improve hatchery practices for the production of P. lividus juvenile. However, more research is required to investigate some of the underlying mechanisms behind the observed biological effects such as delay in larval development when T. suecica was used as larval feed, increased broodstock fecundity, improved larval survival in the flow-through system and higher gonadal concentration of some fatty acids (mainly DHA) than provided in the feed.

This study investigates early life stages of the southern sea garfish (Hyporhamphus melanochir) and their association with seagrass in Gulf St Vincent, South Australia. The overall aims were to identify and describe the early life stages of H. melanochir and to explore the possible relationship(s) between these life stages and seagrass habitat with the emphasis on seagrass as a requirement for spawning or as a food source. The reproductive biology of female H. melanochir from the commercial fishery was assessed by microscopic examination of ovaries, oocyte size distributions, gonadosomatic indices, and macroscopic ovarian stages. Five stages of oocyte development were identified and described: perinucleolar, yolk vesicle, yolk globule, migratory nucleus and hydrated. A coherence between histological and whole oocyte descriptions is demonstrated. Hyporhamphus melanochir are characterised as multiple spawners with group-synchronous oocyte development and indeterminate fecundity. A protracted spawning season from October to March was indicated by the occurrence of ripe ovaries and increases in gonadosomatic index. Females reach sexual maturity at 193 mm standard length, and batch fecundity ranged from 201-3044 oocytes depending on fish size. Spawning shoals are segregated by sex, as indicated by commercial samples, with a biased female-to-male ratio of 4.5:1 during the spawning season (1.2:1 during the non-spawning season). In addition, features of the oocyte surface were closely examined, which revealed that the filaments on the chorion of the hydrated oocyte are adhesive. These adhesive filaments presumably allow the fertilised egg to become attached to vegetative substrate by adhesion and/or entanglement. H. melanochir larvae were discriminated from another hemiramphid species, river garfish (H. regularis), which is also known to occur in the study area, based on species-specific amplification of part of the mitochondrial control region using a multiplex polymerase chain reaction (PCR) assay. The species were easily discerned by the number and distinct sizes of PCR products [H. melanochir, 443 bp; river garfish (H. regularis), 462 and 264 bp]. Although based on a single gene, this molecular method will correctly identify the species of individuals in at least 96% and 94% of tests for H. melanochir and H. regularis, respectively. Subsequent to verifying the identification of species by molecular discrimination, the larval development of H. melanochir and H. regularis were described. Larvae of H. melanochir and H. regularis had completed notochord flexion at hatching and are characterized by their elongate body with distinct rows of melanophores along the dorsal, lateral and ventral surfaces; small to moderate head; heavily pigmented, long straight gut; persistent preanal finfold; and extended lower jaw. Fin formation occurs in the sequence: caudal, dorsal and anal (almost simultaneously), pectoral, pelvic. Despite the similarities between both species and among hemiramphid larvae in general, H. melanochir larvae are distinguishable from H. regularis by: having 58-61 vertebrae (v. 51-54 for H. regularis); having 12-15 melanophore pairs in longitudinal rows along the dorsal margin between the head and origin of the dorsal fin (v. 19-22 for H. regularis); and the absence of a large ventral pigment blotch anteriorly on the gut and isthmus (present in H. regularis). A logistic regression analysis of body measurements also revealed interspecies differences in the combined measurements of eye diameter and pre-anal fin length. Both species can be distinguished from morphologically similar larvae found in southern Australia (other hemiramphids and a scomberosocid) by differences in meristic counts and pigmentation. Hyporhamphus melanochir larvae were successfully collected throughout Gulf St Vincent using a neuston net; however, attempts to sample eggs were unsuccessful. Abundances of larvae in the gulf averaged 4.8 and 12.3 larvae 1000⁻ ² of surface water in December 1998 and December 2000, respectively. Larvae exhibit fast growth, as indicated by otolith growth increments, with backcalculated spawning dates falling within the October-March spawning season. Spatial analysis of larval distributions revealed a positive spatial autocorrelation, i.e. non-randomness or clustering of similar abundance values. Most larvae were found in the upper region of the gulf, and the prevalence of seagrass habitat throughout this region supports the view that the demersal eggs of H. melanochir become attached to seagrass and/or algae following spawning. A gyre in waters of the upper gulf, influenced by prevailing southerly winds, the Coriolis effect, and land boundaries, may explain retention of larvae. The importance of seagrass beds to H. melanochir spawning is also supported by anecdotal evidence and available literature on eggs of other Beloniformes, which are also demersal and attach to marine plants. Dual stable isotope analysis (δ¹³ and δ¹⁵N) of larval, juvenile and adult H. melanochir and several potential food sources from the Bay of Shoals was carried out to estimate the importance of zosteracean seagrass towards the assimilated diet of H. melanochir. Although the diet of H. melanochir larvae is probably planktonivorous, their isotopic signatures partly reflect the parental diet due to the influence of pre–existing tissue in addition to growth. According to mixing model calculations, the signatures of juveniles can be explained by a diet consisting of 23–37% Zostera, 0–10% Halophila and the remainder zooplankton, whilst the diet of adults consists of 53–58% Zostera and the remainder zooplankton. These findings indicate an increasing dependence upon Zostera with growth of H. melanochir. The results of this study enhance the completeness of our understanding of the fisheries biology and ecology of H. melanochir. Significant contributions are provided in reproductive biology and larval biology, seagrass beds (in combination with mixed algae) are demonstrated to be an important habitat for spawning, and Zostera seagrass is shown to be a necessary food source in the diet of juveniles and adults. / Thesis (Ph.D.)-- University of Adelaide, School of Earth and Environmental Sciences, 2005.

This study investigates early life stages of the southern sea garfish (Hyporhamphus melanochir) and their association with seagrass in Gulf St Vincent, South Australia. The overall aims were to identify and describe the early life stages of H. melanochir and to explore the possible relationship(s) between these life stages and seagrass habitat with the emphasis on seagrass as a requirement for spawning or as a food source. The reproductive biology of female H. melanochir from the commercial fishery was assessed by microscopic examination of ovaries, oocyte size distributions, gonadosomatic indices, and macroscopic ovarian stages. Five stages of oocyte development were identified and described: perinucleolar, yolk vesicle, yolk globule, migratory nucleus and hydrated. A coherence between histological and whole oocyte descriptions is demonstrated. Hyporhamphus melanochir are characterised as multiple spawners with group-synchronous oocyte development and indeterminate fecundity. A protracted spawning season from October to March was indicated by the occurrence of ripe ovaries and increases in gonadosomatic index. Females reach sexual maturity at 193 mm standard length, and batch fecundity ranged from 201-3044 oocytes depending on fish size. Spawning shoals are segregated by sex, as indicated by commercial samples, with a biased female-to-male ratio of 4.5:1 during the spawning season (1.2:1 during the non-spawning season). In addition, features of the oocyte surface were closely examined, which revealed that the filaments on the chorion of the hydrated oocyte are adhesive. These adhesive filaments presumably allow the fertilised egg to become attached to vegetative substrate by adhesion and/or entanglement. H. melanochir larvae were discriminated from another hemiramphid species, river garfish (H. regularis), which is also known to occur in the study area, based on species-specific amplification of part of the mitochondrial control region using a multiplex polymerase chain reaction (PCR) assay. The species were easily discerned by the number and distinct sizes of PCR products [H. melanochir, 443 bp; river garfish (H. regularis), 462 and 264 bp]. Although based on a single gene, this molecular method will correctly identify the species of individuals in at least 96% and 94% of tests for H. melanochir and H. regularis, respectively. Subsequent to verifying the identification of species by molecular discrimination, the larval development of H. melanochir and H. regularis were described. Larvae of H. melanochir and H. regularis had completed notochord flexion at hatching and are characterized by their elongate body with distinct rows of melanophores along the dorsal, lateral and ventral surfaces; small to moderate head; heavily pigmented, long straight gut; persistent preanal finfold; and extended lower jaw. Fin formation occurs in the sequence: caudal, dorsal and anal (almost simultaneously), pectoral, pelvic. Despite the similarities between both species and among hemiramphid larvae in general, H. melanochir larvae are distinguishable from H. regularis by: having 58-61 vertebrae (v. 51-54 for H. regularis); having 12-15 melanophore pairs in longitudinal rows along the dorsal margin between the head and origin of the dorsal fin (v. 19-22 for H. regularis); and the absence of a large ventral pigment blotch anteriorly on the gut and isthmus (present in H. regularis). A logistic regression analysis of body measurements also revealed interspecies differences in the combined measurements of eye diameter and pre-anal fin length. Both species can be distinguished from morphologically similar larvae found in southern Australia (other hemiramphids and a scomberosocid) by differences in meristic counts and pigmentation. Hyporhamphus melanochir larvae were successfully collected throughout Gulf St Vincent using a neuston net; however, attempts to sample eggs were unsuccessful. Abundances of larvae in the gulf averaged 4.8 and 12.3 larvae 1000⁻ ² of surface water in December 1998 and December 2000, respectively. Larvae exhibit fast growth, as indicated by otolith growth increments, with backcalculated spawning dates falling within the October-March spawning season. Spatial analysis of larval distributions revealed a positive spatial autocorrelation, i.e. non-randomness or clustering of similar abundance values. Most larvae were found in the upper region of the gulf, and the prevalence of seagrass habitat throughout this region supports the view that the demersal eggs of H. melanochir become attached to seagrass and/or algae following spawning. A gyre in waters of the upper gulf, influenced by prevailing southerly winds, the Coriolis effect, and land boundaries, may explain retention of larvae. The importance of seagrass beds to H. melanochir spawning is also supported by anecdotal evidence and available literature on eggs of other Beloniformes, which are also demersal and attach to marine plants. Dual stable isotope analysis (δ¹³ and δ¹⁵N) of larval, juvenile and adult H. melanochir and several potential food sources from the Bay of Shoals was carried out to estimate the importance of zosteracean seagrass towards the assimilated diet of H. melanochir. Although the diet of H. melanochir larvae is probably planktonivorous, their isotopic signatures partly reflect the parental diet due to the influence of pre–existing tissue in addition to growth. According to mixing model calculations, the signatures of juveniles can be explained by a diet consisting of 23–37% Zostera, 0–10% Halophila and the remainder zooplankton, whilst the diet of adults consists of 53–58% Zostera and the remainder zooplankton. These findings indicate an increasing dependence upon Zostera with growth of H. melanochir. The results of this study enhance the completeness of our understanding of the fisheries biology and ecology of H. melanochir. Significant contributions are provided in reproductive biology and larval biology, seagrass beds (in combination with mixed algae) are demonstrated to be an important habitat for spawning, and Zostera seagrass is shown to be a necessary food source in the diet of juveniles and adults. / Thesis (Ph.D.)-- University of Adelaide, School of Earth and Environmental Sciences, 2005.

This study investigates early life stages of the southern sea garfish (Hyporhamphus melanochir) and their association with seagrass in Gulf St Vincent, South Australia. The overall aims were to identify and describe the early life stages of H. melanochir and to explore the possible relationship(s) between these life stages and seagrass habitat with the emphasis on seagrass as a requirement for spawning or as a food source. The reproductive biology of female H. melanochir from the commercial fishery was assessed by microscopic examination of ovaries, oocyte size distributions, gonadosomatic indices, and macroscopic ovarian stages. Five stages of oocyte development were identified and described: perinucleolar, yolk vesicle, yolk globule, migratory nucleus and hydrated. A coherence between histological and whole oocyte descriptions is demonstrated. Hyporhamphus melanochir are characterised as multiple spawners with group-synchronous oocyte development and indeterminate fecundity. A protracted spawning season from October to March was indicated by the occurrence of ripe ovaries and increases in gonadosomatic index. Females reach sexual maturity at 193 mm standard length, and batch fecundity ranged from 201-3044 oocytes depending on fish size. Spawning shoals are segregated by sex, as indicated by commercial samples, with a biased female-to-male ratio of 4.5:1 during the spawning season (1.2:1 during the non-spawning season). In addition, features of the oocyte surface were closely examined, which revealed that the filaments on the chorion of the hydrated oocyte are adhesive. These adhesive filaments presumably allow the fertilised egg to become attached to vegetative substrate by adhesion and/or entanglement. H. melanochir larvae were discriminated from another hemiramphid species, river garfish (H. regularis), which is also known to occur in the study area, based on species-specific amplification of part of the mitochondrial control region using a multiplex polymerase chain reaction (PCR) assay. The species were easily discerned by the number and distinct sizes of PCR products [H. melanochir, 443 bp; river garfish (H. regularis), 462 and 264 bp]. Although based on a single gene, this molecular method will correctly identify the species of individuals in at least 96% and 94% of tests for H. melanochir and H. regularis, respectively. Subsequent to verifying the identification of species by molecular discrimination, the larval development of H. melanochir and H. regularis were described. Larvae of H. melanochir and H. regularis had completed notochord flexion at hatching and are characterized by their elongate body with distinct rows of melanophores along the dorsal, lateral and ventral surfaces; small to moderate head; heavily pigmented, long straight gut; persistent preanal finfold; and extended lower jaw. Fin formation occurs in the sequence: caudal, dorsal and anal (almost simultaneously), pectoral, pelvic. Despite the similarities between both species and among hemiramphid larvae in general, H. melanochir larvae are distinguishable from H. regularis by: having 58-61 vertebrae (v. 51-54 for H. regularis); having 12-15 melanophore pairs in longitudinal rows along the dorsal margin between the head and origin of the dorsal fin (v. 19-22 for H. regularis); and the absence of a large ventral pigment blotch anteriorly on the gut and isthmus (present in H. regularis). A logistic regression analysis of body measurements also revealed interspecies differences in the combined measurements of eye diameter and pre-anal fin length. Both species can be distinguished from morphologically similar larvae found in southern Australia (other hemiramphids and a scomberosocid) by differences in meristic counts and pigmentation. Hyporhamphus melanochir larvae were successfully collected throughout Gulf St Vincent using a neuston net; however, attempts to sample eggs were unsuccessful. Abundances of larvae in the gulf averaged 4.8 and 12.3 larvae 1000⁻ ² of surface water in December 1998 and December 2000, respectively. Larvae exhibit fast growth, as indicated by otolith growth increments, with backcalculated spawning dates falling within the October-March spawning season. Spatial analysis of larval distributions revealed a positive spatial autocorrelation, i.e. non-randomness or clustering of similar abundance values. Most larvae were found in the upper region of the gulf, and the prevalence of seagrass habitat throughout this region supports the view that the demersal eggs of H. melanochir become attached to seagrass and/or algae following spawning. A gyre in waters of the upper gulf, influenced by prevailing southerly winds, the Coriolis effect, and land boundaries, may explain retention of larvae. The importance of seagrass beds to H. melanochir spawning is also supported by anecdotal evidence and available literature on eggs of other Beloniformes, which are also demersal and attach to marine plants. Dual stable isotope analysis (δ¹³ and δ¹⁵N) of larval, juvenile and adult H. melanochir and several potential food sources from the Bay of Shoals was carried out to estimate the importance of zosteracean seagrass towards the assimilated diet of H. melanochir. Although the diet of H. melanochir larvae is probably planktonivorous, their isotopic signatures partly reflect the parental diet due to the influence of pre–existing tissue in addition to growth. According to mixing model calculations, the signatures of juveniles can be explained by a diet consisting of 23–37% Zostera, 0–10% Halophila and the remainder zooplankton, whilst the diet of adults consists of 53–58% Zostera and the remainder zooplankton. These findings indicate an increasing dependence upon Zostera with growth of H. melanochir. The results of this study enhance the completeness of our understanding of the fisheries biology and ecology of H. melanochir. Significant contributions are provided in reproductive biology and larval biology, seagrass beds (in combination with mixed algae) are demonstrated to be an important habitat for spawning, and Zostera seagrass is shown to be a necessary food source in the diet of juveniles and adults. / Thesis (Ph.D.)-- University of Adelaide, School of Earth and Environmental Sciences, 2005.

This study investigates early life stages of the southern sea garfish (Hyporhamphus melanochir) and their association with seagrass in Gulf St Vincent, South Australia. The overall aims were to identify and describe the early life stages of H. melanochir and to explore the possible relationship(s) between these life stages and seagrass habitat with the emphasis on seagrass as a requirement for spawning or as a food source. The reproductive biology of female H. melanochir from the commercial fishery was assessed by microscopic examination of ovaries, oocyte size distributions, gonadosomatic indices, and macroscopic ovarian stages. Five stages of oocyte development were identified and described: perinucleolar, yolk vesicle, yolk globule, migratory nucleus and hydrated. A coherence between histological and whole oocyte descriptions is demonstrated. Hyporhamphus melanochir are characterised as multiple spawners with group-synchronous oocyte development and indeterminate fecundity. A protracted spawning season from October to March was indicated by the occurrence of ripe ovaries and increases in gonadosomatic index. Females reach sexual maturity at 193 mm standard length, and batch fecundity ranged from 201-3044 oocytes depending on fish size. Spawning shoals are segregated by sex, as indicated by commercial samples, with a biased female-to-male ratio of 4.5:1 during the spawning season (1.2:1 during the non-spawning season). In addition, features of the oocyte surface were closely examined, which revealed that the filaments on the chorion of the hydrated oocyte are adhesive. These adhesive filaments presumably allow the fertilised egg to become attached to vegetative substrate by adhesion and/or entanglement. H. melanochir larvae were discriminated from another hemiramphid species, river garfish (H. regularis), which is also known to occur in the study area, based on species-specific amplification of part of the mitochondrial control region using a multiplex polymerase chain reaction (PCR) assay. The species were easily discerned by the number and distinct sizes of PCR products [H. melanochir, 443 bp; river garfish (H. regularis), 462 and 264 bp]. Although based on a single gene, this molecular method will correctly identify the species of individuals in at least 96% and 94% of tests for H. melanochir and H. regularis, respectively. Subsequent to verifying the identification of species by molecular discrimination, the larval development of H. melanochir and H. regularis were described. Larvae of H. melanochir and H. regularis had completed notochord flexion at hatching and are characterized by their elongate body with distinct rows of melanophores along the dorsal, lateral and ventral surfaces; small to moderate head; heavily pigmented, long straight gut; persistent preanal finfold; and extended lower jaw. Fin formation occurs in the sequence: caudal, dorsal and anal (almost simultaneously), pectoral, pelvic. Despite the similarities between both species and among hemiramphid larvae in general, H. melanochir larvae are distinguishable from H. regularis by: having 58-61 vertebrae (v. 51-54 for H. regularis); having 12-15 melanophore pairs in longitudinal rows along the dorsal margin between the head and origin of the dorsal fin (v. 19-22 for H. regularis); and the absence of a large ventral pigment blotch anteriorly on the gut and isthmus (present in H. regularis). A logistic regression analysis of body measurements also revealed interspecies differences in the combined measurements of eye diameter and pre-anal fin length. Both species can be distinguished from morphologically similar larvae found in southern Australia (other hemiramphids and a scomberosocid) by differences in meristic counts and pigmentation. Hyporhamphus melanochir larvae were successfully collected throughout Gulf St Vincent using a neuston net; however, attempts to sample eggs were unsuccessful. Abundances of larvae in the gulf averaged 4.8 and 12.3 larvae 1000⁻ ² of surface water in December 1998 and December 2000, respectively. Larvae exhibit fast growth, as indicated by otolith growth increments, with backcalculated spawning dates falling within the October-March spawning season. Spatial analysis of larval distributions revealed a positive spatial autocorrelation, i.e. non-randomness or clustering of similar abundance values. Most larvae were found in the upper region of the gulf, and the prevalence of seagrass habitat throughout this region supports the view that the demersal eggs of H. melanochir become attached to seagrass and/or algae following spawning. A gyre in waters of the upper gulf, influenced by prevailing southerly winds, the Coriolis effect, and land boundaries, may explain retention of larvae. The importance of seagrass beds to H. melanochir spawning is also supported by anecdotal evidence and available literature on eggs of other Beloniformes, which are also demersal and attach to marine plants. Dual stable isotope analysis (δ¹³ and δ¹⁵N) of larval, juvenile and adult H. melanochir and several potential food sources from the Bay of Shoals was carried out to estimate the importance of zosteracean seagrass towards the assimilated diet of H. melanochir. Although the diet of H. melanochir larvae is probably planktonivorous, their isotopic signatures partly reflect the parental diet due to the influence of pre–existing tissue in addition to growth. According to mixing model calculations, the signatures of juveniles can be explained by a diet consisting of 23–37% Zostera, 0–10% Halophila and the remainder zooplankton, whilst the diet of adults consists of 53–58% Zostera and the remainder zooplankton. These findings indicate an increasing dependence upon Zostera with growth of H. melanochir. The results of this study enhance the completeness of our understanding of the fisheries biology and ecology of H. melanochir. Significant contributions are provided in reproductive biology and larval biology, seagrass beds (in combination with mixed algae) are demonstrated to be an important habitat for spawning, and Zostera seagrass is shown to be a necessary food source in the diet of juveniles and adults. / Thesis (Ph.D.)-- University of Adelaide, School of Earth and Environmental Sciences, 2005.

INTRODUCTION: Estuaries have received considerable attention from biologists in southern Africa in the past thirty years. Professor J.H. Day and the Zoology Department of the University of Cape Town have undertaken ecological surveys of a large number of estuaries of a variety of types, laying the important groundwork for more detailed studies of single species or problems. Their studies showed that the fish populations of southern African estuaries consist mainly of marine species. Why and when these species enter or leave the estuaries and whether they grow or breed in them were not established. The only relatively detailed study of a fish in southern African estuaries was that of Talbot (1955) who investigated aspects of the growth, feeding and reproduction of Rhabdosargus globiceps (Cuvier). No quantitative information has been published on any of the fish species, perhaps due to the difficulty of obtaining such data; results such as population estimates, mortality rates and growth rates are particularly unreliable when the area being investigated is in direct contact with the sea. Studies relating to even a single species of fish are hampered by immigration and emigration between estuary and sea, which make the population continuously variable. The larger estuaries are also difficult to sample adequately, especially with regard to netting, except perhaps on a very large scale using expensive commercial equipment. Along the southern African coast many of the smaller estuaries with a limited catchment area are cut off from the sea for most of the year due to erratic rainfall and longshore drift of sand. These closed or 'blind' estuaries of south eastern Africa, of which there are at least thirty in the eastern Cape Province, provide excellent areas for studying estuarine fish populations which enter when the estuary is open to the sea, but become isolated once it closes. These fish which become cut off in closed estuaries are subject to a new series of conditions. They are exposed to the wider temperature and salinity fluctuations of an estuary and they are forced to utilise the food resources of the estuary. Additionally they may be subjected to predation from piscivorous birds, and those predatory fish which are also cut off in the estuary. These factors will affect the size, mortality, and growth rates of a fish population. Since the fish are isolated no recruitment from the sea can take place, and any increase in numbers would have to come from breeding within the estuary. It is inevitable that the most numerous species should receive attention first. Rhabdosargus holubi (Steindachner) (Synonomy : Sargus holubi Stndr, Austrosparus tricuspidens Smith, Rhabdosargus tricuspidens (Smith))(Plate 1) is one of the most abundant fish in the estuaries of the eastern Cape Province. According to Smith (1965) it is endemic to southern Africa, occurring from the Cape to Zululand, being most common between Mossel Bay and East London. It is largely replaced by Rhabdosargus sarba (Forskal) in the north of its range and by Rhabdosargus globiceps (Cuvier) in the south. Commonly called the 'flatty' or silver bream it is considered only as being of nuisance value by fishermen, seldom exceeding 30 cm in length. In this study the ecology of R.holubi was examined in relation to the closed West Kleinemond estuary which is a small 'blind' estuary typical of those found along the eastern Cape coast. The growth, mortality and population size of R.holubi in this estuary were investigated and linked where possible with predation and detailed studies into the tolerances, food, and feeding of the species. It was not known at the commencement of the study whether breeding could occur in estuaries, although Smith (1965) stated that R.holubi does breed in estuaries. During the course of the sampling programme in the West Kleinemond estuary, measurements of the growth and population structure of two other species of fish were recorded for comparison with R.holubi: the sparid Lithognathus lithognathus (Cuvier) (White Steenbras) and the piscivorous carangid Hypacanthas amia (L.) (Leervis or garrick). Laboratory studies on R.holubi were possible due to the fact that large numbers of this species could be caught in eastern Cape estuaries and transported to Grahamstown, where they could be maintained in captivity for up to four months in recirculating seawater aquaria. During the course of the study over 2000 fish were kept in captivity at different times for a variety of experiments. Another 2500 were preserved for gut analyses and lipid determinations, while over 5000 were captured, examined and released in the field.

As the estimated market demand for the Agulhas sole Austroglossus pectoralis exceeds the annual catch from trawlers, this species is a potential aquaculture candidate. Broodstock conditioning and gamete preservation is part of research and development aiming at establishing a breeding protocol for a new aquaculture species. Based on a literature review of the morphology of pleuronectiform spermatozoa, this study was designed firstly, to contribute to the field of spermatozoan morphology by describing the ultrastructure of A. pectoralis spermatozoa. This was followed by an experiment to cryopreserve mature spermatozoa to provide baseline data for future studies on this and related species. The testis of A. pectoralis was a paired structure encased in a membrane, the tunica albuginea. The primary testis was located on the dorsal surface of the rib cage and the secondary testis on the ventral side. The testis was of an unrestricted spermatogonial type, based upon observations of spermatogonia along the entire length of the lobule. Mature spermatozoa of A. pectoralis had an acrosome-free ovoid head 1.68 ± 1.6μm in length and 1.7 ± 1.6μm in diameter, a short mid-piece of 0.5 ± 0.1μm in length, containing 7 irregularly shaped mitochondria forming a ring-like structure at the base of the nucleus. The flagellae were 47.4 ± 4.8μm in length, most with two plasma membrane lateral fin-like projections. However, some flagellae had either zero or three lateral fin projections. Cross-sections of the flagellae showed an axenome with a 9+2 microtubule configuration. The proximal and distal centriols were coaxal, situated deep within the nuclear fossa. The structure of A. pectoralis spermatozoa conformed to the type 1 ect-aquasperm, also found in externally fertilizing species. This type has been suggested to be the plesiomorphic form in Neopterigians. Finally, this study contributed to a cryopreservation protocol for A. pectoralis spermatozoa by testing the two cryoprotectants dimethyl sulphoxide (DMSO) and glycerol. Glycerol, at a concentration of 10%, offered better cryoprotection than DMSO. This was established using flow cytometry analysis of post-thaw nuclear membrane integrity after 64 days of storage in liquid nitrogen. The toxicity of DMSO to isolated cellular proteins may have resulted in DMSO-treated sperm having the highest percent (35.2% ± 3.2%) of non-viable cells compared with 23.0% ± 2.5% and 27.8% ± 3.4% for glycerol and the control, respectively. The presence of sucrose in the Modified Mounib Medium extender solution may explain why 45.5% ± 5% of the sperm cells were potentially viable in the control treatment. Initially, the white margined sole Dagatichthys marginatus (Soleidae) was selected as the most suitable candidate for flatfish aquaculture in South Africa. Thus, the aim of this study was to investigate the cryogenic potential and ultrastructure of D. marginatus spermatozoa. However, due to a skewed sex ratio, there were not enough males available to study this species. A skewed sex ratio is common amongst soleids, thus, the need to develop effective cryopreservation methods and to develop an understanding of sperm morphology so that the best time for cryopreservation can be chosen. In conclusion, this first description of spermatozan morphology of A. pectoralis contributed to our understanding of soleid sperm ultrastructure. In addition, a comparison of testis appearance between fish sampled just prior to spawning season and fish with mature sperm provided information on the spawning season of this species. The findings from the cryopreservation experiment suggested that glycerol was a feasible cryoprotectant for this species when sperm was prepared under field conditions.

The potential role of sanctuary areas for biological control of Schistosomiasis and development of sustainable fisheries was investigated at Cape Maclear, Lake Malawi National Park (LMNP). There has been a recent increase in the incidence of schistosomiasis infection which is a threat to the local community as well as the tourism industry which is the major source of income to LMNP as well as Chembe Village. At the same place there is increasing fishing pressure due to growing human population and declining fish resource. The increase in the incidence of schistosomiasis transmission was attributed in part to over-fishing of the molluscivorous fish which are believed to control the vector snails for schistosomiasis. Four molluscivorous fish species, Trematocranus placodon; Trematocranus microstoma; Mylochromis sphaerodon and Mylochromis anaphyrmus were reported to account for more than 90% of the fishes (by numerical abundance) which feed on the gastropods above 15 metre depth. The gastropod numbers was reported to be highest at 1.5 to 4.5 metre depth. Of the four molluscivores, T. placodon was proposed as a biological control agent for schistosomiasis based upon the previous observations of its feeding habits in artificial conditions. Captive propagation of T. placodon for reintroduction at Cape Maclear in Lake Malawi has been proposed. The present study aimed at providing baseline data required to test the hypotheses that: 1) Over-fishing of the molluscivorous fish has resulted to the increased incidence of schistosomiasis at Cape Maclear. A sub hypothesis to this was that an extension of the LMNP can act as a sanctuary area for the biological control of schistosomiasis by protecting molluscivorous fish which could control schistosomiasis vector snails. 2) A park initially designed to protect the colourful rock dwelling fish and for promotion of tourism may not effectively protect the food fish. To test the first hypothesis, the biology and ecology of T. placodon were investigated with a view to evaluating the effect this species could have on the schistosomiasis vector snail population and hence the control of bilharzia in the lake. The proportions of various gastropod species at Cape Maclear was compared with those found in T. placodon guts. Comparisons of T. placodon abundance and demographic structure inside and outside LMNP were made. To test the second hypothesis, this study investigated the food fish species that use LMNP 100 m protected zone and some basic ecological factors to appreciate the extent to which the adjacent fishery might benefit from their use of the park waters. T. placodon numerical abundance (number of individuals per unit area) ranged from 5.7 to 40.5 /200 m² and it significantly (P< 0.05) varied between sampling sites. Otter Point and Mitande which are inside the park had the lowest abundance as compared to the other three sites; Nguli inside the park; Fisheries and Nchenga outside the park. Two sites in the park, Otter Point and Mitande, had a greater proportion of mature T. placodon individuals than all other sites. The abundance of T. placodon fluctuated significantly from month to month at Nchenga, Nguli and Fisheries (X² test, P<0.0001 for all the three sites) and insignificantly (P>0.05) at Otter Point and Mitande (X² test). T. placodon densities found in the present study corresponded to the peak density of 30 individuals / 200 m² reported in 1986 but did not correspond to that of 1.0 / 200 m² for 1994. There was no evidence to support the previous reports that T. placodon abundance had decreased tremendously from 1986. The reason suggested to account for the discrepancies of T. placodon abundance reported in the present study and other studies was inadequate sampling in the previous studies which did not take into account spatial and temporal variability in T. placodon abundance. The findings reported in this thesis show that there is no need for captive propagation of T. placodon to be reintroduced into the lake at Cape Maclear and that it may prove to be unsuccessful. However, since juvenile T. placodon dominated in abundance at the three sites along the major beach which is outside the park boundaries, it is suggested that the park boundaries be extended to this area so that T. placodon should be protected to allow individuals to grow to bigger size which would be more effective for gastropod control. T. placodon between 60 mm and 80 mm TL fed on benthic insects, phytoplankton and from detritus material. Individuals between 80 mm and 100 mm fed on a mixture of benthic insects, fish scales and small gastropods and at sizes greater than 100 mm individuals specialized feeding on gastropods. Gastropods of five genera were taken and they were: Melanoides , Bulinus, Gabiella, Lanistes and Bellamya. Of these genera Melanoides fonned the greatest part of T. placodon diet. Bulinus was the second most abundant genus but compared to Melanoides its proportion was very small. Of the three Bulinus species taken by T. placodon, B. globosus, is a confirmed vector for Schistosoma haematobium which is prevalent at Cape Maclear. This species was eaten in the least quantities. A comparison ofthe five gastropod proportions in T. placodon diet and in the habitats they occupy showed that Melanoides were taken in proportionately more quantities than Bulinus at most sites. These findings contrasted the previous reports that T. placodon preferred Bulinus to Melanoides. By applying the optimal foraging theory which predicts that an animal species searching for food will go for the type of prey with the highest profitability, it is concluded that the Bulinus cannot be eliminated completely by molluscivores because if their population size falls below a certain level, the fish will switch to other gastropod types. It is concluded that the increase in schistosomiasis may not be necessarily due to overfishing the molluscivorous fish but could be due to the fact that there has been an increase in the proportion of the B. globosus albeit in small numbers which are infected with schistosomiasis parasites. An integrated approach to schistosomiasis control at Cape Maclear comprising vector control, improved water supply, sanitation and health education is suggested since no method can be effective in isolation. Few food fish species were observed to use the park at various times, varying from one species to another with regards to duration, life history stages and abundance. Only a few fish species taken by the adjacent artisanal and commercial fisheries were represented among those observed in the park. This was attributed to the limited diversity of habitat types covered. Only small population size of some species visited the protected area and only part of the life cycle of some species were observed in the park. The use of the park area was seasonal for some species and the protected zone boundaries can be crossed more than once within a day because 100 m distance is just a few minutes swim by fish. Under such circumstances the park cannot function as an effective sanctuary for food fish. An increase of the park size may be a better option to effectively protect the food fish.