Dissertations / Theses on the topic 'Wrasses'

The biology of four species of Choerodon (Labridae), the blue tuskfish C. cyanodus, the bluespotted tuskfish C. cauteroma, the baldchin groper C. rubescens and the blackspot tuskfish C. schoenleinii was studied in Shark Bay in Western Australia. These species are fished commercially and/or recreationally in this large subtropical marine embayment, which is a world heritage area. The biology of C. rubescens was also studied in the Abrolhos Islands, which are located ~ 300 km to the south of Shark Bay, where this labrid is an important commercial and recreational fish species. The broad aims of this project were to determine the following for the above four Choerodon species in Shark Bay. (1) Whether they are protogynous hermaphrodites, as is the case with many labrids. (2) The biological variables required for developing management plans for these species, such as the timing of spawning, the lengths and ages at both maturity and sex change, size and age compositions and growth parameters, and (3) the habitat types occupied during their life cycles and also of the purple tuskfish Choerodon cephalotes. Finally, comparisons are made between the age and size compositions, growth and reproductive biology of C. rubescens in Shark Bay and the Abrolhos Islands. Where relevant, the underlying hypotheses for the individual studies conducted during this PhD are included in the following chapters. A macroscopic and histological examination of the gonads of the full size range of C. cyanodus, C. cauteroma, C. rubescens and C. schoenleinii, together with an analysis of the length and age compositions of female, transitional (individuals changing sex) and male individuals, demonstrated that each of these species is a protogynous hermaphrodite, i.e. individuals change sex from female to male during their life cycle. The gonads of all small (> < ca 100 mm) and young (> < ca 1 year old) individuals of each species comprised solely ovarian tissue and thus the individuals of each species began life as a female. All individuals subsequently become sexually mature as females and then later in life some will change to males. Since this was found to be the only method of sex change in these species, they are termed monandric. Individuals that were changing sex contained 'undelimited type 2' gonads sensu Sadovy and Shapiro (1987). These gonads contained both ovarian and testicular tissue that was intermixed and not separated by connective tissue. The males of each species possessed secondary testes, which retained structures of the ovary they had replaced, such as a membrane-lined ovarian lumen, lamellae and ovary wall. Furthermore, histological sections indicated that sperm were transported towards the outer walls of the testes, where the multiple sperm sinuses present in that region were presumably responsible for transporting sperm to the cloaca, rather than to a singular sperm duct as is the case with gonochoristic species. The typically large size and different colour of the males of C. rubescens, C. schoenleinii and C. cauteroma and the bias in the sex ratios of their adults towards females suggests that the males of each of these species are either haremic, i.e. permanently territorial, or form leks, i.e. are temporarily territorial during their spawning seasons. In these three species, the presence of ripe testes that are far smaller than ripe ovaries and the release by females of eggs in batches are consistent with a single male spawning with an individual female, as commonly occurs in haremic/lekking species. In contrast to the above species, C. cyanodus was not sexually dichromatic, the sex ratio was not biased towards either sex and the weight of ripe testes remained relatively constant as body weight increased. The latter implies that the relative investment of energy by males into testicular development during the spawning season declines with increasing fish size. Thus, the males of C. cyanodus may be opportunistic spawners when small, possibly spawning in groups, and may tend towards a haremic or lek mode of life when larger. The respective lengths and ages at which 50% of the females of C. cyanodus C. cauteroma and C. schoenleinii attained sexual maturity (L50m, A50m) in Shark Bay were ca 129, 196 and 253 mm and 2.3, 2.0 and 3.5 years of age. The corresponding L50m and A50m for C. rubescens in Shark Bay and the Abrolhos Islands were ca 274 and 279 mm, respectively, and 2.7 and 4.1 years of age, respectively. The respective lengths and ages at which 50% of the females of C. cyanodus, C. cauteroma and C. schoenleinii changed to males (L50c, A50c) in Shark Bay were 221, 310 and 556 mm and 4.1, 6.4 and 10.4 years of age. The length at which C. rubescens changed sex (L50c) was significantly greater in Shark Bay (545 mm) than in the Abrolhos Islands (479 mm), whereas the reverse pertained with respect to the age at sex change (A50c), i.e. 10.5 vs 11.9 years of age. Since some females were found in the oldest age classes of each species in Shark Bay and in the population of C. rubescens in the Abrolhos Islands, some of the females of each species do not apparently change sex. The trends exhibited by the gonadosomatic indices of females and males and the stages of ovarian development in sequential months demonstrated that the spawning periods of each species varied. Thus, C. rubescens (in both Shark Bay and the Abrolhos Islands) and C. cauteroma spawn predominantly in spring, whereas spawning occurs in late spring/early summer in C. schoenleinii and in summer in C. cyanodus. As C. schoenleinii, C. cyanodus and C. cauteroma occur predominantly within the inner gulfs of Shark Bay, the offset in the timing of their spawning periods would be likely to reduce any potential for competition between the larvae of those three species for resources. The trends exhibited by the mean monthly marginal increments in sectioned otoliths with differing numbers of opaque zones demonstrated that, in each species, those opaque zones were laid down annually. Thus, the numbers of opaque zones in the sectioned otoliths of individuals of each species could be used, in conjunction with the birth date and time of year when those zones are delineated, to determine their approximate ages at capture. The maximum ages recorded for the four Choerodon species in Shark Bay ranged only from 12 to 16 years. However, in that environment, the maximum lengths of C. rubescens (649 mm) and C. schoenleinii (805 mm) were far greater than those of C. cauteroma (424 mm) and C. cyanodus (382 mm). In contrast to the situation with C. rubescens in Shark Bay, this species reached a substantially older maximum age (22 years), but slightly shorter length (629 mm), and grew at a slower rate in the Abrolhos Islands, possibly reflecting the influence of greater productivity in Shark Bay and/or greater densities of this species in the Abrolhos Islands. Although a few C. rubescens and C. schoenleinii reach large sizes in Shark Bay, most of the individuals of these species were less than 400 mm, their minimum legal length (MLL) for capture. This raises the possibility that these two sought after species, i.e. the seventh and ninth most abundant species in the recreational fishery in Shark Bay, are subjected to substantial fishing pressure. Sampling for C. cyanodus was considered representative of the sites that this species occupies in Shark Bay and the sampling methods would have been likely to have captured the full size range of this tuskfish. Thus, the failure to catch any C. cyanodus greater than 400 mm indicates that, in Shark Bay, this species does not grow to the far greater lengths of about 600 mm reported for this species as a maximum by Allen (1999). Furthermore, the 400 mm MLL for this species in Western Australia precludes the retention by fishers of this species in this environment. Choerodon cauteroma was caught at lengths up to 424 mm, which is greater than the maximum of 360 mm reported for this species (Allen, 1999). Although there is no MLL for C. cauteroma, recreational fishers are restricted to a bag limit of four fish per person per day, as is the case with all other tuskfish species. Since fishers target large fish preferentially and the largest size classes of each of the species of tuskfish are dominated by males, heavy fishing pressure has the potential to remove a large proportion of the males of the Choerodon species that are fished in Shark Bay, i.e. C. rubescens, C. schoenleinii and C. cauteroma, and also of C. rubescens in the Abrolhos Islands. Since the ratio of females to males in catches of C. rubescens taken by the commercial fishery in the Abrolhos Islands are ca 1:1 and yet the typical adult sex ratio is heavily biased towards females (ca 14:1), that fishery is removing a substantial proportion of the males from the population. Protogynous hermaphroditic species are apparently able to respond to such pressure on the males by initiating a change in sex by the larger females. However, there is evidence from studies of other protogynous species that heavy size-selective fishing can lead to a reduction in the size and age at which a species changes sex and ultimately to a collapse in the stock. The results of visual surveys, when taken in conjunction with the locations of the catches of each of the five Choerodon species, demonstrated that C. rubescens lives on reefs in 'oceanic' waters along the western boundary of Shark Bay, whereas C. schoenleinii, C. cyanodus, C. cauteroma and C. cephalotes are found predominantly in the two inner gulfs of this large embayment. Choerodon cephalotes lives almost exclusively in seagrass beds, while C. schoenleinii and C. cyanodus occupy predominantly inner gulf reefs and rocky shorelines and C. cauteroma occurs in all of those three habitats. Choerodon cauteroma was the only species that underwent an obvious size-related shift during its life cycle, moving from seagrass to hard substrates, such as inner gulf reefs and rocky shorelines, as it reached adulthood. The biological and habitat data produced during this thesis will provide fisheries and environmental managers with the types of information that will enable them to develop management plans for conserving tuskfish species and their habitats in Shark Bay. The biological data for C. rubescens in the Abrolhos Islands will be able likewise to be used to develop plans for conserving the stock of this species in waters in which it is heavily fished.

The biology of four species of Choerodon (Labridae), the blue tuskfish C. cyanodus, the bluespotted tuskfish C. cauteroma, the baldchin groper C. rubescens and the blackspot tuskfish C. schoenleinii was studied in Shark Bay in Western Australia. These species are fished commercially and/or recreationally in this large subtropical marine embayment, which is a world heritage area. The biology of C. rubescens was also studied in the Abrolhos Islands, which are located ~ 300 km to the south of Shark Bay, where this labrid is an important commercial and recreational fish species. The broad aims of this project were to determine the following for the above four Choerodon species in Shark Bay. (1) Whether they are protogynous hermaphrodites, as is the case with many labrids. (2) The biological variables required for developing management plans for these species, such as the timing of spawning, the lengths and ages at both maturity and sex change, size and age compositions and growth parameters, and (3) the habitat types occupied during their life cycles and also of the purple tuskfish Choerodon cephalotes. Finally, comparisons are made between the age and size compositions, growth and reproductive biology of C. rubescens in Shark Bay and the Abrolhos Islands. Where relevant, the underlying hypotheses for the individual studies conducted during this PhD are included in the following chapters. A macroscopic and histological examination of the gonads of the full size range of C. cyanodus, C. cauteroma, C. rubescens and C. schoenleinii, together with an analysis of the length and age compositions of female, transitional (individuals changing sex) and male individuals, demonstrated that each of these species is a protogynous hermaphrodite, i.e. individuals change sex from female to male during their life cycle. The gonads of all small (> < ca 100 mm) and young (> < ca 1 year old) individuals of each species comprised solely ovarian tissue and thus the individuals of each species began life as a female. All individuals subsequently become sexually mature as females and then later in life some will change to males. Since this was found to be the only method of sex change in these species, they are termed monandric. Individuals that were changing sex contained 'undelimited type 2' gonads sensu Sadovy and Shapiro (1987). These gonads contained both ovarian and testicular tissue that was intermixed and not separated by connective tissue. The males of each species possessed secondary testes, which retained structures of the ovary they had replaced, such as a membrane-lined ovarian lumen, lamellae and ovary wall. Furthermore, histological sections indicated that sperm were transported towards the outer walls of the testes, where the multiple sperm sinuses present in that region were presumably responsible for transporting sperm to the cloaca, rather than to a singular sperm duct as is the case with gonochoristic species. The typically large size and different colour of the males of C. rubescens, C. schoenleinii and C. cauteroma and the bias in the sex ratios of their adults towards females suggests that the males of each of these species are either haremic, i.e. permanently territorial, or form leks, i.e. are temporarily territorial during their spawning seasons. In these three species, the presence of ripe testes that are far smaller than ripe ovaries and the release by females of eggs in batches are consistent with a single male spawning with an individual female, as commonly occurs in haremic/lekking species. In contrast to the above species, C. cyanodus was not sexually dichromatic, the sex ratio was not biased towards either sex and the weight of ripe testes remained relatively constant as body weight increased. The latter implies that the relative investment of energy by males into testicular development during the spawning season declines with increasing fish size. Thus, the males of C. cyanodus may be opportunistic spawners when small, possibly spawning in groups, and may tend towards a haremic or lek mode of life when larger. The respective lengths and ages at which 50% of the females of C. cyanodus C. cauteroma and C. schoenleinii attained sexual maturity (L50m, A50m) in Shark Bay were ca 129, 196 and 253 mm and 2.3, 2.0 and 3.5 years of age. The corresponding L50m and A50m for C. rubescens in Shark Bay and the Abrolhos Islands were ca 274 and 279 mm, respectively, and 2.7 and 4.1 years of age, respectively. The respective lengths and ages at which 50% of the females of C. cyanodus, C. cauteroma and C. schoenleinii changed to males (L50c, A50c) in Shark Bay were 221, 310 and 556 mm and 4.1, 6.4 and 10.4 years of age. The length at which C. rubescens changed sex (L50c) was significantly greater in Shark Bay (545 mm) than in the Abrolhos Islands (479 mm), whereas the reverse pertained with respect to the age at sex change (A50c), i.e. 10.5 vs 11.9 years of age. Since some females were found in the oldest age classes of each species in Shark Bay and in the population of C. rubescens in the Abrolhos Islands, some of the females of each species do not apparently change sex. The trends exhibited by the gonadosomatic indices of females and males and the stages of ovarian development in sequential months demonstrated that the spawning periods of each species varied. Thus, C. rubescens (in both Shark Bay and the Abrolhos Islands) and C. cauteroma spawn predominantly in spring, whereas spawning occurs in late spring/early summer in C. schoenleinii and in summer in C. cyanodus. As C. schoenleinii, C. cyanodus and C. cauteroma occur predominantly within the inner gulfs of Shark Bay, the offset in the timing of their spawning periods would be likely to reduce any potential for competition between the larvae of those three species for resources. The trends exhibited by the mean monthly marginal increments in sectioned otoliths with differing numbers of opaque zones demonstrated that, in each species, those opaque zones were laid down annually. Thus, the numbers of opaque zones in the sectioned otoliths of individuals of each species could be used, in conjunction with the birth date and time of year when those zones are delineated, to determine their approximate ages at capture. The maximum ages recorded for the four Choerodon species in Shark Bay ranged only from 12 to 16 years. However, in that environment, the maximum lengths of C. rubescens (649 mm) and C. schoenleinii (805 mm) were far greater than those of C. cauteroma (424 mm) and C. cyanodus (382 mm). In contrast to the situation with C. rubescens in Shark Bay, this species reached a substantially older maximum age (22 years), but slightly shorter length (629 mm), and grew at a slower rate in the Abrolhos Islands, possibly reflecting the influence of greater productivity in Shark Bay and/or greater densities of this species in the Abrolhos Islands. Although a few C. rubescens and C. schoenleinii reach large sizes in Shark Bay, most of the individuals of these species were less than 400 mm, their minimum legal length (MLL) for capture. This raises the possibility that these two sought after species, i.e. the seventh and ninth most abundant species in the recreational fishery in Shark Bay, are subjected to substantial fishing pressure. Sampling for C. cyanodus was considered representative of the sites that this species occupies in Shark Bay and the sampling methods would have been likely to have captured the full size range of this tuskfish. Thus, the failure to catch any C. cyanodus greater than 400 mm indicates that, in Shark Bay, this species does not grow to the far greater lengths of about 600 mm reported for this species as a maximum by Allen (1999). Furthermore, the 400 mm MLL for this species in Western Australia precludes the retention by fishers of this species in this environment. Choerodon cauteroma was caught at lengths up to 424 mm, which is greater than the maximum of 360 mm reported for this species (Allen, 1999). Although there is no MLL for C. cauteroma, recreational fishers are restricted to a bag limit of four fish per person per day, as is the case with all other tuskfish species. Since fishers target large fish preferentially and the largest size classes of each of the species of tuskfish are dominated by males, heavy fishing pressure has the potential to remove a large proportion of the males of the Choerodon species that are fished in Shark Bay, i.e. C. rubescens, C. schoenleinii and C. cauteroma, and also of C. rubescens in the Abrolhos Islands. Since the ratio of females to males in catches of C. rubescens taken by the commercial fishery in the Abrolhos Islands are ca 1:1 and yet the typical adult sex ratio is heavily biased towards females (ca 14:1), that fishery is removing a substantial proportion of the males from the population. Protogynous hermaphroditic species are apparently able to respond to such pressure on the males by initiating a change in sex by the larger females. However, there is evidence from studies of other protogynous species that heavy size-selective fishing can lead to a reduction in the size and age at which a species changes sex and ultimately to a collapse in the stock. The results of visual surveys, when taken in conjunction with the locations of the catches of each of the five Choerodon species, demonstrated that C. rubescens lives on reefs in 'oceanic' waters along the western boundary of Shark Bay, whereas C. schoenleinii, C. cyanodus, C. cauteroma and C. cephalotes are found predominantly in the two inner gulfs of this large embayment. Choerodon cephalotes lives almost exclusively in seagrass beds, while C. schoenleinii and C. cyanodus occupy predominantly inner gulf reefs and rocky shorelines and C. cauteroma occurs in all of those three habitats. Choerodon cauteroma was the only species that underwent an obvious size-related shift during its life cycle, moving from seagrass to hard substrates, such as inner gulf reefs and rocky shorelines, as it reached adulthood. The biological and habitat data produced during this thesis will provide fisheries and environmental managers with the types of information that will enable them to develop management plans for conserving tuskfish species and their habitats in Shark Bay. The biological data for C. rubescens in the Abrolhos Islands will be able likewise to be used to develop plans for conserving the stock of this species in waters in which it is heavily fished.

Genetic variation was examined in bluehead wrasse, Thalassoma bifasciatum, populations along the Florida Keys. Interpretion of mean heterozygosity (H), percent polymorphic loci per population (P), genetic similarity, and F and G statistics demonstrated a clear division of wrasse populations into "northern" and "southern" groups. Correlation and cluster analyses indicated the six reefs can be grouped in a similar fashion based on genetic and environmental data. The conclusion from this analysis of wrasse populations in the Florida Keys is that substantial population subdivision occurs in response to differing selective pressures created by heterogeneous environments.

The use of wrasse (Pisces: Labridae) as cleaner fish to combat infections with the parasitic copepods Lepeophtheirus salmonis (Kroyer) and Caligus elongatus (Nordmann) (sea-lice) in the culture of Salmo salar L. (Atlantic salmon) is now common. Infections with these parasites has caused considerable losses in the industry since its formative years. The use of the wrasse species Ctenolabrus rupestris (L. ) (goldsinny), Centrolabrus exoletus (L. ) (rockcook), Symphodus melops (= Crenilabrus melops) (L. ) (corkwing) and Labrus mixtus L. (cuckoo) as cleaner fish was first suggested in 1988. The use of these species in the industry is now widespread in Scotland, Ireland and Norway. The fish used are normally caught from the wild before being stocked with S. salar smolts during their first year at sea. The fish are routinely collected from waters close to the farm sites to be stocked. As most of the S. salar sea production sites in Scotland are located on the west coast of the country, the wrasse to be used in these sites are normally collected from these waters. The movement of wild fish into farm pens presents a risk of disease transfer from wrasse to S. salar and vice versa. Prior to their use as cleaner fish, these four species of wrasse had received little attention as subjects of scientific study. As a result, there was very little information available in the literature regarding their diseases. The present study was undertaken to investigate the potential pathogens present in wild populations in Scottish coastal waters, and, in particular, which of these pathogens, if any, could be transmitted to the S. salar. The study also investigated the susceptibility of wrasse to the two major viral diseases of S. salar to which they would be exposed in pens. In order to fully assess the pathogenicity of the potential disease agents under farm conditions, it was first necessary to establish the normal morphology of the wrasse species. Hence, a study of the morphological features of wrasse, with particular emphasis on those features important in the health of the fish was undertaken. Wrasse were shown to differ in many aspects from salmonids but shared many morphological features with other perciforme fish. Major differences from salmonids were evident in the skin, fins, pancreas, intestine, gonads and heart. There were also aspects of their morphology which differed from other perciforme fish, notably the structure of the heart. These features were regarded to be adaptations to the specific demands of their feeding strategies and habitats. This study was the first of its kind undertaken for wrasse and showed some early contraindications for the use of wrasse in culture; most notable was the marked lipid accumulation in, and resultant degeneration of, the liver resulting from the consumption of high energy S. salar feeds. Once the normal morphological features were established, it was possible to examine the disease status of wrasse. Wild fish were sampled from three different locations on the west coast of Scotland. These sites were all geographically distinct and were all used as sources of wrasse for the S. salar farming industry. Samples of wrasse were also obtained from farm sites supplied with wrasse from these wild sites, and an additional number of other geographically distinct farm locations. As a comparison wrasse were also obtained from a wrasse captive breeding facility and another captive location unrelated to the S. salar industry, a public aquarium. The fish from all of these sampling sites were examined fully for the presence of parasites, bacteria and, in some cases, viruses. Histological examination was also carried out on all of the fish studied. A total of 24 new parasite host records, and two tentative ones, were recorded from the four wrasse species studied. These new parasite records included protozoa, digeneans, nematodes and crustacea. Parasite infections were found to vary in prevalence, abundance and intensity in respect to the geographical characteristics of sampling sites and also the length of time spent in S. salar pens. It was concluded that the separation of wrasse from their natural diet and habitat influenced the degree of parasitism. None of the parasites found to infect wrasse were observed to cause any significant pathology in their hosts other than localised tissue responses. The possibility of transfer of wrasse parasites to S. salar was also investigated experimentally in a series of infections in which parasites dissected from wrasse were introduced to S. salar smolts by means of a novel gavage method. None of the parasites used established in the S. salar, indicating that there is little risk of transfaunation of parasites between wrasse and S. salar. However, this aspect requires further work due to the low number of parasites available and the subsequent low numbers of S. salar infected. Bacterial isolates were obtained from wrasse held in S. salar pens but were not found in any of the fish collected from the wild. Most of the bacterial strains isolated would normally be considered as opportunistic pathogens of fish. It was concluded that the relatively high levels of stress, both environmental and physical, that wrasse are subjected to under farm conditions were instrumental in the number of bacterial infections seen in wrasse. Only one pathogenic bacterial infection was seen in any of the fish sampled. This was an isolate of Aeromonas salmonicida, the agent known to cause the disease furunculosis, isolated from a wrasse obtained from one of the farm samples. Other authors have reported that this bacterium has already caused substantial losses of wrasse under farm conditions. It was concluded that Aeromonas salmonicida will prove to be a major pathogen of wrasse held in S. salar pens. No viruses wereI isolated from any of the wrasse studied. The susceptibility of wrasse to the most significant pathogens of S. salar under farm conditions was also subjected to investigation. In addition to sea-lice infection, the industry lists Infectious Pancreatic Necrosis (IPN) and Pancreas Disease (PD) as of primary importance for further research. Both of these diseases cause substantial losses in the industry. The susceptibility of wrasse to both of these disease conditions was investigated by means of experimental infections. In the case of IPN wrasse were infected by bathing with two different infective doses, a low dose which would be expected to induce the disease in S. salar parr and a second dose substantially higher than the first. The C. rupestris used were found to be susceptible to IPN. The wrasse developed some of the pathological characteristics typical of the disease in S. salar, however, other pathological signs were peculiar to wrasse. The recovery rate from the disease seen in wrasse was far more rapid than that recorded from S. salar. Shedding of the virus in the faeces of infected C. rupestris was also demonstrated. This study has illustrated for the first time the susceptibility of wrasse to IPN and that they can shed the virus in their faeces. This suggests that infected wrasse could be a source of continual reinfection in an affected sea site. Experimental infections of C. rupestris with PD followed a standard protocol for the reproduction of the disease in S. salar. Infection was by means of intraperitoneal injection with putatively infective material obtained from S. salar affected with PD. Two infection doses were used, the lowest dose used had been proven to be effective in inducing the disease in S. salar parr while the second dose, ten times higher than the first, had been shown to be effective in reproducing PD in S. salar smolts. The C. rupestris infected did not develop any of the typical signs of the disease seen in S. salar. It was, therefore, concluded that wrasse were not susceptible to PD.

Fishes represent half of all of the living species of vertebrates that have been described to date. Most species of fish are marine, with at least a third found exclusively in coral reefs. Yet the phylogenetic relationships of coral reef fishes and the drivers of their diversification remain poorly understood, particularly for taxonomic groups at the incipient stages of speciation. Despite the increasing application of high-throughput sequencing techniques to other taxonomic groups, much of the research conducted on coral reef fishes still relies on more traditional sources of information, such as morphology and mitochondrial sequence data. These methods are unreliable in resolving taxonomically problematic groups such as the Labridae, where many groups are still rapidly radiating, and the processes driving this are not well understood. It is therefore prudent to use a combined, integrative approach using both morphological and high-throughput sequencing techniques. This thesis uses the aforementioned techniques, integrated with morphological studies, to tease apart the relationships for members of the Labridae, in particular the fairy and pencil wrasses (Cirrhilabrus and Pseudojuloides respectively). Additionally, it includes taxonomic descriptions of eight new species, as well as investigations into general themes on coral reefs, including, but not limited to hybridisation, deep reef communities, and historical biogeography.

This study investigates the characteristics and use of spawning sites by the bluehead wrasse Thalassoma bifasciatum in Barbados, and movements of wrasses from source reefs to proximate and isolate recipient reefs. Pair spawning rate increased with increasing projection height; group spawning rate increased with increasing proximity to the downcurrent reef edge. Daily group spawning rates, but not pair-spawning rates, were higher when daily current speeds were lower, suggesting that fertilisation rates in group spawns may be more sensitive to current speed than fertilisation rates in pair spawns. Migration rate to isolated reefs was 16% that to proximate reefs. Immigration rate to recipient reefs decreased with increasing distance from the source reef and increased with increasing population density on the source reef. Immigration rates to proximate reefs were phase, sex, and size-specific, and were strongly influenced by phase, sex, and size-specific differences in home range size of wrasses. Immigration to isolated reefs was also phase and sex-specific. (Abstract shortened by UMI.)

The biogeography of wrasses within the Western Indian Ocean (WIO) is poorly understood, with regard to origins and genetic differentiation or connectivity among the regions of the WIO. This region is a good model for studying the influence of physical complexities and biogeographic breaks in shaping patterns of differentiation in wrasses. Three reef-associated fish species, Cheilio inermis, Thalassoma hebraicum and T. lunare, were selected to examine the factors that have influenced patterns of differentiation across the WIO. Each species was sampled from various localities of the WIO, the Red Sea and Indo-West Pacific Islands. Sequence data were generated from two mitochondrial gene fragments (cytochrome b and ATPase 6) and one nuclear locus (the first intron of the ribosomal protein S7 gene). Genetic analyses were used to calculate genetic diversity indices within species, which were then compared among species. The relationships among haplotypes and alleles were constructed using median-joining networks. Where necessary, neighbour-joining trees (NJ) were constructed to examine relationships among haplotypes and alleles for the Thalassoma species. Population structure was analysed using AMOVA and pairwise ФST to compare and calculate differentiation between the WIO localities. Mismatch distributions were used to examine population growth and decline or stability, and demographic parameters were used to calculate time of population expansion. There was high haplotype (h = 0.88 to 0.98) and low nucleotide diversities (π = 0.003 to 0.008) among all species for mitochondrial markers. For S7 intron I, high allelic (A = 0.95 to 0.98) and low nucleotide diversities (π = 0.002 to 0.014) were observed for all species. The pairwise ФST values revealed little to great (ФST = -0.02 to 0.67) genetic differentiation between localities, across all species for the three gene regions. The pairwise comparisons indicated the differentiation in C. inermis of Tanzania and Kenya from Mozambique and Nosy Be (Madagascar). The widespread C. inermis also revealed the differentiation of Kenya and Tanzania. For C. inermis, the AMOVAs of ATPase 6 and cytochrome b data indicated high differentiation among defined locality groups. The groups were defined according to geographic proximity. However, the AMOVA of the nuclear gene (S7 intron I) did not find variation among defined locality groups. Cheilio inermis revealed a sequence divergence of 0.4%. The divergence that was found in C. inermis was not enough to suggest a cryptic species within the WIO. Overall, the widespread and monotypic C. inermis revealed genetic differentiation within the WIO. Thalassoma hebraicum generally revealed little genetic differentiation across the WIO. The AMOVAs of the three gene regions showed no variation among specimens of the defined locality groups. However, some differentiation was found between localities. The pairwise comparisons of T. hebraicum revealed the differentiation of Seychelles from the African mainland and Madagascar. Southern Africa was observed to be differentiated from Nosy Be and Zanzibar. The observed differentiation could be caused by oceanic barriers such as the South Equatorial Current (SEQC), East African Coastal Current (EACC), and the Comoros Gyres and eddies in the Mozambique Channel, and Agulhas Current. Thalassoma lunare revealed genetic isolation between the WIO and the Red Sea as well as within the WIO. The genetic isolation between the WIO and the Red Sea is probably due to the historical isolation by the Bab al Mandab and contemporary barriers such as the cold upwelling cells in Somalia. The differentiation of Maldives from the African mainland and Seychelles could be due to distance and the upwelling cells created by monsoon winds. Mismatch distributions suggested that C. inermis and T. hebraicum had undergone demographic expansion during the Pleistocene (92 678 to 40 219 years ago). The results of the current study are similar to those from previous studies of WIO reef fish species, and the results of the present study could have potential implications for conservation and fisheries management. Single genetic markers and single species studies do not detect all barriers to dispersal in the WIO, thus they are insufficient to inform conservation management. Thus, the use of multispecies and genetic markers in the current study can be adopted by other studies of the marine taxa of the WIO.

Cleaner fish, including ballan wrasse (Labrus bergylta) have been proposed as a sustainable solution to sea lice infestations affecting farmed Atlantic salmon (Salmo salar) globally. However, in order to become sustainable, ballan wrasse need to be farmed. This thesis investigated the establishment of captive broodstock and protocols to optimise hatchery performance and productivity of ballan wrasse. High throughput sequencing was used to develop a panel of novel single nucleotide polymorphic markers (SNPs). These SNPs were used to investigate the phylogeographic structuring of ballan wrasse populations within northern geographic ranges including the UK and Norway. Results indicated fine scale population structuring within the UK suggesting that founder broodstock should be sourced locally to minimise the risk of genetic introgression with wild ballan wrasse. Secondly, captive breeding was benchmarked from harems to determine total egg production over the spawning season. Data quantified the spawning periodicity and seasonal changes in egg quality parameters. In addition, microsatellite markers identified the parental contribution to each spawning event of captive broodstock. Results confirmed, for the first time, the repeat-batch spawning behaviour and suggested that spawning events were single-paired matings. Furthermore, bottlenecks in commercial production were investigated including the benthic adhesive eggs and complex spawning behaviours of ballan wrasse within broodstock tanks. Experiments were conducted to optimise the spawning dynamics and egg productivity using fragmented spawning zones and coloured substrates. Finally, an effective bath treatment for removal of the adhesive gum layer of eggs using the proteolytic enzyme alcalase® was found to assist in egg disinfection and incubation. Overall, this research provides important baseline data on the management of broodstock and the optimisation of hatchery protocols to improve the commercial productivity and performance of ballan wrasse for use as a biological control of sea lice of farmed Atlantic salmon.

Labrus bergylta (ballan wrasse) have recently emerged as a key resource to aquaculture through proven efficacy in controlling infestations of sea lice (Leclercq et al., 2014a). However, due to complex ecology, and a complete lack of sexual dimorphism gender identification endures as a key restriction to optimising broodstock management therefore male selection and establishing optimal sex ratios is difficult (Talbot et al., 2012). L. bergylta, are noted to demonstrate unusually coloured plasma ranging in hue from green to blue with the haem catabolite biliverdin established as the causal pigment in the majority of cases (Abolins, 1961). As most vertebrates excrete biliverdin, or rapidly metabolise it to prevent toxicity, accumulation to such excess is a phenomenon which merits attention. Notably, correlation between plasma biliverdin and gender has been reported in some Labridae. Although patterns vary between species, the abundance or characteristics were such that sexual identity could be established (Gagnon, 2006). Pigment analysis was therefore proposed as a potential sex-marker in L. bergylta. In the initial experimental phase (Chapter 3), the ultimate aim was to isolate and identify the blue pigment from L. bergylta plasma, and to develop a method of quantification. The initial phase confirmed the target pigment was biliverdin IXα by visible spectroscopy, TLC, HPLC, MSMS, and a series of reactions. Following this, a protocol was developed (Chapter 2) to quantify the pigment. This method was applied accross plasma sampled from four geographically distinct wild populations with established biometrics including age, mass, length, gender and external phenotype. Subsequent analysis revealed that although pigment abundance did not vary relative to ontogeny, and there was no difference in concentration between the binary genders, plasma biliverdin was depleted in individuals undergoing sex change. Although this conclusion was complicated by significant biliverdin variation relative to origin and phenotype, which were interrelated based on relative distributions across populations, further analysis of plasma pigment in related species identified that biliverdin accumulation was associated with protogynous species. Considering the anti-oxidant capacity of biliverdin and other potentially relevant functions, this was indicative of association with the tissue remodelling processes which accompany inversion. During Chapter 3 it was noted that the biliverdin appeared tightly bound to a protein moiety. Based on the hypothesis that the pigment was actively managed and accumulated in L. bergylta plasma by this association, the next phase of experiments (Chapter 4) was an exploration of biliverdin and its binding protein in L. bergylta. The experiments revealed plasma biliverdin comigrated with the protein such that it was depleted from solution at the same rate indicating that all of the pigment was associated. Subsequent electrophoretic experiments using the fractionation products supported this, and UV fluourescence identified fragments of interest in the 25-28 kDa region. To confirm observations from the previous cross species comparison, the study was similarly expanded to include other Labrini. This revealed that although the 25 kDa band was common to all species, and genders, the 28 kDa band was collocated with the protogynous, and as such hyperbiliverdinaemic species. The 28 kDa band was sequenced using MSMS, and was identified as similar to the lipocalin Apolipoprotein A1. In combination with the properties of biliverdin, and considering that ApoA1 is analogous to serum albumin in many telesots, this supported the chromoprotein association as the main mechanism of biliverdin accumulation in such species. Further to the proposed function of biliverdin with inversion processes, and considering relevant literature, the active properties of ApoA1 suggested additional associations with prolonged altered states of metabolism which considering the ecology of L. bergylta would include gender transition, overwintering torpor and prolonged micronutrient limitation, all of which occur simultaneously. Other potential roles include modulating inflammatory responses, inhibiting pathogenic incursions and acting as an external point of contact innate immune response. From this, it was concluded that the data fully supported the previous assertions of biliverdins relevance in protogynous species, and identified a number of properties which could be of great interest to the industry in terms of welfare. The final experimental phase (Chapter 5) had two main aims. The first was to establish whether protogynous inversion could be artificially induced in L. bergylta as a means of generating male fish, and whether size had any effect on the process. The second was then to utilise controlled induction for tracking biliverdin mobilisation across the process to test the previous hypothesis. The preliminary trial demonstrated that both androgen inhibition and non-aromatisable testosterone could stimulate inversion in female L. bergylta. From this, the second trial then determined that although there was a dose dependant effect in that high androgen dosages appeared to compress the inversion process, relative size was not a factor. Gonad histology was used to create a unified scale of protogynous transition which could be expressed as a gradient to structure the biliverdin analysis. Although the biliverdin data demonstrated cryptic trends at the higher resolution gender scales, when the endpoint was condensed back to the binary gender scale employed previously (Chapter 3), the prior assertion of depletion during transition, and therefore the association with sex change associated tissue remodelling was supported. Ultimately this thesis revealed links between the biliverdin macromolecule and the highly unusual metabolic and physiological demands of gender transition in sequentially protogynous hermaphroditic temperate wrasse species.

Ballan wrasse (Labrus bergylta) is a new species in aquaculture, the newly initiated intensive production is a response to the increasing demands for wrasse as cleaner fish in salmon and trout farming. The dietary requirements of the ballan wrasse larvae are largely unknown. The nutritional quality of the live feed commonly used in the farming of marine larvae today, rotifers and Artemia sp., may be suboptimal to the dietary need of developing ballan wrasse larvae, and contribute to the problems with growth, survival and skeletal anomalies currently observed in the rearing of this species. Copepods are the natural feed for pelagic marine larvae. To use intensively cultivated copepod nauplii in the farming of marine species, either as a supplement to, or instead of rotifers or Artemia sp., have lead to improved growth and normal development, earlier onset of ossification, and less skeletal anomalies. The aim of this study was to examine the effects of four different feeding regimes from 4 to 30 days post hatch (dph), on larval growth, development and quality. The larvae in the Copepod and Cop7 treatments were fed cultivated copepod nauplii of the species Acartia tonsa from 4 to 30 dph, and from 4 to 10 dph, respectively. The larvae in the RotMG and RotChl treatments were fed enriched and unenriched rotifers (Brachionus ibericus Cayman), respectively. All groups were fed Artemia fransiscana nauplii from 24 to 51 dph, and formulated feed from 40 to 61 dph. The experiment was terminated on 61 dph. Growth and survival was recorded, and the quality difference between the larvae from the different treatments was assessed through several quality parameters; observations of larval behaviour, response to handling stress, bone ossification, and skeletal anomalies.The larvae fed copepods instead of rotifers showed better growth and stress tolerance, were more effective predators, showed earlier onset of ossification of the axial and fin ray skeleton, and had significantly less skeletal anomalies per larvae compared to the larvae fed rotifers, when the most common anomaly (twisted arches) was excluded. Larval size was consistently more related to degree of ossification than age. Growth was the parameter most notably affected by the diet, and significant differences in dry weight, standard length and myotome height was found already on 8 dph.

The use of ballan wrasse (Labrus bergylta) for the treatment of salmon lice (Lepeophtheirus samonis) in salmon farming is rapidly growing due to resistance to chemical treatments. The demand for high quality and large amounts of ballan wrasse has thus led to the development of ballan wrasse farming. Due to the small size of the newly hatched larvae, the first-feeding of the larvae has become a bottleneck area. The size and nutritional properties of the live feed chosen can greatly affect both growth and development. Using copepods as live feed in the first part of the feeding regime should give a positive effect on the larval growth and development as they are the natural prey of marine fish larvae and are nutritional superior to rotifers.The aim of this study was to use 1H-NMR to examine the differences in larval metabolic composition as a function of the different types of live feed, as well as to find the most suited method for preparation of ballan wrasse samples for metabolic analysis. To examine this, ballan wrasse larvae was fed using four different feeding regimes. The live feed used was different between the four groups for the first 30 days post hatch. After this all groups were fed the same diet. Larvae in the Copepod group were fed only the copepod Acartia tonsa. The Cop7 was fed copepods the first 7 days, then enriched rotifers. The RotMG group was fed enriched rotifers, and the RotChl group was fed unenriched rotifers. After 30 days post hatch, all treatments was fed enriched Artemia before weaning to formulated feed. Results from the present study indicate that the use of copepods in first-feeding of ballan wrasse resulted in an increase in growth compared to larvae fed rotifer diets. Metabolic analysis of the larvae show that changes in the levels of certain metabolites can be used as biomarkers for growth and development. Differences seem to be correlated to feed composition, with TMAO and taurine standing out as reliable biomarkers for growth differences

Sea lice have been causing problems in Norwegian salmon farming industry for decades; increasing mortality and reducing the welfare of farmed and wild fish. Various chemical agents have been used to combat sea lice, but increased tolerance and resistance have shifted the focus more onto biological combatants in the form of cleaner wrasse. The ballan wrasse (Labrus bergylta) is the most promising species for aquaculture, but problems with poor larval growth and survival suggests that the feeding practices in use today remain suboptimal.At present, the feeding regimes consist of enriched rotifers, followed by Artemia sp. and weaning onto formulated feed. However, the nutritional profile of copepods is known to be highly beneficial for marine fish larvae, and increased larval growth, survival and development have been reported when copepods are administered as live feed. In the present study the fish larvae were fed intensively cultivated copepod nauplii (Acartia tonsa) in order to evaluate the effects on somatic growth, survival and muscle development, compared to when being fed rotifers (Brachionus ibericus). Four different feeding regimes were used; the Copepod treatment received copepod nauplii 4-30 dph, while larvae from the Cop7 treatment received copepod nauplii 4-10 dph, followed by short-term enriched rotifers (11-30 dph). Larvae from the treatments RotMG and RotChl were given short-term enriched and cultivated rotifers respectively, 4-30 dph. Larvae from all treatments went through a transition phase onto Artemia sp. 24-30 dph, and were weaned onto formulated feed 40-50 dph. The larval somatic growth and survival were studied from start to end (2-61 dph), while muscle development was studied until 21 dph, in order to address the direct effects of copepod nauplii vs. rotifers. The results in the present study revealed that larvae fed copepod nauplii achieved higher growth rates initially, and greater dry weight already at 8 dph (also at the end of the experiment) compared to larvae fed rotifers. From 4-21 dph the increase in muscle mass was significantly larger in larvae from the Copepod treatment; accounted for by higher hyperplastic and hypertropic growth. The number of red and white fibres, size of deep white fibres, and total muscle area was found to correlate strongly with larval standard length.

Salmon production is threatened by sea louse &lt;i&gt;Lepoptheirus salmonis&lt;/i&gt;, affecting both the environment and the industry?s economy. Use of the cleaner fish ballan wrasse (&lt;i&gt;Labrus bergylta&lt;/i&gt;) is a promising method of salmon delousing with good results. The individuals used in salmon farms have so far derived from wildfish catches, but there is now interest in ballan wrasse aquaculture, as a more sustainable and environmental-friendly solution.During this experiment ballan wrasse larvae were fed with either enriched rotifers &lt;i&gt;Brachionus&lt;/i&gt; sp., followed by enriched Artemia franciscana (Rot treatment) or with reared copepods &lt;i&gt;Acartia tonsa&lt;/i&gt; (Cop treatment) until 45 day post hatch. This is probably the first study on ballan wrasse larvae with copepods as exclusive first feed. Larvae of the two treatments were compared for growth, survival and expression of seven genes (cyc1, cox5a, mnsod, fxn, crls1 and pla2g6 ) whose encoding products are localized in the mitochondrion. All genes were related to oxidative phospholyration, with two of them being parts of the final complexes of the electron transport (cyc1, cox5a).Cop larvae had a significantly higher growth during the whole experiment, while survival did not differ significantly. Gene expressions had good correlations with the larval standard length, implying that body size is more reliable than age for denoting the larval development. All genes except pla2g6 were higher expressed for the Cop treatment during the first 8 days post hatch, suggesting higher mitochondrial activity and energy (ATP) generation for the initial larval period. Specifically cox5a expression corresponded with larval dry weight increase, implying a strong molecular effect of the initial diet quality on cellular energy generation and growth. The present study underlines the importance of diet quality during the early days of the ballan wrasse life and results confirm other studies stating that reared &lt;i&gt;A. tonsa&lt;/i&gt; is an optimal fish larval live feed for this period. Results also imply that copepods have a positive impact on the mitochondrial respiration, especially for the early larval days.

Salmon lice (Lepeophtheirus salmonis) have been an increasing problem for the Norwegian aquaculture industry in recent years, and several chemicals have been used by the industry to get rid of the louse. However, resistance and reduced sensitivity towards a large quantity of these has resulted in a blooming interest for the use of wrasse as a biological method of sea louse control. The ballan wrasse (Labrus bergylta) is considered the most promising species for aquaculture, but problems with poor survival, growth and skeletal deformities suggest that a suboptimal first feeding practice may be used. At present, commercial farmers first feed the ballan wrasse larvae on enriched rotifers from 4 to approximately 30 dph, followed by Artemia sp. until weaning on formulated feed is successful. Copepods are considered the natural prey of most marine fish larvae, and usually a greater larval growth, survival and development are observed when they are used as first feed instead of rotifers. This has been attributed to the copepods high fraction of essential fatty acids in their polar lipid fraction, in addition to their great amounts of protein and free amino acids. The present study was conducted to evaluate the effect of using intensively reared copepods (Acartia tonsa naupliii) as early live feed for the ballan wrasse larvae on the larval growth and survival, and early organ growth and development, compared to using rotifers (Brachionus ibericus). Four different feeding regimes were used, varying in the live feed provided during the first 30 days. Larvae from the “Copepod”-treatment were fed exclusively with A. tonsa during this period. Larvae from the “Cop7”-treatment were fed A. tonsa from 4 to 10 dph, with a transition to enriched rotifers. Fish larvae from the “RotMG”-treatment were fed enriched rotifers the whole period, while the “RotChl”-treatment had a diet consisting of unenriched rotifers. All treatments had a transition to Artemia from 24-30 dph, and were weaned to formulated feed from 40-50 dph. Results from the present study indicated that intensively reared A. tonsa was more suitable as early live feed for ballan wrasse larvae compared to enriched or unenriched rotifers. Increased growth rates were obtained while feeding the larvae with copepods, and it resulted in larvae with significantly higher dry weight at the end of the experimental period (61 dph). No difference in larval growth was observed when feeding with enriched or unenriched rotifers, however larvae fed unenriched rotifers had a significantly lower survival than larvae from all other treatments. Higher organ volume growth rates were observed when copepods were used as feed, and the organ volumes were found to relate to the larval standard length. At 21 dph, the Copepod larvae had a significantly higher proportion of musculature than larvae from the other treatments, and the intestine appeared to be more developed and mature.

Hogfish (Lachnolaimus maximus; Walbaum 1792) from Southeast Florida were aged using sectioned otoliths and growth rates were calculated using the von Bertalanffy growth equation. The samples were collected from Broward County (n=209); other regions of Southeast Florida (n=18), the Florida Keys (n=35) and Bahamas (n=43). Growth rates were determined for each of these areas and were then compared to previously reported growth rates from other regions including the eastern Gulf of Mexico and Florida Keys. There was significant separation at the 95% confidence level between growth rates from each reagion. The average maximum fork length increased, from the Florida Keys (336mm) to Southeast Florida (414-mm) by 78-mm. However, the annual survival rate was the same (S=61%) between these two regions and the maximum age of Southeast Florida (age 12) was still half that of the previously reported eastern Gulf (age 23). Broward County was divided into three reef zones each at different depths (5-m, 10-m, and 20-m) and growth rate and survival rate were compared between zones. Results showed a decrease in maximum fork length with reef depth (857-mm, 420-mm, 352-mm), as well as an increase in mean age (age 3, 4, 5), maximum age (9, 10, 12), and survival (42%, 65%, 73%), respectively. The decrease in observed growth rate of an area as a whole (e.g. Florida Keys) may represent an example of Lee’s phenomena caused by increased top-down selective fishing pressure. However, the growth rates of individual hogfish are most likely a result of differences in habitat and food resource availability. This study provides baseline age and growth information for hogfish in Southeast Florida prior to the recent changes to the fishery regulations, which will help fisheries management better understand the effects of alternative management strategies.

Elevated levels of trace elements in the environment are of great concern because of their persistence, and their high potential to harm living organisms. The exposure of aquatic biota to trace elements can lead to bioaccumulation, and toxicity can result. Furthermore, the transfer of these elements through food chains can result in exposure to human consumers. Sea-fill or coastal fill sites are among the major anthropogenic sources of trace elements to the surrounding marine environment. For example, in the Maldives, Thilafushi Island is a sea-fill site consisting of assorted municipal solid waste, with multiple potential sources of trace elements. However, there is limited data on environmental trace element levels in the Maldives, and although seafood is harvested from close to this site, there is no existing data regarding trace element levels in Maldivian diets. Following the Christchurch earthquakes of 2011,

The mechanisms that determine the age and size structure of a population are not well understood for temperate reef species and are functionally divided into pre-settlement and post-settlement processes. Their relative importance in structuring populations can be inferred from spatial comparisons of adult age composition and growth rates, provided factors that obscure or bias the ability to detect signals from early life history are considered. These factors include migration, fishing mortality and methodological issues such as ageing errors and size-selectivity of sampling gears. The biological features that promote Notolabrus fucicola as a suitable species for demographic research and the highly variable hydrology of the east and southeast coasts of Tasmania, provide an opportunity for examination of the relative roles of pre-and post-recruitment processes in structuring temperate reef populations. In addition, N. fucicola is the primary target of a commercial live wrasse fishery that has developed over the last decade in Tasmania, necessitating an understanding of its population dynamics for effective management. Notolabrus fucicola were collected using baited traps from six locations around the east and southeast coasts of Tasmania. A methodology for age estimation was developed using thinly sliced transverse sagittal otolith sections and the periodicity and timing of increment formation, location of the first annual increment and precision of age estimations were validated. A size-selectivity function for the traps was calculated directly from sampling an isolated population of N. fucicola of known size structure, generated from a simulated fish-down using tags. The selectivity function was used to reconstruct the raw size and age structures from which fishing mortality was estimated. Fishing mortality varied spatially, with strong mortality prior to the minimum legal size limit implied at some sites. Spatial comparisons of growth were not biased by fishing mortality and indicated that growth rates followed the sea surface temperature gradient, but did not follow trends in fishing mortality, diet or reef structure. Given that growth is sensitive to resource limitation, the spatial trend in growth rates suggested that populations of N. fucicola were not structured by post-settlement density-dependent mortality. Spatial and temporal comparisons of selectivity-corrected age compositions, whilst primarily detecting differences in the level of fishing mortality between sites, indicated low variation in relative year class strength. This result is consistent with recent findings suggesting that N. fucicola are well adapted to maximize larval survival despite the highly variable coastal hydrography. Thus, it is likely that populations of the temperate wrasse N. fucicola are primarily structured by larval supply.

Temperate reef fishes in general have received less attention than their tropical counterparts, despite their prominence in shallow reef ecosystems and developing fisheries. The temperate wrasses are a case in point, where a commercial fishery in southeastem Australia has developed over the last decade, targeting the large species in the genus Notolabrus. More research directed at the biology and ecology of temperate wrasses is required to improve the understanding of wrasse population dynamics and inform the management of fisheries targeting these species. This thesis describes original studies investigating the biology and ecology of all life-history stages of Notolabrus spp. in Tasmania. The development of Notolabrus spp. is described from eggs to post-settlement juveniles, based on reared larvae and wild caught pre- and post-settlement specimens. The early life stages of the two species of Tasmanian Notolabrus are morphologically very similar and cannot be readily identified by traditional methods, such as meristic or pigmentation pattern analyses. A random fragment length polymorphism (RFLP) assay is developed to discriminate all the 5 Tasmanian species in the pseudolabrine group (Notolabrus and closely related genera). Ageing fish using otolith microstructure requires species-specific validation of the onset and periodicity of increments. The presence of druly increments in the otoliths of N. fucicola and N. tetricus is validated using oxytetracycline (OTC). The initiation of daily increments is determined, based on comparisons of the sagittal otoliths of reared larvae and the primordial region in wild caught post-settlement specimens. Daily increments are initiated in the larval otolith at yolk sac absorption, which may occur up to 14 days after fertilisation at 10 °C. Otolifh microstructure is used to determine spawning and settlement dates of N. fucicola and N. tetricus at three sites on the east coast of Tasmania, over three years. The planktonic larval duration of N. facicola varies from 40 - 87 days, back-calculated from the settlement check observed in sagittal otoliths. The settlement season for these species extends for more than 100 days over spring and summer, a period of high variability in the oceanographic conditions off the Tasmanian coast. Patterns of sea surface temperature (SST) are correlated with peaks of spawning dates, larval duration, and post-settlement growth. Settlers show earlier spawning dates in years with higher SSTs. Mean larval duration decreases as SST increases in each year. Growth is significantly faster in later settlers within a year class. A trade-off is apparent between earlier settlers spending longer in the risky planktonic phase, but arriving on the reef earlier and hence having longer to grow in the post-settlement habitat before winter, as against later settlers' shorter exposure to the planktonic environment but shorter growth phase post-settlement before water temperatures and productivity plunge. The growth of demersal juvenile and adult N. facicola is modelled using otolith and tag- recapture data. Robust methods of comparing growth models are developed, involving likelihood ratio tests, and percentile confidence intervals and 2D plots of parameters estimates derived using bootstrapping. Traditional models for describing growth in fishes, such as the von Bertalanffy growth function show high levels of correlation in their parameter estimates, and are best compared using likelihood ratio tests, and inspection of parameter plots. Notolabrus facicola growth varies seasonally, between sites, and at one site growth difference between sexes is detected. Such growth variability poses a challenge to management based on models assuming uniform growth. Growth variability in N. facicola may explain why the species is a gonochorist, rather than a protogynous hermaphrodite as in most other wrasses species. Sex change may have been abandoned because attaining and/or maintaining a clear size advantage over other members of a social group to control mating opportunities may be impossible under unpredictable growth conditions.

Estimates of the energetic costs associated with locomotion in free-living animals are difficult to find for aquatic animals. This is largely due to the technical difficulties of applying methods commonly used for air-breathing terrestrial taxa (e.g. doubly-labelled water) to aquatic environments. Consequently, there is a paucity of information on the activity costs of free-living fishes. Preliminary indications are that swimming can be a significant energetic expense for fishes. Swimming may therefore affect the allocation of metabolic energy to other fundamental functions such as growth, reproduction and cellular maintenance and thus is expected to be significant target for natural selection. However, without quantitative information on either the time or energy budgets associated with field swimming behaviours, we are unable to explore this further. In this study I observed both similarities and differences in the swimming activity during foraging of three species of coral reef wrasses. Regardless if measured in terms of time or frequency of events, all three species allocated most of their time to steady swimming (75-98% of total time). Searching and feeding were also prominent behaviours, with the extent varying across species with different feeding ecologies (benthic macrocarnivore, benthic microcarnivore and planktivore). Antagonism also varied across and within species, possibly as a consequence of the varying territoriality among species. When combined with the observed differences in field swimming speeds among species, it appears that wrasses vary their activity budgets according to their particularly trophic ecology and social structure. Combining the field observations of swimming behaviour with lab-based measures of the cost of swimming yielded estimates of the aerobic costs associated with locomotor activity by these wrasses. Such estimates of activity cost varied approximately seven-fold across species. Comparing our results with the limited current data available for field activity costs in other fishes, Stethojulis bandanensis appears to have the highest mass-specific activity cost reported for a fish to date. More broadly, it appears that my estimates for these ectothermic species are well below those for locomotor activity costs calculated for a range of endothermic taxa. Estimating and validating field activity costs in free-living fishes is still a challenge. Moreover, placing these estimates of activity cost within a broader context of field energetics is hindered by a lack of consistency in how values are reported in the literature. A convention on a common currency of costs would greatly facilitate future comparative analyses of field energetics, both within and across taxa from aquatic and terrestrial realms.

Among terrestrial organisms, body size exhibits predictable relationships with many characteristics including growth rate, mortality rate, longevity, reproductive traits, abundance, species richness and habitat use. However, the majority of studies identifying such relationships have looked at a limited range of terrestrial taxa, in particular mammals, birds and beetles. These patterns have received much less attention among marine organisms and consequently their generality is questionable. Factors influencing growth of organisms in terrestrial and marine environments may be fundamentally different. This variation could result in considerable differences in growth processes among marine and terrestrial organisms and influence constraints on body size among species in these environments. The principal aim of this study was to identify whether numerous body size-related patterns observed in terrestrial taxa were repeated in a group of coral reef fishes, and assess reasons for differences when predicted relationships were not detected. This study employed a multispecific comparative approach to examine life history and ecological correlates of body size in coral reef fishes of the wrasse family (Pisces: Labridae), a group in which species range in length from 4cm to over 2m. To account for the influence of evolutionary history of species on the patterns observed, a working hypothesis for a wrasse phylogeny was derived for the sampled species. This phylogeny was integrated into the analyses for Chapters 2 and 3 of this thesis. The study comprised four main data chapters examining relationships between body size and a range of life history traits and other ecological characteristics. In Chapter 2, the relationships between maximum body size of species and growth rate, mortality rate and longevity were examined among ten species of wrasses which encompassed a ten-fold size range. Based on current theory it was predicted that there should be a positive relationship between maximum size of species and maximum age and a corresponding negative relationship between maximum species size and mortality rate. Both of these relationships were detected for the wrasses studied here. Conceptual models indicating ways in which differences among body sizes of fish species can arise were developed and tested. It was found that in some species larger size was simply attained by growing at the same rate as smaller species but for a longer period of time. In other species faster growth enabled the attainment of larger body size but at the cost iiof shortened life-span. There was limited evidence that by growing faster individuals became larger and less susceptible to predation sooner, resulting in larger body sizes and longer life spans. A further idea was that smaller species are smaller because they have determinate growth and stop growing sooner than larger species. Wrasse species studied here exhibited the range from indeterminate to determinate growth but there was no apparent relationship between maximum body size of species and growth strategy. In Chapter 3 covariation between maximum size of species and reproductive characteristics was explored. It was predicted that smaller species should mature and change sex at a smaller proportion of maximum size and proportionally earlier in life than larger species. They were also expected to have greater reproductive effort than larger species. Despite this, none of these relationships between maximum species size and reproductive traits were evident. Relationships between maximum size and size at maturity and sex change were in fact opposite to those expected as smaller species matured and changed sex at a greater proportion of maximum size than larger species. Similarly, short-lived species matured and changed sex proportionally later in life than long-lived species. In general, body size appeared more important than age in determining when maturation and sex change occurred both among and within species. Energy invested per reproductive episode was not significantly related to species body size. In Chapter 4, covariation among body size, growth rate, longevity, reproductive effort and size/age at maturity and size/age at sex change was examined in the wrasse Halichoeres melanurus. Individuals were sampled at four locations along a latitudinal cline. Consistent with patterns identified to date it was predicted that with an increase in latitude there should be a decrease in growth rate, and an increase in body size and longevity. Initial growth rate was slightly slower at the two higher latitude locations and the maximum body size and maximum age of individuals within populations did tend to increase with an increase in latitude. It was also considered that an increase in latitude should be associated with an increase in the proportion of adult size and age attained before maturation and sex change, and an increase in reproductive effort. However, there was no consistent relationship between the latitude at which individuals were sampled and the proportion of maximum size/age attained at maturity or sex change. In addition reproductive effort of individuals did not vary predictably as latitude increased. Individuals collected at the Palm Islands matured relatively earlier, exhibited greater iiireproductive effort and changed sex proportionally earlier in life than those collected at Kimbe Bay, Lizard Island and Heron Island. In Chapter 5, relationships between body size and ecological characteristics including local abundance, species richness, habitat use and depth range, were investigated among all wrasse species present at a range of locations. Based on patterns identified within both marine and terrestrial taxa it was predicted that the smallest species would not be the most abundant with abundances peaking in species of small to intermediate size. Very large species were expected to have low abundances. This relationship was expressed for the wrasse species examined here. In addition, the body size-species richness distribution of wrasses at a number of locations was log normal, with many species of small to intermediate size and low numbers of very small or very large species. Smaller reef fish species were expected to be associated with a smaller range of microhabitats than larger species, be more habitat specific and have smaller depth ranges than larger species. The small wrasse species examined here were found to use a small, intermediate or large diversity of available microhabitats, whereas the larger species consistently used a wide diversity of microhabitats. Depth ranges of small species lay on a continuum from very small to very large, whereas larger species consistently had large depth ranges. Variation between some of the patterns observed in this study and those described in previous studies, demonstrates the need to replicate similar studies in a wide range of organisms inhabiting a wide range of habitats before their generality can be assessed. Repeating similar studies among species within a large range of reef fish families is crucial to determine the utility of species body size as a predictor of life history characteristics and other ecological variables in reef fishes.

One of the greatest challenges faced by fishery managers and ecologists is determining the degree to which natural variation in population dynamics (i.e., species abundance and life history traits) are modified by fishing. Multiscale sampling that encompasses temporal and spatial variation in anthropogenic and environmental factors are key to understanding mechanisms that drive population dynamics. To date, our understanding of these processes has been hindered by a lack of spatially explicit demographic information for many fishery targeted species. This is particularly true for multi-species, Indo-Pacific coral reef fisheries that are under increasing levels of exploitation. For example, across Indo-Pacific reefs, large-bodied wrasses are common components of reef fish assemblages and are exposed to increasing levels of exploitation, yet little is known regarding their life histories or drivers of population variability. This thesis was designed to assess the relative effects of environmental variation and fishery exploitation on the demography and ecology of tropical wrasses both spatially and temporally, to determine patterns of demographic variation and population drivers among and within species.

Tese de mestrado. Biologia (Ecologia Marinha)Universidade de Lisboa, Faculdade de Ciências, 2010<br>A Etologia tem vindo a revelar alguns aspectos da vida animal que, anteriormente, eram mal interpretados ou mesmo desvalorizados. No entanto, actualmente, é uma área cada vez mais explorada que tem vindo a contribuir para beneficio humano. Por outro lado, tendo em conta que o estudo do comportamento animal pode também contribuir para a aplicação de boas medidas de Conservação ambiental, trata-se de um instrumento importante, e por vezes indispensável. Porém, para evitar interpretações erradas e dilemas éticos, estes estudos devem ser conduzidos no ambiente natural das espécies alvo, onde o seu comportamento pode ser observado com o mínimo de interferência humana e sob a influência dos factores ambientais. Considerando o estudo de espécies de peixe, o mergulho com escafandro autónomo tem-se revelado como um método essencial para complementar a informação que é fornecida por outros métodos, uma vez que permite a observação das espécies e dos seus comportamentos no seu meio natural. Em Portugal, a prática do mergulho nem sempre é favorável. No entanto, existem alguns locais que permitem a prática desta actividade durante quase todo o ano. O Parque Marinho Professor Luiz Saldanha é um desses locais, uma vez que está, privilegiadamente, protegido dos intensos ventos de Norte (predominantes em Portugal). O presente estudo decorreu neste parque marinho e teve como principal objectivo estudar os diferentes comportamentos de três espécies de labrídeos e, simultaneamente, avaliar a sua relação com os diferentes habitats. Para isso, foi previamente elaborado um Etograma, onde os comportamentos foram identificados e definidos tendo em conta a literatura existente e as observações preliminares realizadas na área de estudo. A área de estudo é caracterizada por um recife rochoso no subtidal que apresenta uma complexidade considerável devido à erosão das falésias da serra da Arrábida, uma vez que os blocos de pedra que se soltam são de variados tamanhos acabando por gerar uma grande diversidade de micro e macro-habitats. Para além disso, este recife tem uma cobertura algal durante todo o ano, embora exista uma substituição sazonal de umas espécies de algas por outras. Deste modo, tendo em conta que os labrídeos habitam recifes com estas características, procurou-se analisar se, neste grupo de peixes, existia uma relação de dependência entre um habitat especifico e cada uma das categorias de comportamento. A amostragem decorreu através de mergulho com escafandro autónomo, aplicando-se uma amostragem focal (de um minuto), em diferentes indivíduos das espécies Symphodus bailloni, S. melops e S. roissali. A amostragem consistiu em duas campanhas: uma no Outono e outra na Primavera. Na primeira campanha visitaram-se 11 locais (distribuídos pelos três tipos de áreas de protecção do Parque Marinho Professor Luiz Saldanha) enquanto que, na segunda campanha, para além dos mesmos 11 locais houve mais um a ser visitado. Para além dessa amostragem focal, outras espécies de Symphodus (S. cinereus, S. mediterraneus, S. rostratus), consideradas raras na área de estudo, foram registadas através de uma amostragem oportunista, com o objectivo de se confrontar, posteriormente, a presença ou ausência verificadas, com os dados de um projecto que está a decorrer na mesma área e nos mesmos locais. Para ambos os grupos de espécies referidas, foi realizada uma prévia compilação de literatura sobre as mesmas, com o principal intuito de clarificar os seus diferentes padrões de coloração e as diferentes aparências que estas espécies podem apresentar. Para complementar a literatura referida, os mergulhos de treino revelaram-se uma fonte de informação importante, uma vez que permitiram observar alguns detalhes desvalorizados pelos autores consultados. No entanto, também se procurou juntar a informação disponível relativamente aos tipos de habitat onde se podem encontrar estas espécies, os locais onde a sua ocorrência já foi registada, o seu tipo de alimentação, e as suas estratégias de reprodução. Quanto ao registo da amostragem, este consistiu no apontamento de todos os comportamentos realizados por cada um dos indivíduos observados, dos habitats onde estes eram realizados (quer biótico, quer físico) e da distância observada entre os indivíduos e o substrato em cada comportamento realizado. O género, o comprimento total de cada indivíduo, e a profundidade inicial e final de cada observação eram também apontados. Para se proceder à analise dos dados, os comportamentos foram agrupados em oito categorias: Exploração, Alimentação, Repouso, Limpeza, Interacções agonísticas, Corte, Reprodução e Nidificação. Os tipos de habitat, biótico e físico, foram classificados em dois níveis. O habitat biótico, em grupos funcionais de algas ou, em alternativa, todas as espécies de algas. O habitat físico, em micro e macro-habitats. Foram ainda estabelecidas duas classes de tamanho ajustadas aos comprimentos máximos e mínimos observados para cada espécie. Os resultados indicaram uma distribuição diferente, em profundidade, das espécies principais em estudo. Contudo, o uso da complexidade do substrato, pelas diferentes espécies, para realizar categorias distintas de comportamento foi semelhante, embora com ligeiras diferenças. Para além disto, diversos aspectos relacionados com os comportamentos foram igualmente explorados. Foi examinado o tipo de esconderijo usado pelas diferentes espécies, comparando as duas campanhas de amostragem, analisando simultaneamente que classes de tamanho realizaram mais frequentemente o comportamento “esconder”. A intensidade de cada categoria de comportamento (actos por hora) foi calculada para cada espécie separadamente, assim como o número total de comportamentos realizados por cada indivíduo e o número de diferentes comportamentos em cada indivíduo. Assim pôde-se verificar em que campanha existe maior actividade, e que categorias são realizadas com maior frequência. As algas bicadas foram comparadas entre Symphodus spp. e entre campanhas, para evidenciar possíveis preferências na escolha das mesmas. Averiguou-se também se a média de bicadas ao longo da amostragem se mantinha semelhante ou se seguia alguma tendência. Quanto à nidificação, procurou-se detectar padrões no tipo de habitat que os machos escolhiam para construir os seus ninhos, na profundidade a que eram construídos, e nas algas usadas para os construir. Relativamente à posição estática, realizada pelos indivíduos como um convite à inspecção de ectoparasitas por parte das espécies limpadoras, foi procurado esclarecer se este comportamento se realiza apenas na presença de uma espécie limpadora. Também se procurou evidenciar um padrão na distribuição dos indivíduos com ectoparasitas pelas três áreas de protecção. Já as distancias entre os indivíduos e o substrato, registadas para cada comportamento durante a amostragem, foram submetidas a uma análise com o intuito de destacar as categorias de comportamento que estão mais relacionadas com o fundo. Por fim, efectuaram-se análises de sequências de categorias que foram interpretadas em diagramas de transição de estados, para identificar possíveis padrões durante os períodos de actividade reprodutora e os períodos de inactividade reprodutora. Enquanto que os comportamentos durante o Outono tinham maior probabilidade de ser seguidos por comportamentos da categoria Exploração, durante a Primavera surgiram as categorias relacionadas com a reprodução que provocaram, parcialmente, uma alteração no padrão observado anteriormente.<br>Wrasses’dependence on a specific habitat to perform a particular category of behaviour, was investigated. This was conducted after the elaboration of one Ethogram, common for the species’group studied. A focal sampling was applied in individuals of Symphodus bailloni, S. melops and S. roissali, throughout the three types of protection areas of the Arrábida Marine Park, during the Autumn and the Spring seasons. Other Symphodus spp., considerable less abundant in the study area, were opportunistically registered. A previous compilation of the literature concerning the studied species was conducted, mostly, to clarify the diverse colour patterns of these species. The sampling registration was focused in behaviours (grouped in categories) and in the habitat where they were performed, either biotic (being posterior analysed at the algae species level or at the functional groups of algae level) or physical (micro or macro-habitats were defined). Eight categories were established: Exploring, Foraging, Rest, Cleaning, Agonistic Interactions, Courtship, Reproduction and Nidification; and two size classes were defined for each species, taking into account the maximum and minimum TL observed in each one of them. Our results indicate a different distribution of the three principal studied species, by depth. However, their use of the substrate complexity, to perform distinct categories of behaviour, was similar, but not identical. Moreover, diverse aspects related to some of the behaviours, were exploited. Finally, the categories sequential analysis was interpreted to highlight possible patterns during the reproduction activity period and during the reproduction inactivity period. While behaviours, during Autumn, were more likely followed by behaviours of the Exploring category, during Spring, the categories related to reproduction emerged, partially replacing the previously observed pattern.

Sea levels fluctuated following glacial cycles during the Pleistocene, reaching approximately 115-130m below current sea levels in the Indian and Pacific Oceans during the last glacial maximum 17,000 years before present. The effects of these sea level fluctuations on population structure have been shown in many near-shore marine taxa, revealing several common patterns. However, the underlying mechanisms behind these observed patterns are largely unknown. Drops in sea level affect the distribution of shallow marine biota, exposing the continental shelf on a global scale, and displacing coral reef habitat to steep slopes where shelf breaks are shallow. In these circumstances, we expect that species inhabiting lagoons should show reduced genetic diversity relative to species inhabiting more stable outer reefs. Here, I tested this expectation on the scale of an entire ocean-basin with four wrasses (genus Halichoeres): H. claudia (N=194, with ocean-wide distribution) and H. ornatissimus (N=346, a Hawaiian endemic) inhabit seaward reef slopes, whereas H. trimaculatus (N=239) and H. margaritaceus (N= 118) inhabit lagoons and shallow habitats throughout the Pacific. Two mitochondrial markers (cytochrome oxidase I and control region) were sequenced to resolve population structure and history of each species. Haplotype and nucleotide diversity were similar among all four species. The outer reef species showed significantly less population structure, consistent with longer pelagic larval durations and a historically stable population. Mismatch distributions and significant negative Fu’s F values indicate Pleistocene population expansion for all species, and (contrary to expectations) reduced genetic diversity in the outer slope species. These data indicate that lagoonal species may persist through the loss of habitat, but are restricted to isolated refugia during lower sea level stands, which may inflate genetic diversity during high sea levels. Outer reef slope species on the other hand have homogeneous and well-connected populations through their entire ranges regardless of sea level fluctuations. These findings contradict the hypothesis that shallow species are less genetically diverse as a consequence of glacial cycles.<br>text

Tese de mestrado. Biologia (Ecologia Marinha). Universidade de Lisboa, Faculdade de Ciências, 2012<br>Cooperação é um dos mais intrigantes tipos de interações sociais presentes na natureza, pois sai fora do contexto de contante competição e típico comportamento egoísta entre indivíduos, proposto por Charles Darwin na sua teoria da Seleção Natural. A evolução e manutenção de cooperação têm sido considerados autênticos puzzles evolucionários devido à inerente dificuldade em explicar por que razão indivíduos agem em benefício de outros. O termo cooperação é usado para definir interações entre dois ou mais indivíduos cujos benefícios da interação suplantam os custos (investimentos). Este tipo de interações são denominadas por “mutualismos”, quando ocorrem entre indivíduos de espécies diferentes. Os mutualismos de limpeza marinhos estão entre os mais estudados, e a associação entre o peixe limpador obrigatório Labroides dimidiatus e os seus peixe-clientes é um exemplo típico. Estes sistemas sociais de natureza interespecífica implicam a existência de pelo menos um indivíduo limpador (normalmente de tamanho mais pequeno) que inspeciona o corpo, boca e cavidades branquiais de outro indivíduo (denominado cliente) em busca de ectoparasitas, muco, tecido morto e escamas (dos quais se alimenta). Embora muito se saiba sobre os mecanismos comportamentais que descrevem e testam a existência e manutenção de interações cooperativas e mutualísticas, pouco trabalho tem sido feito para compreender os mecanismos fisiológicos subjacentes a estas relações sociais. Recentemente foi proposto que o comportamento social é controlado por processos endócrinos, segundo os quais os esteroides sexuais (testosterona, progesterona e estradiol), esteroides do stress (como o cortisol) e alguns candidatos peptídicos, tais como os neuropéptidos da família da vasopressina (AVP)/oxitocina (OT), desempenham papéis cruciais. O sistema nervoso central é um importante coordenador das relações entre o animal, o ambiente que o rodeia e o comportamento que expressa numa dada situação. Foi recentemente proposta a existência de uma rede neural subjacente à expressão de comportamentos sociais (“social behaviour network”; SBN). Esta é uma rede de áreas cerebrais importante na regulação da expressão de comportamentos sociais. Os neuropéptidos (péptidos produzidos no cérebro com capacidade de influenciar outros neurónios) são capazes de alterar a resposta desta rede neuronal e consequentemente a expressão comportamental. Assim, estas moléculas são potenciais agentes moduladores dos comportamentos expressos no decorrer de interações mutualísticas, através de ações na SBN. O neuropéptido arginina vasotocina (AVT) e o seu homólogo arginina vasopressina (AVP; nos mamíferos) têm sido implicados numa grande variedade de comportamentos sociais em vertebrados, incluindo peixes, anfíbios, répteis, aves e mamíferos. Em teleósteos, a AVT é produzida em neurónios neurosecretores da área pré-óptica (POA), que se divide em três grupos neuronais: parvocelular (pPOA), magnocelular (mPOA) e gigantocelular (gPOA). Os neurónios da pPOA projetam maioritariamente para a pituitária (onde os neuropéptidos são armazenados antes de serem transportados para a periferia através do sistema circulatório), enquanto os neurónios da gPOA projetam sobretudo para outras áreas do cérebro (como por exemplo as incluídas na SBN). Os neurónios da mPOA projetam para a pituitária e também para outras áreas cerebrais. A maioria dos estudos sobre os efeitos da AVT em comportamentos sociais tem-se focado em interações sociais intraespecíficas. Recentemente, foi sugerido que este neuropéptido tem também um efeito importante na modulação de comportamentos sociais a nível interespecífico, no sistema mutualista que inclui o peixe limpador L. dimidiatus. De facto, um trabalho recente mostrou que injeções de AVT diminuem a probabilidade destes peixes iniciarem interações mutualísticas, enquanto injeções de um antagonista tem efeitos inversos. No seguimento do estudo acima mencionado, o presente trabalho tem como principal objetivo caracterizar e comparar o fenótipo neuronal da AVT em duas espécies filogeneticamente próximas, que divergem na expressão de comportamentos mutualísticos: um limpador obrigatório (L. dimidiatus) e um não-limpador (Labrichthys unilineatus). As duas espécies vivem em sistema de harém, sendo o macho o maior indivíduo do grupo social, que possui o território. Os limpadores L. dimidiatus dependem das interações mutualísticas para sobreviver, pois obtêm alimento exclusivamente a limpar os seus clientes. O não-limpador L. unilineatus, não apresenta qualquer comportamento de limpeza durante todo o seu ciclo de vida, alimentando-se exclusivamente do muco produzido pelos pólipos dos corais presentes no seu território. Adicionalmente, pretende-se identificar possíveis diferenças relacionadas com o sexo dos indivíduos. Para tal, foi usada uma técnica de imunocitoquímica no sentido de detetar imunorreatividade à AVT em secções de tecido cerebral. As células imunorreativas foram contadas e medidas em cada um dos grupos neuronais (pPOA, mPOA e gPOA) e foram feitas comparações entre espécies e entre sexos para cada espécie. As duas espécies analisadas apresentaram neurónios AVT-imunorreativos (-ir) limitados à área pré-óptica, o que está de acordo com outros estudos que mostram que este é o principal local de produção de AVT, em teleósteos. Os neurónios parvocelulares, que começam a surgir acima do quiasma óptico, são os mais anteriores e ventrais, e são constituídos por células pequenas e redondas, geralmente monopolares ou sem neurites óbvias. Este grupo celular estende-se para uma posição dorso-caudal, até à região onde começam a aparecer os primeiros neurónios magnocelulares. Estes ocupam uma localização medial na POA, têm aproximadamente o dobro da área dos neurónios parvocelulares e geralmente apresentam um axónio evidente que se estende em direção ao trato pré-óptico-hipofisial. Os neurónios gigantocelulares começam a aparecer numa posição dorsal relativamente aos magnocelulares e estendem-se caudalmente. Estas são as maiores células, apresentam geralmente várias neurites evidentes. Este padrão de distribuição está de acordo com a proposta conservação evolucionária das características neuroanatómicas do sistema AVT em teleósteos. Os resultados quantitativos mostram que a espécie limpadora tem menos e menores células AVT-ir relativamente à não-limpadora, e que estas diferenças são apenas significativas no grupo celular gPOA. Muitos outros estudos têm encontrado diferenças quantitativas a nível intra- e interespecífico neste grupo celular. As células AVT-ir do grupo gPOA têm sido particularmente associadas à expressão de comportamentos agressivos, sendo que espécies com maiores ou mais células, ou com maiores níveis de produção de mRNA nesta área, tendem a ser mais agressivas. Este efeito é provavelmente mediado pelas projeções destas células para zonas do cérebro onde ocorre a modulação dos comportamentos sociais. Adicionalmente, foi também demonstrado que injeções de AVT na POA induzem o afastamento social. Possivelmente, a evolução favoreceu neurónios gigantocelulares AVT-ir mais pequenos e em menor quantidade, consequentemente diminuindo a expressão de potenciais comportamentos agressivos e afastamento social interespecíficos, favorecendo assim o desenvolvimento de interações mutualísticas na espécie limpadora. A espécie não limpadora apresentou ainda uma maior densidade de fibras imunorreativas em volta dos neurónios gPOA, relativamente à espécie limpadora, que embora apresente algumas fibras estas ocorrem em menor densidade. Possivelmente, a elevada densidade de fibras observada na espécie não limpadora está relacionada com uma maior necessidade destas células comunicarem entre si (através de comunicação dendrítica), de forma a sincronizarem a libertação axonal de péptidos noutras áreas do cérebro. Isto pode ser particularmente importante em L. unilineatus para modular a expressão de comportamentos agressivos. No entanto, a partir das secções de tecido usadas neste trabalho, não é possível determinar se as fibras encontradas são extensões dendríticas dos corpos celulares dos neurónios gPOA. Assim, esta sugestão deve ser tomada com alguma precaução e trabalhos futuros deverão usar secções de tecido mais finas para verificar esta situação. O número e tamanho dos neurónios nos outros dois grupos celulares não foi significativamente diferente entre as duas espécies. Os neurónios pPOA e mPOA projetam para a pituitária, onde têm a capacidade de estimular a secreção de ACTH e consequentemente modular a reatividade do eixo hipotálamo-pituitária-interrenal (HPI). Assim, a falta de diferenças nestes grupos celulares entre as duas espécies pode sugerir que: a) os animais não se encontravam sob stress, e a falta de diferenças pode refletir processos fisiológicos semelhantes às duas espécies; b) caso os animais estivessem sob stress, as respostas fisiológicas não foram diferentes entre as duas espécies; c) o confinamento a que os animais estiveram sujeitos durante o transporte representou um stress crónico, causando habituação do sistema AVT nas duas espécies. As duas últimas hipóteses apresentadas são as mais prováveis pois o confinamento espacial é uma fonte de stress fisiológico. Adicionalmente às projeções para a pituitária, o grupo neuronal mPOA também projeta para outras áreas cerebrais capazes de modular comportamentos sociais. A falta de diferenças entre as duas espécies neste grupo celular vem enfatizar a ideia de que a expressão de comportamentos mutualísticos poderá estar associada às projeções centrais das células AVT-ir da área gPOA. Ao contrário do esperado, não foram encontradas diferenças no número ou tamanho dos neurónios AVT-ir entre sexos, em ambas as espécies. Diferenças entre sexos eram esperadas com base em diferenças comportamentais entre machos e fêmeas. Em L. dimidiatus sabe-se que os machos apresentam maiores níveis de agressividade intraespecífica, que é dirigida às fêmeas quando estas são desonestas com os seus clientes (i.e. quando elas ingerem muco em vez de ectoparasitas). Em L. unilineatus, os machos patrulham e defendem o seu território contra possíveis machos competidores. Muitos trabalhos têm associado diferenças entre sexos nas características das células AVT-ir a esteroides sexuais, principalmente androgénios. No entanto, existem outros estudos que também não encontraram diferenças significativas entre sexos. A falta de diferenças entre machos e fêmeas pode refletir uma falta de influência dos androgénios no sistema AVT, nas duas espécies usadas no presente estudo. No entanto o presente estudo não foi desenhado para responder a questões sobre a influência de outras hormonas no sistema AVT, pelo que experiências futuras devem testar a hipótese proposta de uma forma mais fidedigna. Em conclusão, este estudo mostra, pela primeira vez, diferenças no fenótipo neuronal da AVT entre duas espécies filogeneticamente próximas que diferem na expressão de comportamentos mutualísticos. A espécie mutualista Labroides dimidiatus apresenta menos e menores neurónios gPOA, em relação à espécie não mutualista. Neste trabalho eu sugiro que os mais baixos níveis de AVT na espécie limpadora (inferidos pelo menor tamanho e número de neurónios AVT-ir) desempenham um papel importante na expressão de comportamentos mutualísticos, possivelmente modulando a predisposição para aproximação, interação e cooperação com parceiros sociais interespecíficos. Adicionalmente, o grupo neuronal gPOA e as suas projeções extra-hipotalâmicas aparentam ser responsáveis pela expressão dos elevados níveis de pro-socialidade interespecífica típicos do peixe limpador. Embora a maioria dos estudos sobre a relação do sistema AVT e comportamentos sociais se concentrem em relações intraespecíficas, o presente estudo fornece evidências de que este sistema desempenha também um papel importante na modulação de comportamentos sociais entre espécies diferentes, e concretamente comportamentos mutualísticos.<br>Mutualistic interactions are particular social interactions where individuals of different species cooperate in order to gain benefits. A textbook example of mutualism is provided by the bluestreak cleaner wrasse (Labroides dimidiatus), which feeds on clients’ ectoparasites, mucus, scales and dead or infected tissue. Although much is known about the ultimate functions of cooperative behaviour, studies on their underlying proximate mechanisms are scarce. The nonapeptides arginine vasotocin (AVT) and its mammalian homologue, are well known for their function in the modulation of several conspecific social behaviours. Recently, AVT was also shown to play an important role in the modulation of interspecific cooperative behaviours in the cleaner wrasse, as AVT injections decreased the cleaners’ willingness to engage in cleaning interactions. In this study, I aimed to characterize and compare the AVT neuronal phenotypes of two closely related species that live in similar environments, but diverge in the expression of interspecific cooperative behaviours: an obligate cleaner (L. dimidiatus) and a non-cleaner, corallivore labrid (Labrichthys unilineatus). Additionally, I aimed to identify sex differences in AVT neuronal phenotypes, as they might be responsible for the expression of sex-related social behaviours. Species differences were restricted to the gPOA (known to have extrahypothalamic projections). Cleaners had smaller and less numerous AVT immunoreactive (-ir) neurons, compared to the non-cleaners. I propose that this neuronal preoptic group plays an important role in the modulation on interspecific cooperative behaviours. Also, I suggest that smaller and less numerous neurons projecting to extrahypothalamic brain areas, might have been selected to lessen the expression of potential interspecific aggressive behaviours and interspecific social withdrawal, thus facilitating the evolution and maintenance of mutualistic interactions. No intersexual differences were found for both species which might reflect a lack of androgen interaction with the AVT system. Alternatively, the lack of sex differences might reflect that both males and females engage in interspecific cooperative relationships, and the expressed mutualistic behaviours do not differ between sexes.

Cleaning symbioses are a well-studied mutualism among marine species. However, the interactions occurring between cleaner fish and sharks are lacking in research, which makes it a target for further investigation. With this study, intentions were to analyse the behaviour of two kinds of cleaner species, bluestreak cleaner wrasse (Labroides dimidiatus) and moon wrasse (Thalassoma lunare), to be able to distinguish differences in cleaning behaviour on pelagic thresher sharks (Alopias pelagicus). A total of 68,4 hours of video was recorded on the edge of a seamount outside of Malapascua, called Monad Shoal, during 18 days in January 2018. The number of interactions were divided into two categories, where the behaviour was classified as an inspection or a bite and could occur on different patches of the sharks’ body (head, gills, body, dorsal, pectoral, pelvic or caudal fin). In total 118 events occurred which comprised in total 4079 interactions from the two cleaner species. Of these interactions 3626 were considered inspections and 453 were bites. Bluestreak cleaner wrasse conducted 3598 of the inspections and 28 of the inspections were conducted by the moon wrasse. All bites were conducted by bluestreak cleaner wrasses. The results indicated a preference in patches of the body to inspect, where the pelvis got the most inspections on 34,1 %, followed by the pectoral fins on 22,8 %. The dorsal fin and the gills accounted for the least number of bites, with 1,3 % on the dorsal and 1,4 % on the gills. Furthermore, a difference in inspected patches between males and females were discovered, where females got significantly more inspections on their head, gills, body, dorsal and pectoral fin. The pelvis and caudal fin did not show any significant differences.

The three-spot wrasse (Halichoeres trimaculatus) is distributed in and around the coral reefs and shallow rocky areas in the tropical and subtropical Indo-Pacific regions. This species displays a distinct diurnal behavior, burrowing under the sand at dusk and emerging out of the sand at dawn, which appears to be synchronized to the photoperiod. In this thesis, the hypothesis tested was that this unique life-style subjected the animal to daily hypoxia exposure while under the sand at night. The measurements of oxygen concentration in the sand around the fish at night confirmed a complete lack of oxygen. The study had three specific objectives: i) obtain a tissue-specific temporal profile of the hypoxia-related molecular and biochemical responses in wrasse over a 24 h diurnal cycle, ii) determine the responses that were unique to sand dwelling and iii) determine if the responses seen at night in the sand are similar to an anoxic response in this species. Wrasse were maintained in a flow-through seawater aquaria (29 ±1°C), with sand at the bottom for the fish to hide, and kept under natural photoperiod. The fish were sampled at 10:00, 14:00, 18:00, 21:00, 24:00, 3:00, and 6:00 clock time and plasma and tissue (brain, liver, gill, heart and muscle) were collected to determine the molecular and biochemical responses over a 24 h period. Fish were also sampled from aquaria without sand at night to determine the responses that were specific to hiding in the sand, while fish exposed to nitrogen gas bubbling for 6 and 12 h served as the anoxic group. A partial cDNA sequence of the hypoxia-inducible factor (HIF)-1α and neuroglobin (two genes that are hypoxia-responsive) were cloned and sequenced from the liver and brain, respectively, and their expression was determined using real-time quantitative PCR. HIF-1α mRNA abundance was higher in the brain compared to the liver and the gills, while a clear pattern of diurnal change in tissue HIF-1α and brain neuroglobin gene expressions was not observed at night relative to the fish during the day. However, wrasse brain showed a significant reduction in glycogen content at night under the sand and this corresponded with a higher hexokinase activity and increased glucose level suggesting enhanced glycolytic capacity. The plasma glucose and lactate levels were significantly lower at night, while in sand, relative to the day. The lower plasma glucose at night corresponded with a significant drop in liver gluconeogenic capacity (reduction in phosphoenolpyruvate carboxykinase, a key gluconeogenic enzyme, activity), while the lower lactate levels support a lack of activity along with the absence of glycogen breakdown in the muscle. Overall, there was a reduction in the metabolic capacity in the gills, heart, liver and muscle, but not the brain, supporting a tissue-specific metabolic reorganization as an adaptive strategy to cope with sand-dwelling in the wrasse. The molecular and biochemical responses seen in the wrasse at night in the sand was dissimilar to that seen in fish exposed to anoxia, leading to the conclusion that this species is not experiencing a complete lack of oxygen while under the sand. Also, the lack of muscle movement associated with sand dwelling at night limits anaerobic glycolysis for energy production, thereby eliminating lactate accumulation that was evident in fish exposed to anoxia. Taken together, wrasse showed a tissue-specific difference in metabolic capacity at night while hiding under the sand. While the mechanism involved in this tissue-specific energy repartitioning at night is unclear, one hypothesis involves selective increase in blood flow to the brain, while limiting peripheral circulation, as a means to maintain oxygen and glucose delivery to this critical tissue while the fish is hiding under the sand. The higher metabolic capacity of the brain, but not other tissues, at night under the sand suggests that maintaining the brain function is essential for the diurnal life-style in this animal.

Tese de mestrado. Biologia (Ecologia Marinha)Universidade de Lisboa, Faculdade de Ciências, 2014<br>Cooperação é geralmente definida como qualquer interacção entre indivíduos em que os custos de investimento nessa parceria são suplantados pelos seus benefícios. Quando as interacções se desenrolam entre indivíduos de espécies distintas, são referidas como mutualismos. Neles se incluem os mutualismos de limpeza. Os mutualismos de limpeza envolvem duas partes: um limpador, de pequenas dimensões e com uma coloração conspícua, e um cliente, geralmente de maiores dimensões. Os organismos que estabelecem este tipo de cooperação acabam por desenvolver adaptações ao nível morfológico e comportamental que facilitam a interacção. O bodião limpador Labroides dimidiatus, juntamente com um restrito número de espécies das famílias Labridae e Gobiidae, apresenta o mais alto nível de especialização em comportamentos de limpeza, sendo classificado como limpador obrigatório, ou seja, a sua alimentação baseia-se exclusivamente no que consegue remover dos seus clientes. Os limpadores encontram-se em territórios específicos, designados por estações de limpeza, e vivem num sistema de harém (espécie poligínica e protogínica). Numa interacção de limpeza, os limpadores removem da superfície dos seus clientes ectoparasitas e tecido infectado, ferido ou morto. A interacção é geralmente iniciada pelos clientes, que adoptam uma posição específica ficando imóveis e abrindo a boca, opérculos e barbatanas peitorais, mostrando assim que desejam ser limpos. Por usa vez, os limpadores adoptam um comportamento específico de dança, nadando para cima e para baixo. Esta dança não só chama a atenção dos seus clientes, como também serve como estratégia de préconflito, pois reduz a agressividade dos seus clientes. Outra estratégia pré-conflito utilizada pelos limpadores é a estimulação táctil, durante a qual, usando as barbatanas pélvicas e peitorais, “massajam” o corpo do cliente durante a inspecção. As massagens não só exercem um efeito anti-stress sobre o cliente, como também aumentam a sua fitness. Por causa disso, os limpadores manipulam os seus clientes a ficarem mais tempo na estação de limpeza, mesmo depois de um acto desonesto. Os limpadores são desonestos quando removem muco e escamas dos clientes, que os limpadores preferem, mas que é prejudicial aos clientes pois requerem alocação de energia para reposição dos mesmos. Por ser a sua preferência, os limpadores cativam clientes de maior valor nutricional, isto é, com maior carga parasitária e/ou muco de melhor qualidade, a aproximarem-se para serem inspecionados. Para tal, os limpadores dão massagens e limpam clientes de menor valor, mostrando assim uma boa qualidade de serviço aos clientes de maior valor, que são cativados a recorrerem a esses limpadores para uma qualidade de serviço idêntica. No entanto, quando se aproximam para serem inspeccionados, os limpadores são desonestos e mordem o muco destes clientes. Para se certificarem que não voltam a ser enganados, os clientes punem os limpadores de formas diferentes. Os clientes com acesso a mais estações de limpeza fogem imediatamente ao comportamento desonesto, escolhendo outra estação de limpeza para a sua próxima inspecção, enquanto que os clientes residentes perseguem activamente o limpador desonesto. Para estas interacções funcionarem correctamente, o limpador não pode ser sempre desonesto, mesmo que a desonestidade lhe traga benefícios imediatos maiores. A desonestidade faz com que os clientes não queiram voltar à mesma estação, e sem um bom fluxo de clientes, os limpadores não comem o suficiente. Para tal, têm de alternar entre desonestidade e cooperação de forma precisa, para maximizarem os seus ganhos a longo termo. Além do mais, estes limpadores agem de forma diferente com diferentes tipos de clientes, e clientes com os quais tenham tido experiências passadas. Apesar da ecologia comportamental destes mutualismos de limpeza estar já bem estudada, os mecanismos causais destes estão ainda mal percebidos. Sabe-se que o neurotransmissor Dopamina (DA) tem efeitos modulatórios sobre o comportamento social e o comportamento em geral em vertebrados. É também amplamente conhecido o seu envolvimento na capacidade de decisão, discernimento, memória, aprendizagem, plasticidade comportamental e percepção do meio ambiente e das suas mudanças. Todas estas capacidades são necessárias à manutenção dos mutualismos de limpeza e sem elas os limpadores não conseguem ajustar o seu comportamento às diferentes situações com que se deparam. Assim, o objectivo do meu trabalho é perceber qual o papel da DA na modulação do comportamento cooperativo e na aprendizagem desta espécie altamente pro-social. A fim de testar o papel da DA na modulação do comportamento cooperativo, cinco compostos (agonista do receptor D1 – SKF-38393; antagonista do receptor D1 – SCH-23390; agonista do receptor D2 – Quinpirole; antagonista do receptor D2 – Metoclopramida; controlo – solução salina) foram injectados in situ em peixes limpadores capturados na Lizard Island, Grande Barreira de Coral, Austrália (14°40'S; 145°28'E), procedendo-se à observação do comportamento dos peixes injectados. O bloqueio da transmissão de DA por via do antagonista do receptor D1 aumentou os níveis de investimento na cooperação; o antagonista do receptor D2 teve efeitos semelhantes, mas de forma menos acentuada. Estes efeitos são contrários ao esperado, mas consistentes com a bibliografia existente: a diminuição de DA leva a alterações na capacidade de decisão e adaptação do comportamento; o intenso aumento do investimento na interacção (por via da estimulação táctil) conduz a uma alteração de percepção (défice na capacidade de decisão) por parte limpadores já que previne que estes se continuem a alimentar, procurando exclusivamente o contacto físico (estratégia que prejudica mais o limpador do que propriamente o cliente). Este aumento de investimento pode dever-se a uma excessiva insegurança durante a interacção. Uma segunda experiência, realizada em cativeiro no Oceanário de Lisboa, foi destinada a investigar o papel da DA na aprendizagem dos limpadores. Em condições controladas, o processo de aprendizagem dos limpadores pode ser explorado usando placas Plexiglas de diferentes padrões de cor com comida, substituindo os clientes. Foi simulada uma situação onde dois clientes se apresentam em simultâneo ao limpador. Nestas situações, o limpador tem de decidir qual dos clientes inspeciona primeiro, correndo o risco do cliente ignorado se ir embora. Como as espécies de clientes variam no seu valor nutricional, este aspecto foi introduzido através da apresentação de comida em apenas uma das placas (placa correcta), forçando os limpadores a aprenderem a identificar a fonte de comida. Os limpadores foram injectados diariamente com os compostos mencionados anteriormente e testados a aprender duas tarefas que diferem em termos de relevância ecológica. A primeira foi uma tarefa visual (ecologicamente relevante), onde os limpadores tinham de aprender a identificar a placa com o padrão correcto. A segunda foi uma tarefa espacial (ecologicamente não relevante), onde os limpadores tinham de aprender a identificar um lado correcto, independentemente do padrão da placa. A manipulação com o agonista do receptor D1 resultou num aumento da velocidade de aprendizagem nas duas tarefas experimentais independentemente da relevância e, ao contrário do previsto, os antagonistas não provocaram uma aprendizagem mais lenta em nenhuma das tarefas. Isto indica que a DA é essencial na aquisição de associações entre estímulos e recompensas, mas que pode estar a actuar em conjunto com outros sistemas neuronais. Os dados obtidos mostram então que a DA é essencial à aquisição, regulação e manutenção de comportamentos mutualísticos, e contribui para a compreensão das bases fisiológicas e dos mecanismos causais subjacentes aos processos cooperativos entre o peixe limpador L. dimidiatus e a sua clientela.<br>In coral reefs, cleaner species such as the cleaner wrasse Labroides dimidiatus have the ongoing duty of cleaning other reef fish, called clients. Cleaning interactions are a classical example of mutualistic interactions where both parties benefit from it: cleaners remove (eat) ectoparasites and harmful debris off the clients’ body surface, consequently contributing for the improved of reef fish health condition and overall ecosystem welfare. Although cleaning mutualisms are already widely studied from a functional point of view, the physiological mechanisms underlying these cooperative interactions are still not well understood. Dopamine (DA) is a neurotransmitter involved in the regulation of social behaviour and heavily influences decision-making. As such, DA is potentially a good candidate modulator of cleaner wrasses’ behaviour (to some degree). The present study is one step forward into understanding the exact role of DA in cooperative behaviour modulation. To do so, the cleaner wrasses’ dopaminergic system was manipulated by exogenously administering two agonists and two antagonists and recorded their behaviour in situ in Lizard Island, Australia. Cleaner wrasses’ cooperative investment increased without increasing cheating events when D1 activity was blocked, probably due to impairment of their judgment and decision-making competence. D2 blockade also increased cooperative investment but not to the same magnitude, and overall DA stimulation had no effects. Additionally, a second experiment (ex situ) was performed to test the influence of DA on cleaner wrasses’ learning competence. DA stimulation via D1 receptors increased cleaner wrasses’ visual and spatial learning, regardless of ecological relevance. The present study provides evidence that DA is involved in neuromodulation of cooperative behaviour as well as learning competence of a coral reef fish.

碩士<br>國立東華大學<br>海洋生物多樣性及演化研究所<br>100<br>Bluestreak cleaner wrasse, Labroides dimidiatus (Valenciennes, 1839) is well- known as a cleaner fish and the object of intensive marine ornamental aquarium trade in coral reef, and may became a consequence rare. Natural spawning and early ontogeny of L. dimidiatus are described and illustrated by using optical microscope, stereomicroscope and scanning electron microscopy (SEM). The present study also focuses on the subject of tempertature and salinity for the bluestreak cleaner wrasse embryonic and larvae development. Successful natural spawning of L. dimidiatus in captivity from 1 May, 2011 to 30 April, 2012 is described. Four broodstock（male: female = 1:3）laid 647900 eggs during the spawning period and the water temperature fluctuated between 24.1 and 29.2℃. The average of the fertilization rate was 96.9 %, and the hatching rate varied from 2.0 to 100.0 %. Based on the above spawning data may prove to L. dimidiatus year-round spawning fish used in this experiment. Fertilized eggs were spherical, buoyant and had a diameter of 0.65 ± 0.01 mm. Embryonic development lasted 18h 30min at 27.3 ± 0.3℃. Newly hatched larvae were 1.65 ± 0.07 mm in total length（TL）, with 29（15+14） myomeres, and melanophores evenly distributed on myomeres. An oil globule was in the posterior area of the yolk sac. At 3 days post hatch（DPH）the larvae measured 2.20 ± 0.02 mm in TL, with the mouth was developed（150.53 ± 0.02 μm in gape height）. At 4 DPH, with the oil globule and yolk being absorbed, the larvae increased in swiming ability. Under scanning electron microscope, the pores on the egg surface were uniform in size and were protrudent, with a density about 61/100 μm2. The micropyle was shape, about 6.9 μm in diameter. There were ridge substances with a sharp cicular triangle inside the micropyle. Newly hatched larvae, a neuromasts were observed on the head. The effects of temperature（14, 17, 20, 23, 26, 29, 32, 35 and 38℃）on hatching rate and hatching time were assessed. And the effect of salinity（10, 15, 20, 25, 30, 35 and 40 ppt）on hatching rate was also assessed. Results indicate significant（p＜0.05）higher hatching rate（78.9 %）in temperature 20 to 32℃, and showed the significantly（p＜0.05） longer hatching time in17℃（39.22 h）. In the salinity trail, egg showed the significantly（p＜0.05）higher hatching rate（74.4 %）at 30 and 35 ppt. These results indicate that the optimal temperature and salinity for embryonic and larval development of L. dimidiatus are 26.1℃ and 30-35 ppt, respectively.

Research Doctorate - Doctor of Philosophy (PhD)<br>In temperate New South Wales, most fish species in the family Labridae have not previously been investigated with available knowledge restricted primarily to photographic identification guides providing brief notes on species distribution, habitat preferences and identification. This information is inadequate for assessing the impact of labrid harvesting on rocky reef systems and for making informed management decisions for the protection of these fishes. Therefore, this study aimed to fill some of the significant gaps in the understanding of labrid assemblages associated with rocky reefs of temperate eastern Australia. This was accomplished by concentrating primarily on three species - Ophthalmolepis lineolatus, Notolabrus gymnogenis and Pictilabrus laticlavius - which are abundant and co-occur in shallow waters (less than 20 m depth) on the central coast of New South Wales. The methods used in this study included SCUBA surveys of labrid assemblages; in situ observations of labrid behaviour on SCUBA; and acquisition of labrid specimens for the extraction of intestines, gonads and otoliths, and for measurements of fish weight and length. <br /> Labrids were found to be the most species rich family in the study region and were the most abundant of all non-planktivorous fishes. Overall, a higher number of labrid species and a higher number of labrid individuals occurred in sponge garden habitat (15-22 m depth) compared to fringe (3-7 m) and barrens (8-15 m), owing to greater densities of O. lineolatus, Austrolabrus gymnogenis and Eupetrichthyes angustipes. The common labrids, N. gymnogenis, Achoerodus viridis and P. laticlavius, occurred at higher densities in fringe habitat due mostly to a higher representation of juveniles in this habitat. The effect of habitat on labrid assemblages was subject to small-scale variation between sites (separated by hundreds of metres) and experienced temporal changes due primarily to a substantial increase in the abundance of recruits coinciding with late summer and autumn (April-May). Behavioural observations revealed that the three focal species differed substantially in their spatial structure. O. lineolatus were found to be temporary reef residents using home ranges in excess of 2500 m2 for periods of up to 1 year before permanently emigrating outside these temporary home ranges. In contrast, N. gymnogenis exhibit strong site fidelity to reef patches of less than 600 m2 in which they remain for periods in excess of 2 years. Reef patches are shared by up to at least 10 juvenile and female individuals and a single, highly territorial male in a mating system suggestive of resource defence polygyny. An understanding of the spatial structure of P. laticlavius was constrained by its cryptic behaviour, but behavioural observations suggest this species is home ranging and establishes temporary territories for the purpose of feeding and/or reproduction. Intensive ethological observations allowed for the description and quantifying of several major behaviours in which all species typically engaged including encounters and interactions with other fishes, lying, use of shelter, side-swiping, bending, gaping, cleaning by clingfishes (Gobiesocidae) and colour change. The occurrence of these behaviours often demonstrated substantial differences among species (e.g. lying, shelter and bending) and/or experienced shifts with ontogeny (e.g. interactions and area usage). These trends generally remained consistent at different times of the day and periods of the year, and at both locations. Dietary analyses revealed O. lineolatus, N. gymnogenis and P. laticlavius are generalist carnivores feeding on a variety of benthic invertebrates including polychaetes, amphipods, decapods, gastropods, bivalves, polyplacophorans, echinoderms and cirripedes. Differences in the volumetric contribution of prey items in the guts of each species showed that food resources are partitioned among species and observations of foraging behaviour demonstrated a partitioning of microhabitats used for feeding. Ontogenetic shifts in diet and feeding microhabitats demonstrate that food resources are further partitioned within a species. However, overall morphological and behavioural similarities within a species results in greater competition occurring among individuals of the same species than among individuals of different species. This was reflected in higher rates of intra-specific interactions compared with interactions between labrid individuals of different species. Observations of feeding episodes revealed the bite rates of all species were typically unaffected by the time of day and period of year in which sampling occurred, but a location effect occurred for O. lineolatus and P. laticlavius. A reduction in bite rate with ontogeny occurred for N. gymnogenis. The population structure of the three species suggests each exhibits the typical labrid reproductive strategy of protogynous hermaphroditism. O. lineolatus and N. gymnogenis are both monandrous species, but the occurrence of some P. laticlavius males at small sizes and young ages suggests this species may be diandrous. Similarities occurred between O. lineolatus and N. gymnogenis in the size/age at which individuals sexually matured (c.a. 180 mm, 2 years) and changed sex (c.a. 280 mm, 4.6 years), but these events occurred at substantially smaller sizes (95 and 138 mm, respectively) and younger ages (les than 0.9 and 1.9 years, respectively) in P. laticlavius. Sectioned otoliths were used to determine that the longevity of O. lineolatus, N. gymnogenis and P. laticlavius was at least 13.4, 9.6 and 4.8 years, respectively. Ages were validated using marginal increment analysis. Timing of reproduction in each species was asynchronous with peaks in the reproductive activity occurring in late summer to early autumn (February-March) for O. lineolatus, mid winter (July) for N. gymnogenis and mid spring to early summer (October-December) in P. laticlavius.

Research Doctorate - Doctor of Philosophy (PhD)<br>In temperate New South Wales, most fish species in the family Labridae have not previously been investigated with available knowledge restricted primarily to photographic identification guides providing brief notes on species distribution, habitat preferences and identification. This information is inadequate for assessing the impact of labrid harvesting on rocky reef systems and for making informed management decisions for the protection of these fishes. Therefore, this study aimed to fill some of the significant gaps in the understanding of labrid assemblages associated with rocky reefs of temperate eastern Australia. This was accomplished by concentrating primarily on three species - Ophthalmolepis lineolatus, Notolabrus gymnogenis and Pictilabrus laticlavius - which are abundant and co-occur in shallow waters (less than 20 m depth) on the central coast of New South Wales. The methods used in this study included SCUBA surveys of labrid assemblages; in situ observations of labrid behaviour on SCUBA; and acquisition of labrid specimens for the extraction of intestines, gonads and otoliths, and for measurements of fish weight and length. <br /> Labrids were found to be the most species rich family in the study region and were the most abundant of all non-planktivorous fishes. Overall, a higher number of labrid species and a higher number of labrid individuals occurred in sponge garden habitat (15-22 m depth) compared to fringe (3-7 m) and barrens (8-15 m), owing to greater densities of O. lineolatus, Austrolabrus gymnogenis and Eupetrichthyes angustipes. The common labrids, N. gymnogenis, Achoerodus viridis and P. laticlavius, occurred at higher densities in fringe habitat due mostly to a higher representation of juveniles in this habitat. The effect of habitat on labrid assemblages was subject to small-scale variation between sites (separated by hundreds of metres) and experienced temporal changes due primarily to a substantial increase in the abundance of recruits coinciding with late summer and autumn (April-May). Behavioural observations revealed that the three focal species differed substantially in their spatial structure. O. lineolatus were found to be temporary reef residents using home ranges in excess of 2500 m2 for periods of up to 1 year before permanently emigrating outside these temporary home ranges. In contrast, N. gymnogenis exhibit strong site fidelity to reef patches of less than 600 m2 in which they remain for periods in excess of 2 years. Reef patches are shared by up to at least 10 juvenile and female individuals and a single, highly territorial male in a mating system suggestive of resource defence polygyny. An understanding of the spatial structure of P. laticlavius was constrained by its cryptic behaviour, but behavioural observations suggest this species is home ranging and establishes temporary territories for the purpose of feeding and/or reproduction. Intensive ethological observations allowed for the description and quantifying of several major behaviours in which all species typically engaged including encounters and interactions with other fishes, lying, use of shelter, side-swiping, bending, gaping, cleaning by clingfishes (Gobiesocidae) and colour change. The occurrence of these behaviours often demonstrated substantial differences among species (e.g. lying, shelter and bending) and/or experienced shifts with ontogeny (e.g. interactions and area usage). These trends generally remained consistent at different times of the day and periods of the year, and at both locations. Dietary analyses revealed O. lineolatus, N. gymnogenis and P. laticlavius are generalist carnivores feeding on a variety of benthic invertebrates including polychaetes, amphipods, decapods, gastropods, bivalves, polyplacophorans, echinoderms and cirripedes. Differences in the volumetric contribution of prey items in the guts of each species showed that food resources are partitioned among species and observations of foraging behaviour demonstrated a partitioning of microhabitats used for feeding. Ontogenetic shifts in diet and feeding microhabitats demonstrate that food resources are further partitioned within a species. However, overall morphological and behavioural similarities within a species results in greater competition occurring among individuals of the same species than among individuals of different species. This was reflected in higher rates of intra-specific interactions compared with interactions between labrid individuals of different species. Observations of feeding episodes revealed the bite rates of all species were typically unaffected by the time of day and period of year in which sampling occurred, but a location effect occurred for O. lineolatus and P. laticlavius. A reduction in bite rate with ontogeny occurred for N. gymnogenis. The population structure of the three species suggests each exhibits the typical labrid reproductive strategy of protogynous hermaphroditism. O. lineolatus and N. gymnogenis are both monandrous species, but the occurrence of some P. laticlavius males at small sizes and young ages suggests this species may be diandrous. Similarities occurred between O. lineolatus and N. gymnogenis in the size/age at which individuals sexually matured (c.a. 180 mm, 2 years) and changed sex (c.a. 280 mm, 4.6 years), but these events occurred at substantially smaller sizes (95 and 138 mm, respectively) and younger ages (les than 0.9 and 1.9 years, respectively) in P. laticlavius. Sectioned otoliths were used to determine that the longevity of O. lineolatus, N. gymnogenis and P. laticlavius was at least 13.4, 9.6 and 4.8 years, respectively. Ages were validated using marginal increment analysis. Timing of reproduction in each species was asynchronous with peaks in the reproductive activity occurring in late summer to early autumn (February-March) for O. lineolatus, mid winter (July) for N. gymnogenis and mid spring to early summer (October-December) in P. laticlavius.

The study reviewed the techniques used in the global trade of live fish. Based on this review and field and laboratory experiments, methods were identified to improve the survival of bluethroat wrasse, Notolabrus tetricus, horseshoe leatherjacket, Meuschenia hippocrepis and greenback flounder, Rhombosolea tapirina, during capture, holding and transport. Investigation of the physiological responses of fish exposed to lowered temperatures was a central theme including description of cold thermal tolerance and the effects of lowered temperature on oxygen consumption. The Critical Thermal Minima and Incipient Lethal Temperature were determined for bluethroat wrasse and greenback flounder. Greenback flounder exhibited remarkable thermal tolerance with a TL 50(24hr) of 2.3°C (acclimation temperature 15 °C). A flow-through respirometer was used to measure oxygen consumption. Bluethroat wrasse exhibited 20% greater oxygen consumption than greenback flounder under normoxic conditions. In both species, a reduction in temperature of 10°C more than halved the oxygen consumption as predicted by typical biological Q1o values. Greenback flounder maintained or increased oxygen consumption during graded hypoxia (i.e. a non-conformer) and exhibited a critical oxygen tension of less than 25% oxygen in air-saturated seawater. Oxygen consumption of bluethroat wrasse decreased during graded hypoxia (i.e. a conformer) but no significant increase in blood lactate concentration was observed. Lowered temperatures induced and maintained coma in horseshoe leatherjackets and bluethroat wrasse, however mean coma-inducing temperatures were generally <0.5°C higher than the TL50. The induction of coma in bluethroat wrasse did not significantly reduce oxygen consumption beyond the level predicted by temperature alone and did not result in a significant increase in blood lactate. The risk of exceeding thermal tolerance when applying coma-inducing temperatures needs to be carefully assessed before use. Feeding history had a marked impact on thermal tolerance. Greenback flounder deprived of food for 72hr exhibited greater than 90% survival when exposed to temperatures of 3°C whereas fish deprived of food for only 24hr exhibited less than 10% survival at the same low temperature. The magnitude of specific dynamic action (SDA) in greenback flounder was 1.88 times higher than the routine rate of oxygen consumption and total duration was 52hr at 15°C. Oxygen extraction by bluethroat wrasse was significantly reduced during lowered temperature and hypoxia. The ventilation rate and ventilation stroke volume of bluethroat wrasse generally decreased with lowered temperature and afferent oxygen tension. The study demonstrated that food deprivation and temperature reduction are powerful tools to improve the survival of fish during transport.

Tese de doutoramento, Ciências do Mar, da Terra e do Ambiente, Faculdade de Ciências e Tecnologia, Universidade do Algarve, 2015<br>Interspecific cleaning interactions are a classical textbook example of mutualistic cooperation. One of the most notorious cleaning mutualisms involve the Indo-Pacific Bluestreak cleaner wrasse Labroides dimidiatus, which are known to interact with an average of 2300 visitor reef fish per day. In contrast to the increasing knowledge on the functional aspects of cleaning mutualisms in the last decades, their underlying physiological mechanisms are still relatively rare. One major class of neuromodulators that is involved in the control of social behaviour and that seems to be co-opted for the regulation of cleaning behaviour is a group of nonapeptides of the arginine vasopressin /oxytocin family (AVP/OT). The general aim of this study is to attempt to link functional aspects of decision-making underlying the cleaning behaviour with their proximate mechanisms, namely to determine the contribution of the neuropeptides arginine vasotocin (AVT) and isotocin (IT) for the regulation of adjustments of individual cleaner wrasses’ behavioural output. This study yielded several findings. First, I tested the influence of AVT upon the cleaners’ ability to solve two different problems that in principle differ in ecological relevance and are associated with two different memory circuits and found that AVT affected the learning competence of cleaners as individual performance showed distinct response selectivity to AVT dosage levels. However, only in the ecologically relevant task was their learning response improved by blocking AVT via treatment with the antagonist Manning compound. Next I examined if neuropeptides may be implicated in the mechanisms underlying the adjustment of individuals to the existence of partner control mechanisms in cooperative interactions between unrelated individuals and discovered that solely the experimental transient higher dosage of AVT led to a decrease of cleaners’ willingness to feed against their preference, while IT and AVT antagonists had no significant effects. Then I asked if the establishment of privileged ties and the quality of association between cleaner wrasse pairs is correlated with neuroendocrine mechanisms involving forebrain neuropeptides and whether these neuropeptides level shifts relate to individual’s interspecific service quality. Here I found .that variation in pairs’ relationship influence male and female cleaner fish differently and contribute to the variation of brain neuropeptide levels, which is linked to distinct cooperative outcomes. Finally, I explore the link between these neuroendocrine pathways and the expression of mutualistic behaviour in fishes by comparing the brain quantitative distribution of AVT and IT across the overall and in selected areas of the brain; aiming at four closely related species of labrids that differ in the degree to which they depend on cleaning. The levels of both AVT and IT varied significantly across species, as measured in the whole brain or in specific macro-areas. More importantly, significantly higher AVT levels in cerebellum and in the whole brain were found in the obligate cleaner species, which seems to be related to expression of mutualistic behaviour. Overall, my study suggests that the neuropeptidergic system but mostly AVT pathways play a pivotal role in the regulation of interspecific cooperative behaviour and conspecific social behaviour among stabilized pairs of cleaner wrasses.<br>As interacções interespecíficas de limpeza são exemplos clássicos de cooperação mutualística. Um dos mais emblemáticos exemplos de mutualismo de limpeza envolve o bodião limpador do Indo-Pacífico Labroides dimidiatus que interage em média com cerca de 2300 espécies de peixes do recife de coral por dia. Estes bodiões removem ectoparasitas e tecido infectado ou morto da superfície dos peixes que os visitam (referidos como clientes) e dependem exclusivamente das interações mutualísticas de limpeza para obter o seu alimento. Os bodiões limpadores encontram-se em territórios específicos, designados por estações de limpeza, e vivem num sistema de harém (espécie poliginica e protogínica). As interacções de limpeza são geralmente iniciada pelos clientes, que adoptam uma posição específica ficando imoveis e abrindo a boca, opérculos e barbatanas peitorais, que demostram assim que desejam ser limpos. Por sua vez, os limpadores exibem um comportamento específico de dança, nadando para cima e para baixo, chamando assim a atenção dos seus clientes. Uma estratégia de pré-conflito utilizada pelos limpadores é a da estimulação táctil, durante a qual, usando as barbatanas pélvicas e peitorais, “massajam” o corpo do cliente durante a inspecção, e enquanto o faz não se alimenta. Apesar do crescente conhecimento sobre os aspectos funcionais do mutualismo de limpeza, pouco se sabe sobre os mecanismos fisiológicos que estão na base destas interações de cooperação. O grupo dos nonapeptidos da família da Arginina vasopressina/ Oxytocina (AVP/OT) é uma classe importante de neuromodeladores que está relacionada com a regulação do comportamento social e que por conseguinte poderá estar envolvida na modelação do comportamento cooperativo de limpeza. Um estudo recente realizado no campo demonstrou que a administração exógena de AVT no bodião limpador contribui para a diminuição as interações de limpeza, sem afectar da mesma forma as suas relações conspecíficas destes limpadores. O principal objectivo deste estudo é relacionar os aspectos funcionais das tomadas de decisão que determinam o comportamento de limpeza com os seus mecanismos proximais, isto é, tentar perceber de que forma os neuropéptidos, a Arginina vasotocina (AVT) e a Isotocina (IT) (homólogos nos peixes da AVP e OT) contribuem para a regulação da flexibilidade das respostas comportamentais dos bodiões limpadores. Em primeiro lugar, testei em laboratório a influência da AVT sobre a capacidade do bodião limpador resolver duas tarefas sociais que diferiam na sua relevância ecológica, e que estão associadas a dois circuitos de memória diferentes. Verifiquei que a AVT influenciou a capacidade de aprendizagem do limpador e o seu comportamento revelou ser sensível a diferentes dosagens deste neuropéptido. Na tarefa de aprendizagem por “pista” - a tarefa ecologicamente relevante - o limpador melhorou a sua aprendizagem quando a AVT foi bloqueada através da administração de um composto antagonista – o composto Manning. Já na aprendizagem espacial, tarefa ecologicamente não relevante para o contexto de limpeza, apenas o tratamento com uma dose baixa de AVT diminui a capacidade de aprendizagem do bodião limpador. Seguidamente, foi examinado, também em condições controladas, se os neuropeptidos estariam implicados nos mecanismos que ajustam o controlo exercido pelos parceiros sociais envolvidos neste tipo de interação interspecífica. Os bodiões limpadores preferem o muco (com alto valor energético) aos ectoparasitas mas a fim de assegurarem relações duradoiras com os seus clientes precisam comer contra a sua preferência, controlando a impulsividade de ingerirem apenas o que preferem. Verificou-se que apenas uma dose elevada de AVT levou os bodiões limpadores a comerem menos vezes contra a sua preferência. Os antagonistas da AVT e da IT não revelaram ter resultados significativos. Depois foi estudado (em condições naturais) se o estabelecimento de ligações privilegiadas e a qualidade da associação dos casais de bodiões limpadores estariam relacionados com os mecanismos neuro endócrinos que envolvem os neuropétidos no forebrain (área cerebral associada ao comportamento social) e ainda se os níveis de neuropéptidos nesta área cerebral influenciariam a qualidade do serviço de limpeza prestado quando limpavam sozinhos. Ficou demonstrado que a variação do índex de associação dos casais de limpadores influencia machos e fêmeas de forma distinta e contribui para a variação dos níveis de neuropeptidos no cérebro, que por sua vez leva a respostas comportamentais diferentes. Os machos L. dimidiatus que estabeleciam pares com maior índice de associação tinham também valores de IT e eram mais desonestos (comiam a favor da sua preferência) com maior frequência. De igual modo, as taxas de “batota” das fêmeas pareceram relacionar-se com os seus níveis de IT no forebrain, mas menos dependentes do índice de associação. Fêmeas com valores mais elevados de IT e AVT demonstraram ser mais desonestas, mas contrariamente aos machos, as fêmeas apresentaram mais “batoteiras” pertenciam a casais onde existiam maior instabilidade. Por último analisei a ligação entre as vias neuroendocrinas e a expressão do comportamento mutualístico ao comparar os níveis de AVT e IT no cérebro (como um todo e nas diferentes áreas cerebrais que o compõem) em quatro espécies de labrídeos próximos filogeneticamente mas que diferem na expressão do comportamento de limpeza (L. dimidiatus e Labroides bicolor – limpadores obrigatórios, Labropsis australis limpador facultativo e um não limpador - Labricthys unilineatus). Verificou-se que os níveis de AVT e IT variavam significativamente entre espécies quando medidos como um todo e em cada uma das macroáreas cerebrais (cerebelo, tecto óptico, tronco cerebral e forebrain). Os limpadores obrigatórios, L. dimidiatus e Labroides bicolor, apresentaram níveis de AVT significativamente maiores no cerebelo e no cérebro total, o que pode estar relacionado com a expressão do comportamento mutualístico. Os níveis de IT não parecem estar relacionados directamente com a expressão do comportamento mutualístico mas os valores elevados de IT no L. unilineatus sugerem que estes nonapéptidos poderão estar relacionados com o desenvolvimento do dimorfismo sexual, que apenas existe na espécie de não limpador (todas as outras espécies são monomórficas). Em resumo, o meu estudo sugere que o sistema neuropeptidergico, em especial as vias neuronais de AVT desempenham um papel crucial na regulação do comportamento cooperativo interespecífico e comportamento social conspecifico entre casais estáveis de bodiões limpadores.<br>SFRH/BD/41683/2007