Population structure and feeding deterrence in three shallow-water antarctic soft corals

Alcyonium paessleri and Clavularia frankliniana are numerically abundant soft corals in the nearshore (12 to 33 m depth) benthic communities of eastern McMurdo Sound. They are much less abundant in western McMurdo Sound where a third species, Gersemia antarctica, co-occurs in low numbers. The body tissues of these three species are comprised mainly of organic material (53 to 70% dry wt), which is primarily dervied from NaOH-soluble protein and refractory material. The energetic contents of the whole-body tissues of A. paessleri, C. frankliniana and G. antarctica are 15.9, 17.3, and 14.5 kJ g^-1 dry wt, respectively. The mean biomass per individual is 1.81, 0.008, and 45 g dry wt for each respective species. Based on population densities of 7.3, 1337.3, and 0.04 soft corals m^-2 for A. paessleri, C. frankliniana and G. antarctica, respectively, the population energetic densities are estimated to be 210.1, 185.1, and 26.1 kJ m^-2. Despite the relatively rich energetic content of the tissue and apparent vulnerability to predators, very little predation occurs on these soft corals. Two potential predators, the antarctic sea stars Perknaster fuscus and Odontaster validus, exhibited significant chemotactic defensive tube-foot retractions to hexane, chloroform, methanol, and aqueous methanol extracts of each soft coral. In addition, wholebody tissue of each soft coral was rejected by the demersal fish Pseudotrematomus bernacchii and the cryopelagic fish Pagothenia borchgrevinki. In contrast, whole soft-coral tissues sequentially extracted in four increasingly polar solvents were readily ingested by these antarctic fishes, indicating that sclerites do not play a significant role in deterring predators. Our results indicate that these antarctic soft corals contain bioactive compounds which deter common predatory seastars and fishes.     

Investigation of the relationship between invertebrate predation and biochemical composition,energy content,spicule armament and toxicity of benthic sponges at McMurdo Sound,Antarctica

The biochemical and energetic composition, spicule content, and toxicity of benthic sponges was investigated in McMurdo Sound, Antarctica from October through December 1984. The predominant organic constituent of sponges was soluble and insoluble protein. Levels of total protein ranged from 17.0 to 55.9% dr. wt. Levels of lipid and carbohydrate were low, ranging from 2.1 to 9.6 and 0.6 to 3.5% dr. wt, respectively. Levels of ash were high and variable (32 t0 79% dr. wt), reflecting species-specific differences in spicule contents. Calculated energy contents of sponges were low, with a mean of 9.8±3.5 kJ g^-1 dr. wt; ranging from 5.1 kJ g^-1 dr. wt in Sphaerotylus antarcticus to 17.4 kJ g^-1 dr. wt in Dendrilla membranosa. Insoluble protein accounted for the greatest contribution to the energetic composition of the sponges, while lipid and carbohydrate combined contributed to less than 25% of the overall energy. Normalized spicule volumes of sponges ranged from 0.15 to 0.38 cm³ g^-1 dr. wt. Ichthyotoxicity assays indicated that 9 (56%) of 16 antarctic sponge species were toxic. The most highly toxic species were Mycale acerata and Leucetta leptorhapsis. The high incidence of toxicity in antarctic sponges indicates that the current hypothesis suggesting a simple inverse relationship between toxicity and latitude in marine sponges is invalid. There was little correspondence between the energetic composition or spicule contents of the sponges and feeding patterns (electivity indices) of sponge-eating predators. Although the asteroid Perknaster fuscus antarcticus specializes on the highly toxic, fast-growing M. acerata, most antarctic sponge-eating predators appear to be generalists which feed on the more abundant, non- to mildly-toxic, sponge species. This feeding strategy is based on exploitation of low energy, sedentary prey, which require a minimal energy output to harvest.     

Habitat use by sponge-dwelling brittlestars

Cryptic organisms often associate with sessile invertebrates for refuge in space-limited environments. To examine interspecific habitat associations on coral reefs, tube- and vase-shaped sponges were surveyed for associated brittlestars at six sites on the coral reefs off Key Largo, Florida. Of 179 sponges encountered, Callyspongia vaginalis was the most abundant (43.0%), followed by Niphates digitalis (39.7%), and Callyspongia plicifera (4.5%). Three of eight sponge species surveyed did not differ from C. vaginalis in two physical refuge characteristics: oscular diameter and inner tube surface area. Brittlestars (416 total), all of the genus Ophiothrix, were only found in C. vaginalis, N. digitalis, and C. plicifera. The most abundant brittlestar, O. lineata (326), occurred on C. vaginalis (99.0%) and N. digitalis (1.0%), while O. suensonii (67) occurred on C. vaginalis (79.1%), N. digitalis (19.4%), and C. plicifera (1.5%). There was no pattern of co-occurrence of O. lineata and O. suensonii on C. vaginalis. The abundance of O. lineata increased with surface area of C. vaginalis. Differential habitat use was observed in O. lineata, with small individuals (<5 mm disk diameter) located inside and on the surface of sponge tubes and large individuals (5 mm) solely inside tubes. The number of large O. lineata in C. vaginalis never exceeded the number of tubes per sponge, and tagged O. lineata remained in the same sponge for at least 3 weeks. In density manipulations, no pattern of intraspecific competition among large O. lineata was observed; however, there was evidence for interaction between size-classes. Brittlestars selected live sponge habitat over a non-living refuge, suggesting a mechanism for sponge habitat recognition. Sponge-dwelling brittle stars prefer some tube- and vase-shaped sponge species despite similar oscular diameters and surface areas. Surprisingly, these preferred sponge species are known from previous studies to be chemically undefended against generalist fish predators; therefore, brittlestars that inhabit these sponges do not gain an associational chemical defense. Sponge habitat use by O. lineata may be governed by intraspecific interactions to maintain habitat and access to food. While past studies have suggested that O. lineata is an obligate sponge commensal, the present study suggests that O. lineata has a species-specific association with the tube-sponge C. vaginalis.Communicated by O. Kinne, Oldendorf/Luhe     

Chemical defense in the eggs and embryos of antarctic sea stars (Echinodermata)

Fifteen species of reproductively mature echinoderms (11 sea stars, 3 sea urchins, 1 sea cucumber) were collected from McMurdo Sound, Antarctica, during austral spring and summer of 1985 and 1986; eggs and embryos were obtained, and were tested for ichthyonoxicity using the common marine killifishFundulus grandis as a model predator. Chemical deterrents occurred in the large, yolky eggs of the pelagic lecithotrophic sea starPerknaster fuscus and the planktotrophic sea starPorania antarctica. Brooded embryos of the sea starsDiplasterias brucei andNotasterias armata were also noxious. Significant ichthyonoxicity was not detected in the remaining 7 species of sea stars, 3 sea urchins, and 1 sea cucumber. Chemical deterrents were generally effective at concentrations below a single egg or embryo per agar test-pellet. Although chemicals found in these eggs and embryos are noxious to an allopatric fish, they may not be noxious to sympatric fish.     

Chemical defenses, nutritional quality, and structural components in three sponge species: Ircinia felix, I. campana, and Aplysina fulva

Allocating chemical defenses to regions or tissues most at risk for predatory attack may provide protection while simultaneously minimizing associated metabolic costs. Chemical defense allocation patterns were investigated in the aspiculate sponges Ircinia felix, I. campana, and Aplysina fulva collected between July 2005 and April 2006 from J Reef off the coast of Georgia, U.S.A. It was predicted that chemical defenses would be (1) higher in the outermost 2 mm layer of the sponge; (2) positively correlated with tissue nutritional quality; and (3) correlated with structural components such as spongin fibers. Whereas defensive chemicals were concentrated in the outer 2 mm of A. fulva, the Ircinia species had higher concentrations in deeper tissue layers. Furthermore, no significant positive or negative correlation between chemical defenses and nutritional quality or levels of structural components was observed in these sponges. Overall, these results do not support the prediction that predation pressure by fish and large mobile invertebrates significantly impacts chemical defense allocation in these sponges.     

Sponge-feeding fishes of the West Indies

In an analysis of the stomach contents of 212 species of West Indian reef and inshore fishes, sponge remains were found in 21 species. In eleven of these, sponges comprised 6% or more of the stomach contents; it is assumed that these fishes feed intentionally on sponges. Sponges comprise over 95% of the food of angelfishes of the genus Holacanthus, over 70% of the food of species of the related genus Pomacanthus, and more than 85% of the food of the filefish, Cantherhines macrocerus. Lesser quantities of sponges are ingested by the remaining fish species. Fishes that feed on sponges belong to highly specialized teleost families, suggesting that this habit has evolved in geologically late time. The small number of fish species that concentrate on sponges as food suggests that the defensive characters of sponges—mineralized sclerites, noxious chemical substances, and tough fibrous components—are highly effective in discouraging predation. The two sponges most frequently eaten by fishes have a low percentage of siliceous spicules relative to organic matter, but among the 20 next most frequently consumed species no striking correlation occurs with respect to spicule content. Color and form of the sponge show no special correlation with frequency of occurrence in fish stomachs. Three species of fishes appear to concentrate on one species of sponge, but in these cases over 60% of the food taken consists of a variety of other organisms. Those fishes, more than half of whose diet consists of sponges, tend to sample a wide variety of species. No strong evidence is provided by our data that fish predation is a significant factor in limiting sponge distribution in the West Indian region.     

Cytotoxic haplosclerid sponges preferred: a field study on the diet of the dotted sea slug<Emphasis Type="Italic"> Peltodoris atromaculata</Emphasis> (Doridoidea: Nudibranchia)

Trophic specializations are widespread among opisthobranch molluscs. One purported example from the Mediterranean Sea is the dotted sea slug Peltodoris atromaculata. It has been hypothesized that this species is strongly monophagous on the sponge Petrosia ficiformis. However, the small amount of evidence that has been found for this hypothesis is based just on laboratory tests. Here we study the feeding habits and the diet of Peltodoris atromaculata in its natural habitat. We observed and videotaped 161 individuals together with the organisms on which they were found (their living substrata). Feeding scars were identified and videotaped as well. Individuals and their living substrata were sampled for further analysis in the laboratory. The composition of faeces of Peltodoris, especially undigested sponge spicules, was analyzed by light and scanning electron microscopy and compared to the composition of the living substrata. Most of the faecal samples consisted of undigested sponge spicules. Although Peltodoris was found on 11 species of sponges, only 2 of them, Petrosia spp. and Haliclona fulva, form its diet (76% out of n=121 samples). In accordance with this, feeding scars in the habitats were exclusively observed on these two sponges. Estimation of electivity indices suggests that Haliclona is preferred over Petrosia. One remarkable feature of the exclusive feeding of Peltodoris on Petrosia and Haliclona is that both sponges share specific fulvinol-like polyacetylenes that show cytotoxic activity in bioassays. Potential benefits and evolutionary aspects of this trophic specialization are discussed. Besides sponge-containing faeces, we found spicule-free faeces (24%, n= 29). These were very small in volume compared to sponge-containing faeces, and only few distinct structures were present. However, the use of food other than sponges is not necessarily indicated by this, because the spicule-free faeces might also represent left-overs from the stomach and digestive gland after sponge spicules have been released.Communicated by O. Kinne, Oldendorf/Luhe     

Evidence for morphology-induced sediment settlement prevention on the tubular sponge<Emphasis Type="Italic"> Haliclona urceolus</Emphasis>

Several mechanisms are known to assist the survival of sponges in highly sedimented environments. This study considers the potential of sponge morphology and the positioning of exhalant water jets (through the osculum) in the adaptation of Haliclona urceolus to highly sedimented habitats. This sponge is cylindrical with an apical osculum, which is common in sedimented subtidal habitats at Lough Hyne Marine Nature Reserve, Cork, Ireland. Fifteen sponges were collected, preserved (killed with the structure and morphology maintained) and then replaced in a high sediment environment next to a living specimen (at 24 m). After 5 days, the sediment settled on both living and preserved sponges was collected and dried. No sediment was collected from living sponges, while preserved specimens had considerable amounts of settled sediment on their surfaces. The amount of sediment collected on these preserved specimens was significantly linearly correlated with sponge dry weight, maximum diameter and oscula width (R²>0.70, P<0.001, df=14). Observations of flow direction (using coloured dye) through H. urceolus showed that water is drawn into the sponge on its underside and exits via a large vertically pointing osculum. Sponge morphologies (shape) have often been considered as a means of passive adaptation to a number of different environmental parameters with oscula position enabling entrained flow through the sponge in high flow conditions. However, this study shows how the combination of sponge morphology (tubular shape) and positioning of the osculum may enable H. urceolus to survive in highly sedimented environments. Similar mechanisms may also aid the survival of some deep-water sponge species with similar morphologies.Communicated by J.P. Thorpe, Port Erin     

Patterns of selective predation by juvenile,benthivorous fish on estuarine macrofauna

Benthic feeding on macrofauna was studied in juveniles of the sparids Lithognathus lithognathus and Rhabdosargus holubi in the upper reaches of the Gamtoos Estuary, South Africa. Fish and benthic macrofauna were sampled simultaneously, and the selection of invertebrate prey assessed. Both fish species strongly selected for corophioid amphipods and consumed other benthic taxa in low numbers. R. holubi also exploited aquatic autotrophs, while L. lithognathus had a narrow prey-spectrum, feeding almost exclusively on the tube-dwelling amphipod Grandidierella lignorum. G. lignorum was the most abundant prey species, both in the benthos and the fish's diet. This species also showed prominent behavioural differences between the sexes; males were markedly more active on the sediment surface than females, who rarely left their tubes during the day. Males switched from an infaunal to epifaunal microhabitat in search of receptive females, concurrently increasing their exposure to fish predators. Consequently, L. lithognathus selected significantly more males than female amphipods, causing a marked bias towards females in the sex ratio and age-structure of the amphipod population. Juvenile amphipods were less preyed upon, presumably as a result of lower prey-detection or capture efficiency by the predators. Accepting current notions about predation as an important structuring element for benthic communities, our data also stress the prominence of size-and sex-selective predation in structuring individual prey populations.     

Population biology and antipredator defenses of the shallow-water Antarctic nudibranch Tritoniella belli

The circumpolar nudibranch Tritoniella belli Eliot occurs in abundance in shallow-water benthic communities of McMurdo Sound, Antarctica. Density estimates based on belt transects averaged collectively 0.46 individuals m⁻² at three study sites between depths of 6 and 30 m in November 1996. At two of the sites, population densities increased linearly between 18 and 30 m depth (up to 0.7 and 1.15 individuals m⁻² at 30 m depth). Individuals at all sites were rare or absent at depths shallower than 12 m. Size frequencies of individuals at the sites were similar, and a pooled analysis revealed a unimodal distribution skewed highly towards juvenile size classes. This suggests both recent recruitment and constant rates of mortality across size classes. The relationship between foot length and wet weight best fits an exponential growth equation, indicative of an allometric growth pattern. Distribution of T. belli in the field suggests that it is a habitat and diet generalist. Potential invertebrate predators include sea anemones and seastars, both of which co-occur in abundance in McMurdo Sound. Laboratory experiments indicate that the sea anemone Isotealia antarctica can capture and ingest T. belli. However, 70% of T. belli that are captured escape from the tentacles or, following ingestion, are rejected from the gastrovascular cavity. The seastars Odontaster validus, Perknaster fuscus, and Acodontaster conspicuus, avoid contact with T. belli, but if forced into contact with mantle tissues, retract their tube-feet. Mucus secreted from the mantle tissues, coated on to the tips of glass rods, and presented to seastar tube-feet, causes significantly longer tube-foot retraction times than control rods. Moreover, pieces of freeze-dried krill coated with mantle mucus are consumed significantly less often than untreated control pieces of krill by a benthic scavenging fish (Pseudotrematomas bernacchi). Employing seastar tube-foot retractions as a bioassay, we found the bioactive compound(s) are soluble in ethyl acetate, indicating they are lipophilic or moderately hydrophilic in nature. Chemical defenses in the mucus of T. belli probably contribute to its high abundance in Antarctic benthic communities. Received: 6 October 1997 / Accepted: 24 March 1998     

Biochemical composition,energy content and chemical antifeedant and antifoulant defenses of the colonial Antarctic ascidian<Emphasis Type="Italic"> Distaplia cylindrica</Emphasis>

The colonial ascidian Distaplia cylindrica occurs both as scattered individual colonies or in gardens of colonies in fine-grained soft substrata below 20 m depths off Anvers Island along the Antarctic Peninsula. Individual colonies, shaped as tall rod-like cylinders and anchored in the sediments by a bulbous base, may measure up to 7 m in height. D. cylindrica represent a considerable source of materials and energy for prospective predators, as well as potential surface area for fouling organisms. Nonetheless, qualitative in situ observations provided no evidence of predation by sympatric predators such as abundant sea stars, nor obvious biofouling of colony surfaces. Mean energy content of whole-colony tissue of D. cylindrica was relatively high for an ascidian (14.7 kJ g^–1 dry wt), with most of this energy attributable to protein (12.7 kJ g^–1 dry wt). The sympatric omnivorous sea star Odontaster validus consistently rejected pieces of D. cylindrica colonies in laboratory feeding assays, while readily ingesting similarly sized alginate food pellets. Feeding deterrence was determined to be attributable to defensive chemistry, as colonies of D. cylindrica are nutritionally attractive and lack physical protection (conspicuous skeletal elements or a tough outer tunic), and O. validus display significant feeding-deterrent responses to alginate food pellets containing tissue-level concentrations of organic extracts. In addition, high acidity measured on outer colony surfaces (pH 1.5) as well as homogenized whole-colony tissues (pH 2.5) are indicative of surface sequestration of inorganic acids. Agar food pellets prepared at tissue levels of acidity resulted in significant feeding deterrence in sea stars. Thus, both inorganic acids and secondary metabolites contribute to chemical feeding defenses. D. cylindrica also possesses potent antifoulant secondary metabolites. Tissue-level concentrations of hydrophilic and lipophilic extracts caused significant mortality in a sympatric pennate diatom. Chemical feeding deterrents and antifoulants are likely to contribute to the abundance of D. cylindrica and, in turn, play a role in regulating energy transfer and community structure in benthic marine environments surrounding Antarctica.Communicated by P.W. Sammarco, Chauvin     

Oxygen dynamics and transport in the Mediterranean sponge Aplysina aerophoba

The Mediterranean sponge Aplysina aerophoba kept in aquaria or cultivation tanks can stop pumping for several hours or even days. To investigate changes in the chemical microenvironments, we measured oxygen profiles over the surface and into the tissue of pumping and non-pumping A. aerophoba specimens with Clark-type oxygen microelectrodes (tip diameters 18–30 μm). Total oxygen consumption rates of whole sponges were measured in closed chambers. These rates were used to back-calculate the oxygen distribution in a finite-element model. Combining direct measurements with calculations of diffusive flux and modeling revealed that the tissue of non-pumping sponges turns anoxic within 15 min, with the exception of a 1 mm surface layer where oxygen intrudes due to molecular diffusion over the sponge surface. Molecular diffusion is the only transport mechanism for oxygen into non-pumping sponges, which allows total oxygen consumption rates of 6–12 μmol cm⁻³ sponge day⁻¹. Sponges of different sizes had similar diffusional uptake rates, which is explained by their similar surface/volume ratios. In pumping sponges, oxygen consumption rates were between 22 and 37 μmol cm⁻³ sponge day⁻¹, and the entire tissue was oxygenated. Combining different approaches of direct oxygen measurement in living sponges with a dynamic model, we can show that tissue anoxia is a direct function of the pumping behavior. The sponge-microbe system of A. aerophoba thus has the possibility to switch actively between aerobic and anaerobic metabolism by stopping the water flow for more than 15 min. These periods of anoxia will greatly influence physiological variety and activity of the sponge microbes. Detailed knowledge about the varying chemical microenvironments in sponges will help to develop protocols to cultivate sponge-associated microbial lineages and improve our understanding of the sponge-microbe-system.     

Biodiversity and rockfish recruitment in sponge gardens and bioherms of southern British Columbia,Canada

In the Strait of Georgia and Howe Sound, British Columbia, colonies of individual cloud sponges, growing on rock (known as sponge gardens) receive resource subsidies from the high biodiversity of epifauna on adjacent rock habitats. Bioherms are reefs of glass sponges living on layers of dead sponges. In the same area as the sponge gardens, newly discovered bioherms in Howe Sound, BC (49.34.67 N, 123.16.26 W) at depths of 28- to 35-m are constructed exclusively by Aphrocallistes vastus, the cloud sponge. The sponge gardens had much higher taxon richness than the bioherms. The sponge garden had 106 species from 10 phyla, whereas the bioherm had only 15 species from 5 phyla. For recruiting juvenile rockfish (quillback, Sebastes maliger), the food subsidy of sponge gardens appears to be missing on bioherms of cloud sponge, where biodiversity is relatively low. While adult and subadult rockfishes (S. maliger, S. ruberrimus, S. proriger, and S. elongatus) were present on bioherms, no evidence for nursery recruitment of inshore rockfishes to bioherms was observed, whereas the sponge gardens supported high densities of newly recruited S. maliger, perhaps owing to the combination of both refuge and feeding opportunities. These results indicate that sponge gardens form a habitat for early stages of inshore S. maliger, whereas A. vastus bioherms are associated only with older juvenile and adult rockfishes.     

Variable effects of symbiotic snapping shrimps on their sponge hosts

Mutualistic relationships are ubiquitous in tropical coral reefs, but the costs and benefits to partner species are often poorly known. In Caribbean coral reefs, several species of snapping shrimp (Synalpheus spp.) dwell exclusively in marine sponges, which serve as both habitat and food source. A paired experimental design was used to examine the effects of Synalpheus occupancy on predation, morphology, and growth of their sponge host Lissodendoryx colombiensis in Bocas del Toro, Panama (9.351°N, 82.258°W) in June 2009. Shrimp occupancy significantly decreased consumption of sponges by a predatory sea star (Oreaster reticulatus) and also affected sponge morphology; sponges grown without shrimps decreased in canal size, in both the laboratory and the field. Shrimp occupancy had more ambiguous effects on sponge growth. In laboratory experiments, shrimp occupancy benefited sponge growth, although all sponges experienced overall decreases in mass. In field experiments, there were no significant differences in growth between occupied and empty sponges. However, the benefits of shrimp occupancy on sponge growth were negatively correlated with overall increases in sponge size; sponges that decreased in mass during the experiment benefited more from shrimp occupancy than sponges that increased in mass. These costs and benefits suggest that Synalpheus has variable effects on sponges: positive effects on sponges in the presence of predators, and/or when sponges are decreasing in mass (e.g., during periods of physical stress), but a negative effect on sponges during periods of active sponge growth.     

Global conservation status of sponges

Sponges are important for maintaining ecosystem function and integrity of marine and freshwater benthic communities worldwide. Despite this, there has been no assessment of their current global conservation status. We assessed their status, accounting for the distribution of research effort; patterns of temporal variation in sponge populations and assemblages; the number of sponges on threatened species lists; and the impact of environmental pressures. Sponge research effort has been variable; marine sponges in the northeastern Atlantic and Mediterranean and freshwater sponges in Europe and North America have received the most attention. Although sponge abundance has increased in some locations since 1990, these were typically on coral reefs, in response to declines in other benthic organisms, and restricted to a few species. Few data were available on temporal trends in freshwater sponge abundance. Despite over 8500 described sponge species, only 20 are on threatened species lists, and all are marine species from the northeastern Atlantic and Mediterranean. Of the 202 studies identified, the effects of temperature, suspended sediment, substratum loss, and microbial pathogens have been studied the most intensively for marine sponges, although responses appear to be variable. There were 20 studies examining environmental impacts on freshwater sponges, and most of these were on temperature and heavy metal contamination. We found that most sponges do not appear to be threatened globally. However, little information is available for most species and more data are needed on the impacts of anthropogenic‐related pressures. This is a critical information gap in understanding sponge conservation status. Estado Global de la Conservación de Esponjas     

Asexual reproduction in homoscleromorph sponges (Porifera; Homoscleromorpha)

Asexual reproduction by external budding in Homoscleromorpha is reported for the first time. Two Mediterranean sponge species were studied, Oscarella lobularis and O. tuberculata. Buds are formed in the marginal basal part of sponge. Budding takes from 1 to 4 days and is defined in three budding stages. First, small irregular protuberances, consisting of external parental tissue, are formed. Second, they elongate and acquire more regular, nipple-like shape. These protuberances are tube like, their internal cavity derived from parental exhalant canal. The wall consists of three layers: (a) external layer is flagellated exopinacoderm, (b) internal one is flagellated endopinacoderm and (c) intermediate one is a thin layer of mesohyl. Third, a spherical bud with a large central cavity is formed. During budding, we did not observe cell proliferation or transdifferentiation either in budding zone or in any special mitotically active region. The bud attached to the substrate is similar to the rhagon developing after larva metamorphosis, it has a syconoid organization. Morphogenetically, budding in Oscarella differes from that in other sponges. Occurring by epithelial morphogenesis, it is similar to morphallaxis during regeneration. The presence in Homoscleromorpha of an epithelial morphogenesis is unique among sponges. This feature is shared by Homoscleromorpha and Eumetazoa.     

Transmission,plasticity and the molecular identification of cyanobacterial symbionts in the Red Sea sponge <Emphasis Type="Italic">Diacarnus erythraenus</Emphasis>

Cyanobacterial symbionts in the sponge Diacarnus erythraenus from the Red Sea were identified in both adult sponges and their larvae by 16S rDNA sequencing. A single cyanobacterial type was found in all samples. This cyanobacterial type is closely related to other sponge cyanobacterial symbionts. The cyanobacterial rDNA, together with the morphological analysis by electron and fluorescence microscopy, provided evidence for vertical transmission of the symbionts in this sponge. In addition, we show phenotypic plasticity of the symbionts inside the sponge, probably as a result of variability in light availability inside the sponge tissue. Finally, the reproduction of Diacarnus erythraenus is also described.Matan Oren, Laura Steindler have contributed equally to the work.     

Coral cavity sponges depend on reef-derived food resources: stable isotope and fatty acid constraints

The diet of cavity sponges on the narrow fringing reefs of Curaçao, Caribbean was studied. The origin and resources of the bulk food of these sponges, i.e., dissolved organic matter (DOM), were identified using stable carbon and nitrogen isotopes and fatty acid biomarkers. We found that phytoplankton and its derived DOM from the adjacent open sea and from reef overlying water is not the main source of food for most of the sponges examined nor is bacterioplankton. Interestingly, dual stable isotope signatures (δ¹³C_org, δ¹⁵N_org) and fatty acid biomarkers appoint coral mucus and organic matter derived from crustose coralline algae (CCA) as probable food sources for encrusting sponges. Mucus-derived DOM may contribute up to 66% to the diet of examined sponges based on results of dual isotope mixing model analysis. The contribution of CCA (as purported representative for benthic algae) was smaller with values up to 31%. Together, mucus- and CCA-derived substrates contributed for 48–73% to the diet of sponges. The presence of the exogenous fatty acid 20:4ω6 in sponges, which is abundant in coral mucus of Madracis mirabilis and in CCA, highlights these reef-derived resources as sources of nutrition for DOM feeding cavity sponges. The relatively high concentrations of exogenous 20:4ω6 in all sponges examined supports our hypothesis that the bulk of the food of the cavity sponge community is reef-derived. Our results imply that cavity sponges play an important role in conserving food and energy produced within the reef.     

Sponge-feeding by the Caribbean starfish Oreaster reticulatus

The common Caribbean starfish Oreaster reticulatus (Linnaeus) feeds on sponges by everting its stomach onto a sponge and digesting the tissue, leaving behind the sponge skeleton. In the San Blas Islands, Republic of Panama, 54.2% of the 1549 starfish examined from February 1987 to June 1990 at eight sites were feeding, and 61.4% of these were feeding on sponges, representing 51 species. Sponges were fed on disproportionately heavily in comparison to their abundance, which was only 9.7% of available prey. In feeding choice experiments, 736 pieces of 34 species of common sponges from a variety of shallow-water habitats, and also 9 ind of a coral, were offered to starfish in individual underwater cages. Acceptance or rejection of sponge species was unambiguous for 31 of the 34 species, and there was a clear relationship between sponge acceptability and sponge habitat. Starfish ate 16 of 20 species that normally grow only on the reefs, but only 1 of 14 species that live in the seagrass meadows and rubble flats surrounding the reefs. The starfish live in the seagrass meadows and rubble flats, and avoid the reefs, and so the acceptable reef sponges are generally inaccessible until a storm fragments and transports them into starfish habitat. After Huricane Joan washed fragments of reef sponges into a seagrass meadow in October 1988, starfish consumed the edible species. When the seagrass meadow was experimentally seeded with tagged reef sponge fragments in June 1994, O. reticulatus consumed edible species and accumulated in the area seeded. Reef sponges that were living in a seagrass meadow, from which O. reticulatus had been absent for at least 4 yr (from 1978 to 1982), were eliminated when the starfish migrated into the area, and the sponges have been unable to recolonize up to June 1994. O. reticulatus feeding and habitat preferences appear to restrict distributions of many Caribbean reef sponge species to habitats without O. reticulatus and may have exerted significant selective pressure on defences of those sponges that live in O. reticulatus habitats.     

Ecological roles of natural products from the marine sponge Geodia corticostylifera

In the Brazilian coast, high numbers of the small brittle star Ophiactis savignyi usually live associated with the sponge Geodia corticostylifera (Demospongiae, Geodidae), but not with other sympatric sponge species. In order to check whether this association was related only with the physical shelter provided by the sponge body or was chemically mediated, the crude organic extract of G. corticostylifera was added to sponge mimics made of phytagel and spongin skeleton. Control and treated mimics were simultaneously offered to previously sponge-associated O. savignyi in both static seawater and flow-through laboratory experiments. Ophiuroids were allowed to move towards the preferred mimic. The defensive properties of the sponge extract against fish predation and fouling were also evaluated. Chemotaxis assays showed that symbiotic ophiuroids were able to chemically recognize its host sponge, moving significantly more towards mimics containing G. corticostylifera extract. Chemical deterrence assays showed that the natural concentration of the extract of this sponge was also able to inhibit generalist fish predation on field experiments and the attachment of the common mussel Perna perna in laboratory assays. These results indicate that the crude extract of G. corticostylifera plays multiple functions in the marine environment, presumably being responsible for a closer association of this sponge with O. savignyi, providing protection for this ophiuroid and inhibition of epibionts on itself.