The South African abalone Haliotis midae Linnaeus, 1758, is an important commercial, recreational and aquaculture mollusc species. It is the largest of the six haliotid species found in South African waters and has the second largest distributional range aside from Haliotis spadicea which is widely exploited by rock and surf anglers as bait. Analysis of population structure at Great Fish Point revealed that H. midae exhibited a high degree of microhabitat specificity, and that while dietary habits played a role in habitat selection, it was ultimately the activities of predators which confined size classes to particular niches and restricted all animals to nocturnal activities. Large (> 100 mm SL) exposed animals relied on shell thickness and adhesion to combat predators, while small (> 45 mm SL) sub-boulder animals and medium sized (50 - 95 mm SL) animals relied on their cryptic microhabitats and the protective spine canopies of co-resident urchins (Parechinus angulosus) for daytime protection. Populations of H. midae were discontinuously distributed along the coast, occupying small isolated reefs which offered a suitable array of microhabitats and a good food supply. They mostly inhabited shallow intertidal and subtidal reefs, but were occasionally encountered on deeper subtidal reefs at 4 - 5 meters. Mean length- and width-at-age were determined from growth rings composed of alternate conchiolin (dark) and aragonite (white) bands in the internal nacreous shell layer. Growth was described by the Special Von Bertalanffy growth equation: Lt(mm) = 176.998918 (1 - e⁻°·²⁴²⁴¹⁹⁽t ⁺ °·⁴⁹⁵⁴⁹⁴⁾) Wt(mm) = 159.705689 (1 - e⁻°·¹⁹⁵⁴³⁹⁽t ⁺ °·²¹¹⁶⁾) The ageing technique used was validated for animals from Great Fish Point and Mgwalana using independent tag-return data. The same data provided evidence that growth rates varied between animals from Great Fish Point and Bird Island. The growth data also showed that H. midae exhibited a high degree of individual variation in growth rate. Males and females exhibited similar growth rates. Exposed large animals showed a preference for red seaweeds, in particular Plocamium corallorhiza and Hypnea spicifera, while small sub-boulder cryptic animals included larger proportions of brown (Ralfsia expansa) and green (VIva spp.) algae in their diets. Exposed individuals also exhibited a higher degree of selectivity towards prey items, but in general, stomach contents reflected the most abundant seaweed types. Both drift and attached algal species were utilized by H. midae which was a nocturnal feeder. Pigments from red algae were incorporated into the shell layers giving the shells a pink or brick red colour. Haliotis midae is a dioecious broadcast spawner. Gonad Bulk Indices in combination with detailed histological examination of gonads showed that individuals were iteroparous, asynchronous spawners and that the breeding season extended from March through to October, although the peak spawning activity was between April and June. Males and females can spawn partially, totally or not at all, with atresia of residual gametes occurring after spawning. There is no resting stage, and gametogenesis is initiated directly after spawning. The structure of the ovary and testis and the process of gametogenesis is typical of haliotid species. AI: 1 sex ratio was observed from all populations studied. Sexual maturity was first attained in the 40 - 59 mm SL size class, although evidence for the smallest size at first spawning was recorded at 54.6 mm SL for females and 69 mm SL for males. Sizes at 50% sexual maturity were 72.5 mm SL (52.8 mm SW) at Great Fish Point, 72.5 mm SL (57.4 mm SW) at Mgwalana, 73.7 mm SL (51.2 mm SW) at Cape Recife, and 73.5 mm SL (53.8 mm SW) at Kelly's Beach. Haliotis midae was typically highly fecund, although a high degree of variation resulting in poor relationships between fecundity/shell length and gonad weight/shell length. The relationship between fecundity and gonad weight was linear. In the Eastern Cape, H. midae possessed a faster growth rate, smaller size at sexual maturity, smaller maximum size and lower longevity when compared to con specifics in Western Cape waters. A smaller minimum legal size of 93 mm SW is proposed for Eastern Cape animals and it is suggested that the closed season be moved to the peak spawning period between April and June. The benefit of a closed season during the spawning period is questioned, and the feasibility of closed areas as a management option for H. midae in the Eastern Cape is discussed.

Inadequate information of the nutritive physiology and the dietary requirements of abalone are the principle factors that currently limit the development of a formulated feed for the commercial culture of Haliotis midae. The need to develop a method to determine apparent digestibility co-efficient's for abalone in order to facilitate further applied nutritional research was identified. Animals between 50 and 80 mm were collected from natural stocks along the east Cape coast of South Africa at Port Alfred and Great Fish point, and acclimated to laboratory conditions. Initial trials demonstrated that H. midae accepted and preferred a semi-purified diet to the seaweed Plocamium corallorhiza, one of the main components of it's natural diet. A technique of determining apparent digestibility co-efficient's (ADC) using the indirect method with chromic oxide as an inert marker was developed. Digestibility trials yielded higher dry matter (DMADC) and crude protein apparent digestibility co-efficient's (CPADC) for the semi-purified diet than for two species of algae, Gelidium amanzii and P. corallorhiza (83.7% and 95.6%, 70.7% and 80.0%, and 29.9% and 57.3% respectively). The ability of the animals to utilize terrestrial animal and plant ingredients efficiently makes it feasible to use conventional feed ingredients in formulated feeds for H. midae. Trials to determine the effect of different temperatures (15°C, 18°C and 22°C) on DMADC and CPADC of the semipurified diet showed that peak digestibility occurred at 18°C. There was also a positive relationship between temperature and consumption rate. Although no enzyme studies with H. midae have been conducted, the peak ADC's at 18°C is attributed to an increase in enzyme activity at this temperature. Transit time, an inverse function of temperature and consumption, is considered to be responsible for the decrease in the ADC' s at 22°C in conjunction with a possible decrease in enzyme activity at this temperature. A photoperiod trial to investigate the effect of darkness on DMADC and CPADC of the semi-purified diet revealed that digestive efficiency decreased with increasing hours of darkness. There was also a positive relationship between duration of darkness and the rate of consumption. The decrease in ADC's is attributed to decreased transit times as the duration of darkness increased . The contribution of this project to the understanding of abalone nutrition, the development of a formulated abalone feed and systems design for abalone farms is discussed.

Thesis (MSc)--Stellenbosch University, 2002. / ENGLISH ABSTRACT: The species Haliotis midae is the only abalone species of commercial importance to the South African fisheries industry. The species is currently under substantial pressure due to commercial harvest and illegal poaching, to the extent that genetic and biodiversity may be under threat. The species is also being cultured in commercial systems for supply to international markets. The South African production for 2002 amounted to 350 tons. The commercial production of abalone may also have implications with regard to genetic diversity of natural and commercial populations. The aim of this project was to assess the genetic diversity of the natural and commercial populations of abalone at the I&J Abalone Farm, Danger Point, Gansbaai, in South Africa, in order to make recommendation with regard to possible impact on natural diversity as well as internal management systems. Distortion of natural genetic diversity or a loss of internal diversity will have detrimental consequences. Representative samples have been obtained from the surrounding natural population, as well as the commercial populations, including the broodstock and various progeny groups. Both mtDNA and AFLP molecular techniques were used to assess genetic diversity. Data analysis showed that the genetic profile of the commercial population display sufficient genetic variation. The genetic structure of the commercial population also displays no significant deviation from that of the surrounding natural population, i.e. the population of origin. The results through give indication of a small, though unsignificant loss of genetic variation from the broodstock to the subsequent progeny groups. The investigation conclude that the commercial populations of abalone at the I&J Abalone Farm, Danger Point, Gansbaai, in its current format holds no threat to the disruption of the genetic diversity of the surrounding natural population. Further, the commercial population possesses sufficient genetic variation in view of future genetic development. There is signs of a slight loss of genetic variation in the change over from the broodstock to the progeny groups. This would necessitate careful and controlled replacement of the original broodstock with new animals from the commercial progeny groups as part of an ongoing process of domestication. / AFRIKAANSE OPSOMMING: Die spesie Haliotis midae is die enigste perlemoen spesie van kommersiële belang vir Suid Afrikaanse visserye. Die spesie verkeer tans onder toenemende druk as gevolg van kommersiële en onwettige vangste, tot so 'n mate dat genetiese en biodiversiteit daardeur geaffekteer kan word. Die spesie word in toenemende mate kunsmatig gekweek vir voorsiening aan internasionale markte. Suid Afrikaanse produksie vir 2002 beloop reeds sowat 350 ton per jaar. Die kunsmatige produksie van perlemoen het ook moontlike implikasies op die genetiese diversiteit van natuurlike en kommersiële populasies van die perlemoen, Haliotis midae. Die doel van die projek was om die aard van genetiese diversiteit van die natuurlike en kommersiële populasies van perlemoen by die I&J Perlemoen Plaas, Danger Point, Gansbaai in Suid Afrika te ondersoek, ten einde aanbevelings te maak ten opsigte van moontlike impak op natuurlike diversiteit asook ten opsigte van interne genetiese bestuurstelsels. Versteuring van natuurlike diversiteit of verlies van interne diversiteit hou nadelige gevolge in. Verteenwoordigende monsters is versamel vanuit die omliggende natuurlike populasies, sowel as van die kommersiële populasie op die plaas, insluitend die teelmateriaal en verskillende nageslaggroepe. Beide mtDNA en AFLP-molekulêre tegnieke is gebruik tydens die ontleding van genetiese diversiteit. Dataontleding het aangetoon dat die genetiese profiel van die kommersiële populasie voldoende genetiese variasie demonstreer. Die genetiese struktuur van die kommersiële populasie toon verder geen betekenisvolle verskil met die omliggende natuurlike populasie nie, dit wil sê met die populasie van oorsprong nie. Die resultate toon verder 'n geringe, dog nie betekenisvolle verlies van genetiese variasie wat waargeneem word met oorgang van die teelmateriaal na die onderskeie nageslaggroepe. Die bevindinge van die ondersoek is dus dat die kommersiële populasie op die I&J Perlemoen Plaas, Danger Point, Gansbaai, in die huidige formaat, geen bedreiging inhou vir die versteuring van die omliggende natuurlike genetiese diversiteit nie. Verdere bevindinge is dat die kommersiële populasie oor voldoende genetiese diversiteit beskik met die oog op toekomstige genetiese ontwikkeling. Daar is tekens van 'n geringe afname in genetiese diversiteit met die oorgang vanaf die teelmateriaal na die nageslaggroepe. Dit beklemtoon die belang van 'n gekontroleerde vervanging van die oorspronklike teelmateriaal met nuwe teeldiere vanuit die kommersiële populasie as deel van die domestikasieproses.

The two most frequently exploited species along the south east coast of South Africa are the gastropods, Haliotis midae (abalone) and Turbo sarmaticus (alikreukel). H. midae is a high valued commercial species, and suffers intense levels of illegal fishing. T. sarmaticus however, has no commercial value but is the preferred food item for impoverished subsistence communities. Owing to the fact that no legal commercial fishery exists for either species along the south coast, very few studies have been undertaken, especially in the heavily exploited infralittoral. Infralittoral size frequency distributions for both species revealed significant variation in density and size among sites of varying exploitation pressure. Densities ranged between 0 – 2.23 m⁻² (H. midae) and 0.03 – 4.93 m⁻² (T. sarmaticus) and maximum shell lengths ranged from 49.4 – 153.5 mm (H. midae) and 28.3 – 104.4 mm (T. sarmaticus). Relatively high densities and large sizes were found in marine reserves and secluded areas, and low densities and small sizes at sites near to large population centres and within the former Ciskei homeland region. Mean size of the largest 10% of the population, total density and sexually mature density were significantly related to exploitation predictors for both species. In addition, densities of H. midae juveniles were significantly related to exploitation predictors, suggesting that recruitment may be suppressed at the most exploited sites. Exploitation of T. sarmaticus tended to be localized with refuge and subtidal populations persisting. H. midae exploitation was however, far more extensive and intense. Growth of H. midae was investigated using three methods; mark-recapture, cohort analysis and growth banding analysis at Kowie Rocks, Port Alfred. The most useful of these methods for determining growth was a new technique described for growth banding analysis; which was validated using cohort analysis and measurements of shells of known age. This technique was less time consuming and labour intensive than previously described methods. Abalone growth was best described by the Schnute (1981) growth function. Systematic geographic variation in growth was observed for 10 sites along the South African coastline. Significant differences in growth among sites existed for animals between 0-4 years (P < 0.0001) and 4-6 years (P < 0.0001), and in the mean maximum sizes attained (P < 0.001). In general, abalone from the south east/east coast were found to have faster growth rates, smaller mean maximum sizes and attained sexual maturity earlier than those along the south west/ west coast. Haliotis midae recruit and juvenile densities were found to differ significantly among sites of varying exploitation pressure (P < 0.0001) and among months for recruit densities (P < 0.001). Exploited sites had low recruit and juvenile densities compared to unexploited sites and peak recruitment occurred during October/ November 2005. Recruit densities were significantly related to infralittoral adult densities during two of the three sampling months (P<0.05), when recruitment was low. No relationship was observed during the period of high recruitment, with all sites receiving high recruit densities. It was concluded that variation in recruit densities was the result of a combination of both density-dependent relationships (i.e. local spawner density and temporal variability in recruitment intensity) and the possible dispersal capabilities of H. midae. In addition, it was concluded that at present recruitment overfishing was not occurring along the south east coast. Post-recruitment mortality rates were variable but relatively constant, with hypothetical percentage survival and density curves revealing high rates and similar mortality curves among sites. Variation in juvenile densities was consequently a result of initial recruit densities and not variation in post-recruitment mortality. T. sarmaticus populations were found to be regionally sustainable and persisted along the south east coast due to adjacent intertidal and subtidal refuge populations. However, H. midae populations are becoming decimated along the south east coast. From the information obtained in this study new management proposals were suggested and discussed, such as closed areas and region-based management fisheries together with stock enhancement. These suggestions may prove to be feasible alternatives to present management strategies.

Abalone Haliotis midae farming in South Africa is highly intensive, employing pump-ashore, flow-through systems. Despite the known sensitivity of abalone to water quality, there is only a rudimentary understanding of water quality dynamics on South African abalone farms and its effects on abalone production. Furthermore, the potential for reusing the relatively dilute abalone farm effluent to culture other animal species has not been investigated. This study investigated the dynamics of water quality and growth on a South African abalone farm and assessed the suitability of the effluent for the culture of silver kob Argyrosomus inodorus and bloodworm Arenicola loveni loveni. Monitoring of water quality and abalone growth in abalone tanks revealed that oxygen concentrations decreased, while H⁺ ion and free-ammonia (NH₃) concentration increased in a gradient between the inflow and outflow. Abalone growth was positively correlated with oxygen concentration and negatively correlated with free-ammonia and H⁺ ion concentration. The oxygen (O) concentration of the farm influent was dependent upon the influents’ temperature (T) and was described by the relationship O (mg L⁻¹) = 11.244 – 0.208T (r²=0.74). Linear regression analysis of data collected from abalone farm tanks revealed that the concentration of total ammonia at the outflow of abalone tanks (μg TAN L⁻¹) was dependant upon temperature (°C), flow-rate (L s⁻¹ kg⁻¹ H. midae), abalone size (g) and length of time since the tank was last cleaned (d) (n = 125, r² = 0.80). The production of total ammonia (μg TAN s⁻¹ kg⁻¹) was related to temperature, abalone size and days that the tanks remained un-cleaned (n = 125; r² = 0.81). A diurnal cycle of respiration was evident in abalone tanks with higher oxygen consumption and H+ ion production at night. The oxygen concentration of farm effluent was related to temperature, farm biomass and flow rate by means of a linear regression equation (n = 40; r² = 0.69). The results demonstrated the importance of optimising the flow-rate per unit of biomass for various temperatures and sizes of abalone. As abalone size and temperature cannot be controlled under farm conditions, the flow-rate per unit of biomass which the abalone culture system receives will determine the quality of the culture water. The specific growth rate (0.48 ± 0.01 % BW d⁻¹), mortality (1.8 ± 0.5 %), feed conversion ratio (3.0 ± 0.2) and protein efficiency ratio (1.0 ± 0.1) of silver kob kept in either abalone farm effluent or control seawater for 120 days did not differ significantly (t-test, P>0.05). A 90 day growth trial indicated that abalone farm effluent is a suitable culture medium for bloodworm. Bloodworm supplied with control seawater lost weight at 0.19 ± 0.04 % BW d-1, while those given abalone effluent grew at 0.39 ± 0.07 % BW d⁻¹. Mortality was 6 ± 3 % in effluent and 11 ± 8 % in seawater. The bloodworm were efficient at processing solid waste. Abalone farm effluent initially contained 7.7 ± 13 mg L⁻¹ more suspended solids than control seawater, which contained 3.5 ± 0.5 mg L⁻¹, but after passing through bloodworm systems the concentration in abalone effluent was reduced to only 1.4 ± 3.5 mg L⁻¹ above that in control seawater. Therefore, abalone farm effluent could be reused as a culture medium for both silver kob and bloodworm. Future work is needed to investigate aspects of the feasibility of such systems such as growth rates at different sizes and stocking densities.

The largest abalone in South Africa, the perlemoen, Haliotis midae, occurs along approximately two-thirds of the country's coastline, but has only been the target of an extensive commercial fishery in the south-western part of South Africa. Large-scale illegal fishing has however proliferated throughout its entire range over the last 10–15 years, which has had serious effects on stock abundance and once productive populations are facing economic collapse. Abalone stock enhancement has been put forward as an addition or alternative to traditional fisheries management practices, which can potentially rehabilitate overfished abalone populations and enhance natural production. The aim of this project was to investigate the potential of abalone stock enhancement for managing an area in the Eastern Cape Province, which was being subjected to intense illegal fishing pressure. A research approach was adopted to investigate the scale and effects of this poaching and to investigate the survival of artificially cultured abalone seed in the natural environment. A model of a commercial-scale ranching operation was investigated to assess the economic feasibility of such a scheme. Analysis of poaching cases and research samples from Cape Recife indicated high levels of fishing effort that appeared to be causing the observed declines in emergent abalone abundance and average size. Poaching cases from Cape Recife contributed 32% of the total number of cases of known origin in the Eastern Cape Province from 1998–2002, with the majority of the cases (82%) originating from within the Port Elizabeth metropole. There was an exponential increase in the number of poaching cases in the Eastern Cape Province (r²=0.967) and Port Elizabeth over this period but actual annual total catch stabilized, particularly at Cape Recife, where estimates of CPUE declined significantly from 2000–2002 (p<0.001). The annual proportion of emergent abalone under the MLS from Cape Recife was always >85% in confiscations and research collections, and did not change significantly over the period of examination. Comparison of illegal catches with an adjacent site indicated that the emergent abalone population at Cape Recife had a significantly higher proportion of undersized emergent animals (p<0.001) and they were significantly smaller than the abalone from the closest site, Noordhoek, and other areas in Port Elizabeth from 2000–2002 (p<0.001). The apparent declines in emergent abalone abundance indicated by the poaching data were corroborated by sampling of emergent abalone abundance at Cape Recife. Numbers of abalone declined significantly from 125.6 to 53.8 per 20min count (p<0.001), from April 1998–October 2001 and density declined from 1.3 to 0.8m⁻² over a similar period (p<0.001). Juvenile density did not change over this period, although there were significant differences in density observed between two different habitats (p<0.001). Initial releases of cultured juvenile abalone showed that they could be distinguished from their wild counterparts for at least a year after release, by their different shell colouration. Short-term, small-scale trials (7–10 days) using animals between 17 and 30mm SL had mean survival rates of 64–82% (mean 70.4%) and the effect of size on survival was not significant in most cases, although the power to detect differences was low. Attempts to measure the effect of habitat on survival were not conclusive, although refuges under sea urchins appeared to be favoured by both seed and wild abalone, although urchins were not an absolute requirement for survival. Similar medium-term trials (31–74 days) in sheltered sites yielded mean returns of 53.1% using 25mm SL animals and similar trials in a more exposed area had more variable survival rates of between 18.4 and 73.6%, after 25–27 days. A comparison between careful hand-seeding and surface-scattering seed release methods in the open-ocean habitat showed no significant differences in survival rate after 41 days. Larger seed survived significantly better than the smaller seed when scattered on the surface (p<0.0001) and the smaller seed survived significantly better when seeded carefully by hand onto the substrate (p<0.028). Further medium-term releases of large numbers of seed abalone, released using the surface-scattering method in open-ocean habitat, gave mean survival rates of 32.8% over periods of 83–114 days. Assessment of a seeding trial using release modules to seed juvenile animals in the open-ocean was hampered by poor sea conditions and difficult to search substrates, and a mean recovery rate of 3.1% was obtained for the sites that were sampled. Movement of seed was low in the more sheltered sites, and also appeared to be affected by the amount of available habitat. Growth of seed was measured using changes in shell colouration and there were significant differences between areas. An average growth of 1.6mm.month⁻¹ shell length was calculated for all areas. The potential for commercial scale abalone ranching to be used to enhance a territorial user right fishery was investigated using a model of economic feasibility. Internal rates of return of 30.3 and 36.9% were obtained from two harvest regimes using different harvest sizes. There was a high degree of risk associated with these figures owing to a lack of consistent seed survival rate estimates, and the input parameters, which are subject to variation, showed a significant effect on profitability. The operation of this type of ranching scheme is probably not economically feasible as a stand-alone operation but could probably be operated effectively in conjunction with an existing abalone farm. By adding a small wild catch component the profitability of a ranching scheme could be significantly improved. The results of the present study indicate that high levels of poaching at Cape Recife have led to declines in emergent abalone abundance, which will probably lead to stock collapse in the near future. This implies that fisheries managers have failed to fulfil the provisions of fisheries policy in South Africa. The present results from seeding trials show that cultured juveniles can survive and make a contribution to overall stock abundance. Furthermore, while there are obstacles to economic feasibility, commercial ranching has the potential to be a valuable addition to current abalone management strategy.

The profitability of abalone farms is heavily influenced by their production per unit of grow-out space. With farms having physically expanded to the maximum, and with increasing production costs, one of the most realistic ways for farms to increase their production is through optimizing stocking densities. The effect of stocking density on Haliotis midae performance is undocumented and optimal stocking densities for this species have not been determined. Experiments were conducted under farm conditions to investigate the effects of four different stocking densities (16 %, 20 %, 22 % and 24 % of available surface area) on growth, production and health of three different size classes of abalone (15-35 g, 45-65 g, and 70-90 g start weight). Each treatment was replicated four times and trials ran over a period of eight months with measurements being made at four month intervals. Abalone behaviour was observed during the trials in the experimental tanks. Weight gain per abalone decreased with an increase in density for all tested size classes (5.04 ± 0.18 to 2.38 ± 0.17; 5.35 ± 0.21 to 4.62 ± 0.29; 7.97 ± 0.37 to 6.53 ± 0.28 g.abalone-1.month-1 for the 15-35, 45-65 and 70-90 g classes respectively, with an increased density of 16 to 24 %). Individual weight gain of 15-35 g abalone was similar at stocking densities of 16 % and 20 % while weight gain of 45-65 g and 70-90 g abalone decreased when density was increased above 16 %. Biomass gain (kg.basket-1.month-1) was not affected by stocking density in the 15-35 g and 45-65 g size classes (1.29 ± 0.02 and 0.97 ± 0.02 kg.basket-1.month-1 respectively). However, the biomass gained by baskets stocked with 70-90 g abalone increased with stocking density (1.08 ± 0.02 to 1.33 ± 0.02 kg.basket-1.month-1) with an increased density of 16 to 24 %) and did not appear to plateau within the tested density range (16 to 24 %). Food conversion ratio did not differ significantly between densities across all size classes. Stocking density did not have a significant effect on abalone condition factor or health indices. The proportion of abalone above the level of the feeder plate increased with density (7.26 ± 1.33 to 16.44 ± 1.33 with an increased density of 16 to 24 %). As a proportion of abalone situated in the area of the basket, the same proportions were situated on the walls above the feeder plate and on the feeder plate itself irrespective of stocking density (p > 0.05). Higher proportions of animals had restricted access to feed at higher stocking densities (p = 0.03). The amount of formulated feed available on the feeder plate did not differ between stocking densities throughout the night (p = 0.19). Individual abalone spent more time above the feeder plate at higher stocking densities (p < 0.05). The percentage of time above the feeder plate, spent on the walls of the basket and on the feeding surface was not significantly different at densities of 20 %, 22 % and 24 % (p > 0.05) but abalone stocked at 16 % spent a greater percentage of time above the feeder plate on the feeding surface (83.99 ± 6.26 %) than on the basket walls (16.01 ± 6.26 %). Stocking density did not affect the positioning of abalone within a basket during the day or at night. Different size H. midae are affected differently by increases in stocking density in terms of growth performance. Findings from this research may be implemented into farm management strategies to best suit production goals, whether in terms of biomass production or individual weight gain. The fundamental mechanisms resulting in reduced growth at higher densities are not well understood, however results from behaviour observations suggest that competition for preferred attachment space and feed availability are contributing to decreased growth rates. With knowledge of abalone behaviour at different densities, innovative tank designs may be established in order to counter the reduction in growth at higher densities.

Physiological stress in farmed abalone can lead to immunosuppression and increase the susceptibility to bacterial, viral and parasitic disease, often followed by mortality. Thus, handling and poor water quality can reduce farm production efficiency. Probiotics in aquaculture have been effective in a wide range of species in enhancing immunity, survival, improving feed utilisation and growth. Three putative probionts identified as a result of in vitro screening had been beneficial to laboratory-reared abalone in a previous study. The aim of this study was to produce an abalone feed that contains a suite of probionts that may promote abalone growth and health under farming conditions. The objectives were to compare growth and physiological responses (i.e., haemocyte and phagocytosis counts) of abalone fed a commercial feed (Abfeed®S 34, Marifeed, Hermanus) supplemented with probiotics (i.e., the probiotic diet) to abalone fed the commercial feed without probiotic supplementation as a control treatment in a factorial design with handling method as an independent variable. This experiment was conducted at HIK Abalone Farm (Pty Ltd) for a period of eight months with initial weight and length 36.1 ± 0.05 g and 58.6 ± 0.06 mm abalone-1. Another experiment was carried out at Roman Bay Sea Farm (Pty) Ltd with initial weight and length 34.7 ± 0.17 g and 62.3 ± 0.18 mm abalone-1, but this experiment included one factor only, i.e. the presence and absence of the probionts in the feed. At HIK there was no significant interaction between diet and handling on average length and weight gain month-1 after four (p=0.81 and p=0.32) and eight (p=0.51 and p=0.53) months, respectively. Average length (additional handling = 73.9 ± 0.52 mm, normal farm handling = 75.8 ± 0.57 mm) and weight gain (mean: additional handling = 68.5 ± 1.20 g, normal farm handling = 74.3 ± 1.86 g) increased significantly in animals that were handled under normal farm procedure and were either fed probiotic or control diet after eight months (p=0.03 and p=0.02, respectively). There was no iii difference in length gain or weight gain of abalone fed the probiotic diet and those fed the control diet (ANOVA: F(1,16)=0.04, p=0.84; F(1,16)=0.14, p=0.71, respectively). After four months phagocytotic count was significantly different between dietary treatments with mean values of 74.50 ± 10.52 and 63.52 ± 14.52 % phagocytosis count per sample for the probionts and control treatment, respectively (p=0.04), there was no difference after eight months at HIK Abalone Farm. There was no effect of stressor application (p=0.14) and no interaction between dietary treatment and stressor application for this variable i.e., phagocytosis count (p=0.61). There was no difference in feed conversion ratio between treatments with values ranging from 2.9 to 3.8. At Roman Bay Sea farm, there was no significant difference in mean length gain between abalone fed the probiotic and control diet after eight months (repeated measures ANOVA: F(4,28)=16.54. Mean weight gain of abalone fed the probiotic diet was significantly greater than those fed the control diet after eight months (repeated measures ANOVA: F(4,28)=39.82, p(0.00001). There was no significant difference in haemocyte counts between animals fed either probiotic or control diet after four and eight months at Roman Bay Sea farm (p>0.05).

Abalone, Haliotis midae, farmers in South Africa that feed formulated diets reported a periodic drop in abalone growth during periods of increased gonad development. A large drop in abalone biomass was noticed after presumed spawning events. This study was aimed to determine the effect of diet and sex-sorting on gonad development in abalone. Experiments were conducted on a commercial abalone farm from July 2012 to the end of June 2013. Isonitrogenous and isoenergetic diets were formulated with two protein sources. A fishmeal and soybean meal (S-diet) diet and a fishmeal only (F-diet) diet were fed to abalone (50 - 70 g abalone⁻¹) over 12 months. Weight and length gain, gonad bulk index (GBI), visceral index (%) and meat mass index (%) were determined monthly and seasonally. A histological study on the female gonads was conducted. This study also included an experiment to test the effect of sex-sorting (70 - 80 g abalone⁻¹) on growth and body composition with treatments including males (M), females (F) and equal numbers of males and females (MF). Weight gain and length gain were faster in S-diet-fed abalone (RM-ANOVA, F ₍₁, ₁₆₎ = 7.77, p = 0.01; F ₍₁, ₆₉₎ = 49.9, p < 0.001, respectively). Gonad development was significantly affected by the inclusion of soybean meal with S-diet-fed abalone showing higher GBI-values than F-diet-fed abalone (RM-ANOVA, F ₍₁, ₃₃)= 16.22, p = 0.0003). Male abalone had higher GBI-values than females (RM-ANOVA, F ₍₁, ₃₃₎ = 39.87, p < 0.0001). There was no significant difference in average feed conversion ratio (FCR) between diets over time (RM-ANOVA, F ₍₁, ₂₁₎ = 0.008, p = 0.97). However, average FCR-values were significantly highest between November 2012 and March 2013, the presumed spawning season. The visceral mass (gut and gonad) as a proportion of whole mass (visceral index, %) was significantly higher in abalone fed the S-diet (RM-ANOVA; F ₍₁, ₆₉₎ = 68.06, p < 0.0001). There was no difference in meat mass index (%) between diets for both male and female abalone (RM-ANOVA; F ₍₇, ₂₄₈₎ = 0.80, p = 0.60; F ₍₇, ₂₄₁₎ = 1.7, p = 0.11,respectively). Meat mass index significantly decreased from September 2012 to February 2013 coinciding with the period of high GBI-values. The distribution of oocyte maturity stages differed between diets. The majority of oocytes within S-diet-fed abalone were fully mature stage 8 oocytes compared to a majority of stage 7 oocytes in F-diet-fed abalone. Histology corroborated peaks in GBI-values for abalone fed both diets. There was no significant difference in growth, GBI, visceral index (%) and meat mass index (%) between abalone sorted into monosex and mixed-sex populations. Thus, the presence of the opposite sex did not have an effect on growth and gonad mass in H. midae. The phytoestrogens daidzin, glycitin, genistin, daidzein, glycitein and genistein were present in soybean meal and only traceable amounts were found in the F-diet. This study provided evidence that soybean meal present in formulated feed affected growth and gonad development in H.midae. The difference in the distribution of the maturity stages of oocytes was affected by diet. Sex-sorting abalone into monosex and mixed-sex populations had no influence on weight and length gain and gonad development.

Thesis (MSc)--University of Stellenbosch, 2003. / ENGLISH ABSTRACT: The genetic status of H. midae broodstock in the South African aquaculture industry reflects that of random samples originating from undomesticated wild populations. The nature of growth in abalone is very slow, taking between three and four years to reach a marketable size of between 60 to 100 grams. It is therefore of paramount importance to improve this trait in order to ensure global competitiveness and economic viability within the industry. Improving this negative characteristic through conventional selection methods is a long-term venture and alternative means that would yield instantaneous results had to be considered. Crossbreeding was identified as an alternative, short-term strategy to improve growth rate. A crossbreeding experiment was performed between two populations of the abalone, Haliotis midae, from the East (E) and West (W) Coast of South Africa. This was done to investigate the occurrence of heterosis for growth among the crossbred genotypes (East x West, West x East). Fifteen males and females from both the East and the West Coast populations were mated in a complete dialelle crossbreeding experiment to produce four progeny groups (WW, EE, EW and WE). Progeny groups were evaluated for weight (bW) and length gain (bL) over a specific growth period of 9 months. The results provided no evidence of significant differences in weight (P = 0.085) or length gain (P = 0.244) between the four progeny groups, giving no indication of significant heterosis for weight and length gain among the crossbred progenies of these East and West Coast populations. It is recommended that further efforts to obtain improved growth rate in the abalone, Haliotis midae, through crossbreeding only be considered in light of clear evidence of substantial genetic differentiation between such populations. / AFRIKAANSE OPSOMMING: Die huidige status van perlemoen, soos dit voorkom in akwakultuur bedrywighede in Suid Afrika, weerspieël dié van ‘n ewekansige monster vanuit wilde, natuurlike populasies. Perlemoen is inherent ‘n stadig groeiende organisme wat tussen drie en vier jaar neem om tot ‘n bemarkbare grote van 60 tot 100 gram te groei. Dit is dus uiters noodsaaklik om hierdie eienskap te verbeter ten einde die bedryf ekonomies lewensvatbaar en mededingend op wêreld markte te maak. Konvensionele seleksie as ‘n metode om hierdie negatiewe eienskap te verbeter is ‘n langtermyn onderneming wat die identifisering van ‘n korttermyn metode, wat ondmiddellike resultate lewer, noodsaak. Kruisteelt is geïdentifiseer as geskikte korttermyn oplossing aangesien dit onmiddellike resultate lewer. ‘n Kruisteel eksperiment is uitgevoer tussen twee populasies van die perlemoen, Haliotis midae, van die Ooskus (E = East) en die Weskus (W = West) van Suid Afrika. Dit is gedoen om die omvang van heterose vir groeitempo in die gekruisde nageslag (East x West, West x East) te bepaal. Fyftien mannetjies en wyfies van beide die Oos- en Weskus populasies is met mekaar gepaar in ‘n volledige dialleel kruising om vier nageslag groepe (WW, EE, EW en WE) te vorm. Die nageslag is geëvalueer ten opsigte van massa (bW) en lengte (bL) toename oor ‘n spesifieke groei tydperk van 9 maande. Die eksperimentele resultate dui daarop dat die vier nageslag groepe nie betekenisvol van mekaar verskil het ten opsigte van massa (P = 0.085) en lengte (P = 0.244) toename nie en dat daar dus geen aanduiding van heterose vir massa en lengte toename in die nageslag van kruisings tussen die Ooskus en Weskus populasies bestaan nie. Daar word aanbeveel dat kruisteling as ‘n metode van genetiese verbetering van groeitempo in Haliotis midae slegs oorweeg word in die lig van nuwe molekulêre bewyse van genoegsame genetiese differensiasie tussen sulke populasies.