Influence of habitat on abundance and size structure of a large temperate-reef fish, Achoerodus viridis (Pisces: Labridae)

The relationship between densities of Achoerodus viridis (Pisces: Labridae) and reef habitats at various localities within New South Wales (NSW), Australia was examined. Types of habitats were quantified at inner, mid and outer estuarine locations in each of two estuaries (Botany Bay and Port Jackson) to determine whether purported patterns of movement from estuaries could be related to differences in habitat. Although the same types of habitat were generally found at all locations, differences in the proportion of habitat types were found between shallow and deep reefs and among inner, mid and outer estuarine locations for both estuaries. Shallow habitats were usually dominated by Ecklonia radiata, turf and/or fringe habitat in Botany Bay, whereas deep sites were generally dominated by urchin-grazed barrens habitat and, sometimes, sponge- and ascidian-dominated deep reef. Shallow sites in Port Jackson were dominated by a mixture of habitats, as were deep reefs at mid-estuarine locations. Other deep reefs in Port Jackson were dominated by E. radiata (inner estuarine) or barrens (outer estuarine) habitat. Thus, patterns of habitat cover were not consistent between estuaries and numbers of fish could not be related to proportional representation of habitat on reefs along estuarine gradients. Univariate and multivariate analyses showed that there was little evidence that any size class of fish was correlated with the proportional representation of a particular habitat or group of habitats. Counts of fish that focused on barrens and E. radiata forest habitats over a period of 10 yr showed that similar numbers and all sizes of fish were found in the two types of habitat. Greater numbers of small fish were, however, found in the E. radiata forest habitat than in the barrens habitat. Estimates of abundance along the coast of NSW (100s to 1000 km) in a range of habitats (e.g. ascidian-dominated reef, kelp forest, urchin-grazed barrens) showed that there was no indication that a particular habitat consistently had greater numbers of A. viridis than other habitats. Therefore, A. viridis of a range of sizes appears to be flexible in its use of habitats on reefs. Received: 24 December 1997 / Accepted: 23 June 1998     

Differential movement patterns and site fidelity among trophic groups of reef fishes in a Hawaiian marine protected area

We tracked the long-term movements of 70 parrotfishes, surgeonfishes and goatfishes captured inside a small (1.3 km²) marine protected area (MPA: Kealakekua Bay Marine Life Conservation District, Hawaii) by implanting them with small transmitters and deploying underwater monitoring devices inside the bay and along 100 km of the adjacent west Hawaii coastline. Individual fish were detected inside Kealakekua Bay for up to 612 days but many were detected for much shorter periods (median = 52 days). There were species-specific differences in the scale of movements and habitats used, but most fish utilized between 0.2 and 1.6 km of coastline, and individuals of each species showed some degree of diel habitat shift. A wide variety of reef fishes captured inside the MPA swam back and forth across an MPA boundary intersecting continuous reef (i.e., this boundary was porous to reef fish movements), but only 1 of 11 species tagged crossed a wide sandy channel inside Kealakekua Bay suggesting that this feature may function as a natural barrier to movements. Results indicate relatively small MPAs (<2 km of coastline) could provide effective, long-term protection for multi-species assemblages of reef fishes provided that boundaries are situated along major habitat breaks (e.g., large sand channels between reefs) that may serve as natural barriers to reef fish movements. It is crucial that a multi-species approach be used when assessing MPA effectiveness.     

Inter-specific and geographical variations in the fatty acid composition of mangrove leaves: implications for using fatty acids as a taxonomic tool and tracers of organic matter

Fatty acid compositions of the leaves of six species of mangroves were studied to ascertain their use as biomarkers for determining the fate of mangrove organic matter in the habitat and as taxonomic tool. Mangrove leaves were collected from three locations in the western Pacific Ocean: Moreton Bay (MB) (Australia), Hong Kong (China) and Okinawa (Japan). In MB, samples were collected from two sites separated by 15 km: Logan River Estuary (LRE) and Jabiru Island. In addition, along the LRE, leaves were collected from five stations at ∼2–3 km apart. Results show that the analysis of the entire fatty acid profiles of the mangrove leaves is a promising taxonomic tool as the profiles of most species were sufficiently different to be separated in an non-metric multidimensional scaling plot. In addition, geographically separated populations of the same species could also be identified by their fatty acid profiles. In most cases, two non-ubiquitous groups of fatty acids dominated in the mangrove leaves: the polyunsaturated 18:2ω6 and 18:3ω3 and the long chain fatty acids (≥24:0). With respect to the relative contributions of these fatty acids, three groups of species were identified, in which one or both groups of fatty acids may potentially be used as markers of the mangrove organic matter in the estuarine environment.     

Migration and stock studies on the Australian school prawn Metapenaeus macleayi

The results of tagging studies conducted to determine the relationship between estuarine (juvenile) populations and adults at sea suggest that maturing Metapenaeus macleayi leave the estuaries and move along the coast in a northerly direction; the longest migration recorded was 120 km, but most prawns appear to disperse in shallow water (<40 m) within about 70 km of their juvenile habitat. There are approximately 60 estuarine populations of M. macleayi in south-east Australia; the geographical range of prawns emigrating from most estuarine populations overlap (at sea) to some degree with that of adjacent or nearby populations, but there is little or no overlap between any two of the 6 major populations. The mixing of individuals from a major population and from nearby smaller populations is negligible for the purposes of fisheries management, hence the prawns in each of the major estuarine populations and the adults in the coastal area north for about 70 km may be regarded as a unit stock.     

Effect of limpets on the fouling of ships in the Mediterranean

Limpets reduce the cover of algae and barnacles on intertidal rocks, but do not inhabit ship-fouling communities. Patella caerulea (L.) from Israeli Mediterranean shores were successfully transplanted onto 20x20 cm initially non-fouled steel panels in 1983–1984, hung vertically in a commercial port at depths of 1 and 5.5 m. The limpets (15 individuals of ca 20 to 25 cm in length on each panel) controlled the fouling organisms that settled on the panels throughout a month of lowest settlement (December) and throughout a month of highest settlement (August). After 7 d of immersion the limpets restricted the total cover of fouling organisms from 2.4 and 4.8% to 0.6 and 0.2% (at 1 and 5.5 m depths, respectively), and brought down the barnacle density from 1.2 and 2.6 to 0.3 and 0.4 individuals per cm² (at 1 and 5.5 m depth, respectively). After 26 d the equivalent figures were a cover of 72 and 92% reduced to 14 and 9% and a density of 2.9 and 1.1 reduced to 0.3 and 0.2 individuals per cm². The limpets presumably achieved this by preventing the attachment of propagules and by removing already settled organisms through their movements during grazing excursions.     

High levels of genetic subdivision of marine and estuarine populations of the estuarine catfishCnidoglanis macrocephalus (Plotosidae) in southwestern Australia

The cobblerCnidoglanis macrocephalus (Valenciennes) is an endemic marine and estuarine catfish from southern Australia. Conflicting views on the degree of isolation of the estuarine populations underscore general questions about genetic divergence in coastal species. Although estuaries are widely recognized as ecologically important, little work has been done on their role in favouring genetic divergence. In order to estimate the extent of genetic subdivision among nearshore marine and estuarine populations, electrophoretic variation of enzymes was examined in seven marine and six estuarine populations of cobbler from sites spanning 1500 km along the southwest Australian coastline. Among all populations, the mean standardized variance in allelic frequencies (F _ST) for six polymorphic loci was 0.277, a high value comparable to those of other shallow-water teleosts whose life-history characteristics and habitat preferences restrict their dispersal capability. The pattern of genetic identities between populations showed divergence between west and south coast sites. Within these regional groups, however, there was substantial heterogeneity, much of which was associated with estuaries. Among all six estuarine sites, the averageF _ST was 0.333, 40% higher than the value of 0.237 for the marine sites. Low estimates of the genetically effective number of migrants suggest population subdivision between marine and estuarine environments and between similar habitat types. This study indicates the importance of habitat in affecting the connectedness of populations, even in apparently open marine systems.     

Primary cell-culture of the digestive gland of the marine mussel Mytilus edulis: a time-course study of antioxidant- and biotransformation-enzyme activity and ultrastructural changes

The extent to which the American lobster, Homarus americanus (H. Milne-Edwards), utilizes estuarine habitats is poorly understood. From 1989 to 1991 we examined lobster movements in and around the Great Bay estuary, New Hampshire using tag/recapture and ultrasonic telemetry. A total of 1212 lobsters were tagged and recaptured at sites ranging from the middle of Great Bay, 23.0 km from the coast, to Isles of Shoals, 11.2 km offshore. Twenty-six lobsters equipped with ultrasonic transmitters were tracked for periods ranging from 2 weeks to >1 year. Most lobsters moved <5 km toward the coast, with those furthest inland moving the greatest distance. Lobsters with transmitters moved in a sporadic fashion, with residency in one area for 2 to 4 weeks alternating with rapid movement to a new location (mean velocity = 0.3 km d⁻¹, 1.8 km d⁻¹ max.). Site of release influenced distance moved, but there was no significant relationship between lobster size and distance traveled, days at large, or rate of movement. Most movement into the estuary occurred in the spring, while during the remainder of the year there was a strong tendency to move downriver, toward the coast. These seasonal migrations of estuarine lobsters may enhance their growth and survival by enabling them to avoid low salinity events in the spring and fall, and to accelerate their growth in warmer estuarine waters during the summer. Received: 26 January 1996 / Accepted: 22 January 1999     

Genetic variability in bloodworm (Glycera dibranchiata) populations in the Gulf of Maine

Genetic structure of eight Northwest Atlantic populations of the marine polychaeteGlycera dibranchiata Ehlers was examined with starch gel electrophoresis. Samples were collected during summer and fall 1981, and seven polymorphic and four monomorphic loci were consistently scored. Average heterozygosity (0.126) and percent polymorphic loci (59.3) were comparable to the averages reported for marine invertebrates. Minimum genetic distances between populations ranged from 0.003 to 0.093, levels typically associated with local populations of the same species in other taxa. Based on these data, inter- and intra-estuarine migration and gene flow appear to be low. Only two populations, separated by 13 km along the same river in New Brunswick, Canada, were not genetically different from each other. These findings may have relevance for management strategies in bloodworms.     

Animal-sediment relations and community analysis of a Florida estuary

Five species assemblages of the intertidal infauna of Old Tampa Bay, Florida, USA are identified. Two assemblages are judged to constitute distinct communities, while a third is shown to be an interdigitation of the two communities. Dominance by on species is the prevalent pattern within the assemblages. Numbers of deposit feeders are found to be inversely correlated to that of filter feeders, and both trophic types are found to be correlated to the sediment parameters of median grain size, sorting and skewness. Three transects with three stations each were established along the south side of Courtney Campbell Causeway in Tampa Bay, Florida. A faunal sample (0.4 m²), a sediment sample, and a water sample were taken at each station in September, December, and March of 1968/1969. Sediment samples were wet-seived. Animal samples were reduced to numbers of organisms and biomass per species. Trellis diagrams and correlation tests were generated. Support is shown for the trophic group-amensalims hypothesis, however, the silt-clay fraction is apparently of lesser importance to deposit feeders in Florida sediments than in Buzzards Bay sediments. An attempt is made to relate an analysis of the optimal grain size for filter feeders to the trophic group-amensalism hypothesis. A view of communities as abstractions from continua is more realistic than communities as discrete units.     

Using stable isotopes to monitor anthropogenic nitrogen inputs to estuaries

Use of stable nitrogen isotope ratios is one method that has been proposed to indicate anthropogenic nutrient enrichment in estuarine systems. However, the role of stable isotopes as a tool in long-term ecosystem monitoring has not been fully developed. Resident producer and consumer species were collected from marshes dominated by Spartina alterniflora and subject to a range of anthropogenic impacts in Cape Cod, Massachusetts, and in Great South Bay and Jamaica Bay, New York. Tissue isotope ratios of Spartina alterniflora, Ulva lactuca, Fundulus heteroclitus, and Geukensia demissa were analyzed in order to determine which organisms are the most sensitive indicators of changes in anthropogenic nitrogen source and loading. Power analysis was used to determine the sample sizes necessary to detect change in nutrient source using the species sampled. Relationships between the delta15N values of the species sampled and watershed population density and residential development were evaluated. Population density was a better indicator of anthropogenic nitrogen impact than residential development, since most anthropogenic nitrogen in the study marshes was derived from wastewater. Consumer species demonstrated lower within-site variability than producer species and would therefore require smaller sample sizes to detect changes in nitrogen source and loading.     

Dispersal and Use of Corridors by Birds in Wooded Patches on an Agricultural Landscape

Dispersal behavior has important effects on the persistence and recolonization of populations, but is one of the least understood traits of most organisms. Knowledge of patterns of fledgling, natal, and breeding dispersal of birds in a patchy environment will assist in decisions regarding reserve design and protection or construction of corridors. I present data on movement patterns of three migratory bird species, American Robin ( Turdus migratorius ), Brown Thrasher ( Toxostoma rufum ), and Loggerhead Shrike ( Lanius ludovicianus ). These birds are relatively common breeders in south-central North Dakota (U.S.) in riparian woodlands and in shelterbelts (woodlots planted as windbreaks in the open agricultural environment). Field assistants and I individually marked and monitored the movements of more than 500 adults breeding in a network of shelterbelts across an 8 × 11 km study area. Most movement occurred at relatively short distances within a shelterbelt. Movements by adults between shelterbelt sites, although rare, occurred significantly more frequently between sites connected by a wooded corridor than between unconnected sites. For robins, there were on average 2.5 dispersal events between each pair of connected sites, but only 0.17 dispersal events between each pair of unconnected sites (Mann-Whitney test, significant at p < 0.009). Because unconnected and connected sites were similar in average area (1.7 to 1.9 ha), distance to next wooded habitat, and tree-species composition, this result provides a test of the hypothesis that organisms disperse preferentially along connecting corridors.     

Reproductive cycles in two populations of the Pacific sand dollar Dendraster excentricus

East Pacific swimming crabs are poorly known ecologically. We trawled samples along 4 estuarine-continental shelf transects to provide information on the population structure and species composition of portunids along environmental stress gradients. Portunids comprised almost 99% of the benthic crabs. Callinectes toxotes dominates the freshest estuarine areas, but is replaced by C. arcuatus in estuary mouth and shallow-shelf samples. Deeper shelf samples are dominated by Euphylax robustus and Portunus sper. Great numbers of E. dovii, a pelagic species, were taken on the shelf; their size distribution implies intraspecific competition. Evidence is given that differences in salinity tolerance account for differential landward limits, but competition, predation by fishes, increased fouling and decreased available food exclude eurytopic species from physicochemically equable shelf waters.     

Long-term variation of organochlorine residues and assemblages of epibenthic organisms in a shallow North Florida (USA) estuary

A 4-year study (1972–1976) determined long-term trends of organochlorine residues (DDT, DDE, DDD, PCB's, mirex) and trawl-susceptible organisms in a shallow, river-dominated estuary in North Florida (Apalachicola Bay, USA). Moderate levels of such compounds were found in various species prior to the restricted use of DDT in 1972. A subsequent precipitous decline in organochlorine besidues was attributed to decreased upland usage, major flushing of the river basin in early 1973, and various factors associated with estuarine function. No mirex was found in sediments or aquatic organisms. Apparently, the half-life of organochlorines is relatively short in this bay system. Various statistical methods were used to test the relationships of different physico-chemical and biological parameters. During the 4-year study period, seasonal river flow fluctuations dominated water color, turbidity, salinity, nutrients (NO₃), chlorophyll a, and the temporal succession of fishes in the bay. Certain long-term trends of fish associations were noted; relative dominance of key fish species declined and stabilized while bay-wide species richness and diversity increased with time. Qualitative changes in species representation determined the long-term pattern of community variability. This was consonant with a distinctive fish fauna during the first year of sampling. The bay anchovy Anchoa mitchilli was dominant during 9 of the first 12 monts of the project; this influenced the time-related changes in community indices. Temporally clustered fish associations reflected the importance of river flow in the estuarine environment. Direct correlation of fish distribution with the rapid disappearance of organochlorine compounds was complicated by aperiodic natural phenomena such as storms and river fluctuations. Population and community trends appeared consistent with other studies showing similar patterns of dominance of stress-resistant fish populations and related changes in community parameters. In this case, the relatively predictable annual succession of fish associations allowed an appraisal of key forcing functions. Due to the high level of seasonal and annual biological variability in this estuary, there were some problems in the application of linear statistical models to the data base. Although the long-term trend of relative species representation is useful as an index of stress, new techniques are needed to analyze extensive field data so that functions such as trophic interactions are included in the estimation of causal relationships. There are indications that such effects could be related to the impact of organochlorine compounds on estuarine systems.     

The spread of marine non-indigenous species via recreational boating: A conceptual model for risk assessment based on fault tree analysis

Recreational vessel movements are increasingly recognised as an important pathway for the spread of non-indigenous species (NIS) in marine environments. Research on risks posed by recreational vessels has focused on external hull fouling, yet a number of studies reveal the potential for NIS to also be transferred by a range of other vessel components. This paper uses fault tree analysis as a framework for incorporating input from a panel of international experts, to elucidate the consecutive steps that must occur for NIS to be introduced from different components of recreational boats. Our conceptual model reveals the complexity of the invasion process even when only the ‘release’ phase is considered (i.e. the release of NIS from an infected vessel into a new area). The model highlights that, in addition to external fouling of the ‘hull’ (hull, rudder and propeller), important vessel components may also include fouling, sediment or water released from the deck, internal spaces, anchors and fishing/diving gear. The extent to which these components are important is situation-specific, and depends on attributes of the vessel, location and NIS present. Hence, the comprehensive model described here could be modified or simplified to reflect the attributes that are relevant to particular circumstances. We demonstrate this principle using examples of three NIS: the colonial tunicate Didemnum vexillum and the Asian kelp Undaria pinnatifida that both have established in Port Nelson New Zealand after vessel-mediated spread, and the clubbed tunicate Styela clava that was detected on a vessel hull in the port but is not known to have established. Although the modelling and assessment of some of the events identified in the fault trees would be difficult or unrealistic, it is important to acknowledge them in order to provide a comprehensive risk assessment tool. Even where risks are largely unknown, difficult to quantify, or reflect stochastic events, this does not necessarily preclude management intervention.     

Macroinfaunal structure and effects of thermal discharges in a mesohaline habitat of Chesapeake Bay,near a Nuclear Power Plant

The soft-bottom macroinfauna of the Chesapeake Bay near the Calvert Cliffs Nuclear Power Plant, Maryland, USA, was studied quarterly in 1977. A total of 42,900 organisms distributed among 55 taxa was taken in 180 Smith-McIntyre grabs. Two communities and an ecotone between these two communities were identified in association with three sediment types: sand, sand-clay, and clay. Reproductive activities of 8 numerically dominant species are described. The macroinfauna at Calvert Cliffs was largely seasonal; species richness and organism abundances were lowest in late summer. The polychaetes Scolecolepides viridis and Heteromastus filiformis and the bivalves Macoma balthica and Mya arenaria were dominant in March-June. In September-December, the polychaetes Neanthes succinea, Paraprionospio pinnata, and Glycinde solitaria and the bivalve Gemma gemma became dominant. There were clear indications of differences associated with sediment types in total number of species, total number of individuals, and abundances of the dominant species. The shallower and intermediate habitats had higher sand-clay ratios, lower organic carbon contents, and larger numbers of organisms than the deeper clay habitat. Within each depth, no apparent detrimental effects caused by the thermal discharges on total number of species, total number of individuals, abundances of the dominant species, species diversity or evenness were found. Increased abundance was observed at the discharge stations for a few species, especially for Mya arenaria and N. succinea.     

Effects of grazer richness and composition on algal biomass in a closed and open marine system

Most natural local systems exchange organisms with a regional pool of species through migration and dispersal. Such metacommunity processes of interconnected multispecies assemblages are likely to affect local dynamics of both species and processes. We present results from an artificial marine outdoor rock pool system in which we investigated the factors of (1) local grazer richness and composition, and (2) connectivity of local patches to a regional species pool, and their effects on algal biomass. Local species richness of six grazers was manipulated in both open and closed pools, which were embedded in a regional species pool containing all six grazers. Grazer richness showed significant net biodiversity effects on grazing in the closed, but not in the open, system. Grazer composition, on the other hand, showed significant effects on grazing in both open and closed systems, depending on which species were initially present. The two most efficient grazers were able to compensate for less efficient grazers in species mixtures, hence ensuring the function of grazing. The efficiency of top-down control of algal biomass in open systems thus depends on which particular species are lost. Further, differences in grazing between the open and closed system changed over time due to temporal dynamics in grazer composition. The results emphasize the importance of including system connectivity in experimental designs to allow an extrapolation of biodiversity ecosystem-functioning relationships to natural systems.     

Biology of marine fouling in Mangalore waters

A comprehensive study of the fouling organisms encountered at the New Mangalore Port at Panambur and the Fish Landing Jetty on the Nethravati-Gurpur estuary near the old Mangalore Port, India, was made. Barnacles, oysters, bryozoans, polychaetes and hydroids are the dominant groups of foulers encountered in this region. Barnacles are by far the most important foulers. Distinct seasonal abundance in settlement was noticed in the different groups. The major peak settlement of barnacles occurs in November and December. Continuous settlement of this group in the harbour indicates the presence of larvae throughout the year in the marine locality. The absence of freshly settled spat in the estuarine locality shows that these larvae do not foul structures in the Fish Landing Jetty during the monsson period. The pattern of oyster settlement indicates that the spat of Crassostrea madrasensis and c. cucullata settle during January, March, May and October in this region. This group remains in the fouling community throughout the year. Only typical marine bryozoans are harboured in abundance in the fouling community. Polychaetes occur in great numbers during April and December.The differences in the quality and quantity of the fouling communities are controlled by their breeding periodicity,seasonal fluctuations in the hydrographical conditions, and the species composition of the sedentary community of this region.     

Benthic invertebrate assemblages of Delaware Bay

During two consecutive summers, the first quantitative bay-wide survey (207 stations) of benthic invertebrates was conducted in Delaware Bay (USA). In 1972, 109 species were collected at 105 stations; and in 1973, 125 species were collected at 102 stations. A total of 169 different species were collected for both summers. The number of species and number of individuals increased with increasing salinity and increasing median grain size. These relationships were compared and were found similar to those in estuaries and bays throughout the world. Average density was 722 individuals m², which is low compared to other estuaries. The relationship of low secondary production to pollution, macroscopic algae, sediment transport, predation, and hydrography is discussed. Deposit feeders comparised the major feeding type. Local species composition was similar to that in Chesapeake Bay, and dominant species occurred in estuaries throughout the Mid-Atlantic Bight. The benthic invertebrates of Delaware Bay were related to the cosmopolitan mode of estuarine faunas. Faunal assemblages were identified by cluster analysis. The assemblages were associated with sediment type and salinity. It was concluded that Delaware Bay comprises a mosaic of animal assemblages, some of which have relatively sharp boundaries similar to classical level bottom type communities, whereas the boundaries of others are almost impossible to detect, and these represent species distributed along an environmental continum.     

Bioaccumulation of heavy metals in flora and fauna of Hooghly estuary,east coast of India

Spectroscopic determinations of Fe, Zn, Mn, Cu, Co and Ni were carried out in 23 organisms including seaweeds, zooplankters, molluscs and fishes from Hooghly estuary at the confluence of Bay of Bengal. Seaweeds and gastropods were more efficient in accumulating levels while zooplankters bivalves and fishes showed intermediate levels. Analyses were performed on the sediment, water and on the individual dissected organs of a mussel, oyster and teleost fish. Bioaccumulation of Fe, Zn, Mn and Cu in the soft body parts (gills, mantle, liver etc.) of the three species showed a high degree of organ specificity in some cases. Major sources of the micropollutants include natural weathering, catchment runoff, urban and industrial discharges that may pose an ecological risk to the local estuarine ecosystem.     

Comparison of larval fish assemblages in three large estuarine systems, KwaZulu-Natal, South Africa

 The structure of the larval fish assemblages in three large estuarine systems on the KwaZulu-Natal coast of South Africa was examined using a combination of univariate, distributional and multivariate techniques. The database was comprised of a full annual set of larval fish samples taken from each estuarine system: Durban Harbour, Richards Bay Harbour and St Lucia Estuary. The mean monthly densities of each species in each system were used in the species matrix, and the mean monthly values of salinity, temperature and turbidity were used in the environmental matrix. The mean species diversity and evenness index were significantly higher in Durban Harbour (H′ = 1.03, J′ = 0.65) than in the other two systems. The cumulative dominance curve showed that St Lucia Estuary has a high dominance of a few species, with Richards Bay Harbour intermediate and Durban Harbour being the most diverse. Classification and MDS (multiple-dimensional scaling) analyses of larval fish densities in all three systems grouped together into three main clusters on the basis of system. The species similarity matrix (inverse analysis) clustered into five groups at the 25% similarity level. The MDS analysis of the same matrix showed that the groups separated out according to the degree of estuarine association of a species and hence habitat type. The species most responsible for system groupings were: Glossogobius callidus, Gilchristella aestuaria, Stolephorus holodon, Croilia mossambica and Gobiid 12. The “best fitting” of the environmental variables to explain the larval fish community patterns in each system was turbidity on its own (weighted Spearman's rank correlation, ρ_w = 0.55). The relationship of larval densities to environmental conditions was shown to be species-specific with estuarine species (e.g. G. callidus and G. aestuaria), having a strong positive correlation with temperature and turbidity but negative correlations with salinity. In summary, much longer term studies with more sites within each system are needed to assess whether the larval fish assemblages are stable or at an equilibrium (both spatially and temporally) and whether these assemblages are indicative of the relative “health” and nursery function of the system. Received: 28 April 1999 / Accepted: 30 May 2000