M. E. Platell

Partitioning of food resources amongst 18 abundant benthic carnivorous fish species in marine waters on the lower west coast of Australia

The volumetric contributions made by prey and plant material to the diets of 4 elasmobranch and 14 teleost species, collected seasonally by trawling from waters along ca. 200 km of the lower west coast of Australia, have been compared. These benthic carnivores, which were all abundant and collectively contributed 83% to the total number of fish caught, represented nine families (Urolophidae, Scorpaenidae, Triglidae, Platycephalidae, Sillaginidae, Carangidae, Gerreidae, Mullidae and Pempherididae). Some species were numerous in both shallow (5-15 m) and deeper (20-35 m) waters and in both northern and southern regions, whereas others were largely confined to one of these water depths or regions. Comparisons between the diets of the different species, which utilised data collected from individuals throughout the study area, demonstrated that the dietary composition of any given species was almost invariably significantly different from that of every other species. This partly reflected the fact that, while errant polychaetes, gammarid amphipods and tanaids were ingested by all species, their contributions to the diets of the different species varied. Furthermore, echinoderms contributed to the diets of just nine species, and this was substantial only in the case of two sillaginid species, while teleosts were never consumed by six species and only made a marked contribution to the diets of the single species of platycephalid. The diet of each species underwent size-related changes, reflecting a shift from the consumption by smaller fish of prey such as amphipods, mysids and copepods, to the ingestion by larger fish of prey such as polychaetes, carid decapods, isopods and small teleosts. The interspecific and intraspecific differences in dietary compositions would spread the food resources amongst and within species, thereby reducing the potential for competition for those resources within the fish community. Non-metric multi-dimensional scaling (MDS) ordination plots emphasised that the dietary compositions of species within each family possessed some obvious similarities, reflecting similarities in body and mouth morphology and feeding behaviour. However, the extent to which the dietary compositions of the different families were similar or different was often not related to the phylogenetic relationships amongst those families. Furthermore, while differences in mouth size and morphology could sometimes be used to account for differences amongst the diets of the full suite of species, this was not always the case. Thus, the ways in which species feed and use their ancillary feeding structures were also employed to help elucidate the basis for variations in diets.

Comparisons between the influence of habitat type, season and body size on the dietary compositions of fish species in nearshore marine waters

Fish were collected from over bare sand in nearshore shallow waters at three sites that varied in the extent to which they were exposed to wave activity and which were located within a 45-km stretch on the lower west coast of Australia. Sampling was undertaken within a 4-6-week period in each season. The volumetric contributions of different prey to the stomach contents and the mouth characteristics of four species, i.e. Sillago bassensis and Sillago vittata (Sillaginidae), Spratelloides robustus (Clupeidae) and Pseudorhombus jenynsii (Bothidae), were determined. Overall, the dietary compositions of the four species differed significantly from each other and those of fish at both the three different sites and in four consecutive seasons were also significantly different. In comparison with S. bassensis, the morphologically similar S. vittata fed to a relatively greater extent on polychaetes than zooplankters, presumably reflecting in part its greater ability to extend its upper jaw downwards towards the benthos. S. robustus typically targeted calanoid copepods in the plankton, whereas P jenynsii fed on larger benthic prey taxa, reflecting the large differences in mouth morphology and feeding behaviour of these species. Although the diets of S. bassensis and S. vittata were strongly influenced by habitat type and season, the former variable was slightly more important for both of these species. However, the reverse applied with S. robustus. Season strongly influenced the dietary composition of P. jenynsii at the one site at which it was regularly caught. The diets of the two Sillago species and P. jenynsii underwent pronounced size-related changes, which would help distribute the food resources among the individuals of the different size classes of each of these species. In contrast, all size classes of S robustus fed predominantly or exclusively on calanoid copepods at all sites and in all seasons, except at the most sheltered site in winter when these zooplankters were not found in samples taken from the water column. The seasonal and habitat variations recorded in the diets of the fish species in this study imply that these species are able to feed opportunistically, a characteristic that would be of particular value to fish that live in nearshore waters where the relative abundance of the different prey types varies with habitat type and season.

Factors influencing the partitioning of food resources among six fish species in a large embayment with juxtaposing bare sand and seagrass habitats

Six abundant fish species were collected from a large embayment during both day and night and at bimonthly intervals for a year. Gerres subfasciatus and Upeneus tragula occurred mainly over bare sand, while Psammoperca waigiensis, Centrogenys vaigiensis and Apogon victoriae lived mainly in seagrass (Amphibolis antarctica) and A. rueppellii often migrated from seagrass to over bare sand at night. All species except U. tragula fed at night, as well as during the day. The overall dietary compositions of the six species were significantly different from each other, even when the species occurred in the same habitat. G. subfasciatus and U. tragula consumed greater volumes of errant polychaetes, which could be readily targeted in a substrate that does not contain dense rhizome mats. In contrast, P. waigiensis, C. vaigiensis and A. victoriae ingested greater volumes of carid decapods, which are particularly abundant in seagrass and, thus, within the water column. Furthermore, when A. rueppellii moved at night from seagrass to over sand, the consumption of carid decapods declined, whereas that of polychaetes and particularly mysids, which are very abundant over bare sand, increased. However, the composition of the prey consumed by different species within the same habitat also often varied markedly. For example, unlike U. tragula, G. subfasciatus ingested not only sedentary polychaetes, but also considerable volumes of errant polychaetes, reflecting its ability to use a combination of vision and its highly protrusible mouth to target prey both on and just below the substrate surface. Differences in the types and range of prey ingested by the six species could often be related to differences in the overall size, width and/or protrusibility of the mouth. Furthermore, dietary breadth was greatest in species with the largest mouth dimensions. The diets of three species underwent diel changes that could be related to differences in foraging mode and/or prey availability. Thus, a use of vision to detect prey would account for the greater consumption during the day of copepods by G. subfasciatus and of small teleosts by A. rueppellii, while the nocturnal emergence of amphipods and/or tanaids from the substrate explains their greater ingestion by G. subfasciatus, A. victoriae and A. rueppellii at night. Although the smaller individuals of each species consumed larger volumes of prey, such as copepods and mysids, and the larger fish ingested greater volumes of prey, such as decapods and teleosts, the extent of the size-related changes in diet varied markedly amongst species.

To what extent are the dietary compositions of three abundant, co‐occurring labrid species different and related to latitude, habitat, body size and season?

This study demonstrated that the dietary composition of each of three abundant reef-associated labrid species in temperate Western Australia differed significantly with latitude and changed with increasing body size and almost invariably differed among those species when they co-occurred. These results were derived from comparisons and multivariate analyses of volumetric dietary data, obtained from the foregut contents of Coris auricularis, Notolabrus parilus and Ophthalmolepis lineolatus from the Jurien Bay Marine Park (JBMP) and waters off Perth, 250 km to the south. Latitudinal differences in the dietary compositions of each species in exposed reefs typically reflected greater contributions by large crustaceans, bivalve molluscs, echinoids and annelids to the diets in the waters off Perth than in the JBMP, whereas the reverse was true for gastropods and small crustaceans. The diet of each species exhibited similar, but not identical, quantitative changes with increasing body size, with the contributions of small crustaceans declining and those of large crustaceans and echinoids increasing, while that of gastropods underwent little change. Within the JBMP, the dietary compositions of both C. auricularis and N. parilus were similar in exposed and sheltered reefs and the same was true for N. parilus in the sheltered reefs and interspersed areas of seagrass. The latter similarity demonstrated that, in both of those divergent habitat types, N. parilus feeds on prey associated with either the sand or the macrophytes that cover and lie between the reefs. Although the main dietary components of each species were the same, i.e. gastropods, small crustaceans (mainly amphipods and isopods), large crustaceans (particularly penaeids and brachyuran crabs) and echinoids, their contributions varied among those species, which accounts for the significant interspecific differences in diet. Coris auricularis had the most distinct diet, due mainly to an ingestion of greater volumes of small crustaceans, e.g. amphipods and isopods, and lesser volumes of large crustaceans, e.g. brachyuran crabs, which was associated with a relatively narrower mouth and smaller teeth and the absence of prominent canines at the rear of the jaw. The above intra and interspecific differences in dietary composition would reduce, on the south-west coast of Australia, the potential for competition for food among and within these three abundant labrids, each of which belongs to different genera within the Julidine clade.

Dietary composition of the blue swimmer crab Portunus pelagicus L. Does it vary with body size and shell state and between estuaries

The aim of this study was to determine the diets of Portunus pelagicus in the large Peel-Harvey and Leschenault estuaries in south-western Australia in order to ascertain whether the dietary composition of this crab changes with body size during two different moult stages and differs between the two estuaries. Portunus pelagicus, ranging in age and carapace width from ca. 2 months and 12 mm to ca. 112 years and 159 mm, were collected from the shallow basins of the Peel-Harvey and Leschenault estuaries. Examination of the cardiac stomachs of these crabs showed that P. pelagicus does not feed just before or immediately after moulting and that the stomachs of recently-moulted crabs contained significantly more food than those of intermoult crabs. Although the volumetric contribution made by calcareous material to the stomach contents was similarly high in all size classes of recently-moulted crabs, i.e. 47 to 55%, the volumetric contributions made by small bivalves decreased with body size, whereas the reverse occurred with shell fragments of large decapods and, to a lesser extent, polychaetes. The dietary compositions of intermoult crabs were shown by classification and multi-dimensional scaling ordination to differ markedly from those of recently-moulted crabs and to undergo similar progressive ontogenetic changes in both the Peel-Harvey and Leschenault estuaries. Thus, the contribution made by small benthic and epibenthic crustaceans, such as amphipods and tanaids, declined with increasing body size, whereas the reverse occurred with larger prey, such as nereid polychaetes, small decapods and teleosts. The dietary composition of P. pelagicus was influenced more by moult stage, i.e. recently moulted vs. intermoult, than by body size. Although the dietary compositions of P. pelagicus in the two estuaries were not significantly different, they did differ from those recorded from coastal marine waters in the same region, thereby reflecting differences in the potential prey in those two environments.

The diets of two co-occurring marine teleosts, Parequula melbournensis and Pseudocaranx wrighti, and their relationships to body size and mouth morphology, and the season and location of capture

The dietary compositions of the co-occurring gerreid Parequula melbournensis and the carangid Pseudocaranx wrighti have been determined, using samples collected seasonally from a 200 km stretch of coastal water in temperate Australia, in which these species are very abundant. Although the small representatives of P. melbournensis and P. wrighti both fed to a large extent on copepods, the latter ingested a wider variety of prey and thus had a greater dietary breadth. The diets of both species changed markedly as body size increased. The larger representatives of P. melbournensis fed mainly on onuphid and other polychaetes, whereas those of P. wrighti consumed a considerable volume of crustaceans, molluscs, polychaetes other than onuphids, and echinoderms. The above differences account for the dietary breadth being far greater in large P. wrighti than large P. melbournensis. Schoeners index and classification showed that dietary overlap between P. melbournensis and P. wrighti was low, suggesting that food resources were partitioned between these two demersal species. Intraspecific overlap was less amongst the length classes of small fish than among those of larger fish, indicating that any competition for food would be less among small fish, when growth would have been most rapid. The larger P. melbournensis fed mainly on prey types which were relatively sessile and can protrude from the substrate, such as tube-dwelling onuphid polychaetes, whereas the larger P. wrighti fed on a variety of epibenthic and/or more mobile species, such as mysids, amphipods, bivalves, gastropods, nereid polychaetes and echinoderms. Although P. melbournensis foraged to a far greater extent on a sessile fauna that occurs on or in the substrate, it ingested a far smaller amount of sand than P. wrighti, even though this latter species fed on a more epibenthic fauna. The presence of smaller amounts of sand in the stomach contents of P. melbournensis than of P. wrighti, presumably reflects the possession of a smaller and far more protrusible mouth, which enabled its sessile prey to be targetted more precisely. The dietary composition of P. melbournensis and P. wrighti underwent some seasonal changes, presumably reflecting seasonal fluctuations in the densities of prey species, and that of P. melbournensis differed slightly between some sites.

Influence of water depth, season, habitat and estuary location on the macrobenthic fauna of a seasonally closed estuary

The macrobenthic invertebrates in the shallow (<0·5 m), medium-depth (0·75-1·0 m) and deep (2·0-2·5 m) waters of the lower and upper regions of the large basin of the seasonally closed Wilson Inlet were sampled, using corers, in each season between the winters of 1989 and 1990. The fauna was dominated by capitellid, spionid and nereid polychaetes and mytilid, venerid and tellinid molluscs. Almost 75% of the total numbers were contributed by the polychaetes Heteromastus filiformis (Claparede) (30·2%), Capitella capitata (Fabricius) (17·4%), Prionospio cirrifera (Wiren) (7·9%) and Boccardia chilensis (Blake & Woodwick) (5·4%), and the gastropod Hydrococcus brazieri (Tenison Woods) (12·4%). The numbers of species recorded in Wilson Inlet (41) were relatively very low for an estuary. This presumably reflects (i) the low salinities, which would restrict colonization by stenohaline marine species, (ii) the prevention of recruitment from the sea by the closure of the estuary mouth during the summer, at which time many macrobenthic species are producing pelagic larvae, (iii) the very limited tidal movements, and thus mechanisms for transporting larvae from the sea, when the estuary mouth is open, and (iv) the lack of a pronounced salinity gradient and the small variation in sediment composition along the estuary. Characteristics (ii-iv), together with the dispersal of pelagic larvae within the basin by wind mixing, would also account for the fact that faunal composition was influenced less by the region within the basin than by water depth. The densities of species, such as Ceratonereis aequisetis (Augener), Capitella capitata and H. brazieri, were greater in the shallow and medium depths than in deeper water, while the reverse pertained with P. cirrifera. Since densities of the first three species are correlated with the biomass of Ruppia megacarpa Mason, the greater amounts of this aquatic macrophyte in the shallow and medium-depth waters may thereby provide more food and/or protection from predators in those waters than would be the case in deep water. In contrast, P. cirrifera, which is a detritivore, would benefit from the greater proportion of particulate organic material in the deep waters. The greater habitat complexity in the shallow and medium-depths, that is provided by the greater amounts of R. megacarpa and shell debris in those waters, was associated with a greater species richness, diversity, density and biomass of benthic invertebrates. The composition of the fauna underwent cyclical changes, which were related, in part, to the patterns of recruitment of certain major taxa following spawning.

Partitioning of food resources by four abundant, co-occurring elasmobranch species: relationships between diet and both body size and season

The present study has tested statistically the hypothesis that the diets of four abundant and co-occurring elasmobranch species differ and change with body size and season and has determined the extent of any differences, to ascertain their potential for reducing competition for food resources. Non-metric multivariate analyses of volumetric contributions of dietary categories to stomach contents demonstrated that the dietary compositions of the rays Myliobatis australis and Aptychotrema vincentiana and the sharks Heterodontus portusjacksoni and Squatina australis in south-western Australian waters differed. M. australis fed predominantly on benthic invertebrates, whereas A. vincentiana consumed large volumes of teleosts. The durophagous H. portusjacksoni ingested a wide variety of prey, including gastropods, cephalopods, bivalves, echinoderms and teleosts, whereas S. australis ingested mainly teleosts and cephalopods. The extent and pattern of change in the diet with increasing body size varied among species. For example, the diet of H. portusjacksoni changed abruptly to larger, harder-bodied prey at ∼400-mm total length, whereas that of S. australis underwent small, gradual changes with increasing body size. The diets of each species changed seasonally. Inter- and intraspecific variations in dietary composition reduce the potential for competition between and within these abundant elasmobranch species in south-western Australian waters.

Influence of selected factors on the dietary compositions of three targeted and co-occurring temperate species of reef fishes: implications for food partitioning.

The dietary compositions of three medium to large targeted fish species, which co-occur over reefs in temperate waters of south-western Australia, were determined. These data were then used to ascertain statistically the extent to which body size, season and habitat influence the diets of these species and the degree to which food resources were partitioned among and within those species, and thus reduced the potential of interspecific and intraspecific competition. On the west coast, Bodianus frenchii (Labridae) and Epinephelides armatus (Serranidae) spent their whole life over prominent limestone reefs, as did Glaucosoma hebraicum (Glaucosomatidae) in all but juvenile life, when it lived over low-relief, limestone substrata. The dietary composition of each species changed with increasing body size, which, in G. hebraicum, was particularly pronounced at c. 300 mm total length (L(T)) and therefore at the size when this species shifts habitat. When the three species co-occurred over the same reefs, their dietary compositions were significantly different, with that of B. frenchii being by far the most discrete, reflecting a far greater contribution by sedentary taxa. Thus, the diet of B. frenchii was distinguished from those of the other two species in containing substantial volumes of bivalve and gastropod molluscs and echinoid echinoderms and essentially no teleosts. Although the diets of G. hebraicum and particularly E. armatus were dominated by teleosts, and especially for larger individuals, the former species ingested greater volumes of cephalopods and small crustaceans. The pointed jaws of B. frenchii, with their forwardly directed and interlocking anterior incisors, are ideally adapted for biting and retaining their invertebrate prey, which are attached to or reside within reef crevices. In contrast, the mouths of G. hebraicum and E. armatus are broader and rounder and contain numerous small, slender and inward-pointing teeth. These teeth, in conjunction with prominent backward-curved canines in E. armatus, facilitate the capture and retention of fish prey. Observations in situ indicate that G. hebraicum is a suction feeder, while E. armatus is predominantly a ram feeder. Although reef environments on the west and south coasts differ, the diet of B. frenchii on these coasts differed only slightly. Interspecific differences in diet, combined with size-related changes in dietary compositions and the occupation of different habitats by juvenile and adult G. hebraicum, reduce the potential for competition for food resources among and within B. frenchii, G. hebraicum and E. armatus and thus helps facilitate the coexistence of these species which historically have been abundant over reefs in south-western Australia.

DISTRIBUTIONS, SIZE COMPOSITIONS AND DIETS OF TWO ABUNDANT BENTHIC AMBUSH-FEEDING TELEOSTS IN COASTAL WATERS OF SOUTH-WESTERN AUSTRALIA

The scorpaenid Maxillicosta scabriceps (Teleostei) and the platycephalid Platycephalus longispinis (Teleostei) were trawled seasonally for a year from shallow (5-15 m) and deeper (20-35 m) waters in four distantly-located regions of the inner continental shelf of south-western Australia. The former species was more abundant in southern regions, which is consistent with its temperate distribution, while the latter species was more abundant in deep than shallow waters. Unlike M. scabriceps, P. longispinis with total lengths < 110 mm were rarely caught, indicating that this latter species only moves on to the sandy substrate of the inner shelf when it reaches a certain size. As M. scabriceps increased in size, the contributions of mysids, amphipods and oxyrhyncan crabs to the diet declined, while those of carid decapods and brachyrhyncan crabs increased. The invertebrate fauna ingested by the smallest length-class of P. longispinis, 110-139 mm, was similar to that of the same and largest length-class of M. scabriceps. However, unlike M. scabriceps, this length-class of P. longispinis consumed teleosts, despite having a relatively smaller mouth. The contribution of teleosts subsequently increased as P. longispinis increased in size, a feature reflected by the relatively low dietary breadth of large fish. The marked shift in the types of prey ingested by P. longispinis as it increased in size accounted for the fact that, in contrast to M. scabriceps, there was limited dietary overlap between the larger and smaller members of this platycephalid. The ability of P. longispinis to ingest other fish is presumably related to the fact that platycephalids lie just under the substrate surface and are thus concealed from their potential prey, and that they are able to emerge rapidly and pursue their prey. Unlike several other local demersal fish species, M. scabriceps and P. longispinis ingested only a small amount of sediment, implying that these ambush predators target their prey very precisely, a foraging mode that would be facilitated by the epibenthic and conspicuous nature of their prey. The absence of marked differences in the dietary compositions of these two species in different regions, water depths and seasons is consistent with a specific mode of foraging by these species.