Patrick J. Schembri

Feeding behaviour of fifteen species of hermit crabs (Crustacea: Decapoda: Anomura) from the Otago region, southeastern New Zealand

Summary The feeding behaviour of 15 species of hermit crabs from four families (Pylochelidae, Diogenidae, Paguridae and Parapaguridae) and from habitats ranging from rocky intertidal through continental shelf gravelly sediments to finer sediments on the shelf edge and continental slope, is described. Feeding mechanisms found include various methods of deposit-feeding, browsing, suspension-feeding, predation and scavenging. Each species employs one or two primary feeding mechanisms but also a number of secondary mechanisms. Depositfeeding techniques include scooping up detritus-rich sediment, scrubbing detritus from the surface of small granules, scraping it from larger surfaces, picking it out of crevices and other irregularities, and brushing out and ingesting material which adheres to the general body setation. A few species browse on algae and sedentary colonial invertebrates which they slice or pluck off with the chelae. A number of species filter particles out of suspension using either the general s...

Symbiotic associations between hermit crabs and bryozoans from the Otago region, southeastern New Zealand

Thirteen species of bryozoans (six cyclostomes and seven cheilostomes) occurring off the Otago Peninsula (southeastern New Zealand) form symbiotic, possibly mutualistic, associations with hermit crabs. For all but one of these bryozoan species, such an association has not been reported previously. At most only 3 of the 13 bryozoans are obligate symbionts of hermit crabs. Associations are apparently initiated when a bryozoan larva settles on a gastropod shell occupied by a juvenile hermit crab and develops to form a colony which encrusts the whole shell and then continues to grow out beyond the shell aperture in the form of a helicospiral tube. The tube-building bryozoan colony grows in step with the crab, and tube development appears to be controlled by crab morphology and activity. Of the six species of hermit crabs found occupying bryozoan tubes, four were regular tube occupants, although they could also be found in other shelter types. Hermit crabs occupying bryozoan tubes very rarely indulged in shell...

Evolutionary history of the Maltese wall lizard Podarcis filfolensis: insights on the ‘Expansion–Contraction’ model of Pleistocene biogeography

The expansion–contraction (EC) model predicts demographic and range contraction of temperate species during Pleistocene glaciations as a consequence of climate‐related habitat changes, and provides a paradigm for explaining the high intraspecific diversity found in refugia in terms of long‐term demographic stability. However, recent evidence has revealed a weak predictive power of this model for terrestrial species in insular and coastal settings. We investigated the Pleistocene EC dynamics and their evolutionary consequences on temperate species using the Maltese archipelago and its endemic lizard Podarcis filfolensis as a model system. The evolutionary and demographic history of P. filfolensis as inferred from mitochondrial and nuclear sequences data does not conform to the EC model predictions, supporting (i) demographic and spatial stability or expansion, rather than contraction, of the northern and southern lineages during the last glacial period; and (ii) a major role for allopatric differentiation primed by sea‐level dynamics, rather than prolonged demographic stability, in the formation of the observed genetic diversity. When combined with evidence from other Mediterranean refugia, this study shows how the incorporation of Pleistocene sea‐level variations in the EC model accounts for a reverse demographic and range response of insular and coastal temperate biotas relative to continental ones. Furthermore, this cross‐archipelago pattern in which allopatric diversity is formed and shaped by EC cycles resembles that seen between isolated populations within mainland refugia and suggests that the EC model, originally developed to explain population fluctuations into and out‐of refugia, may be appropriate for describing the demographic and evolutionary dynamics driving the high genetic diversity observed in these areas.

The Maltese Islands: climate, vegetation and landscape

The Maltese Islands, situated in the central Mediterranean, occupy an area of only some 316 km2. The climate is typically Mediterranean: the average annual rainfall is c. 530 mm of which some 85% falls during the period October to March; the mean monthly temperature range is 12--26 °C, and the islands are very windy and sunny. Although small, the Maltese Islands have a considerable diversity of landscapes and ecosystems which are representative of the range and variety of those of the Mediterranean region. The islands are composed mainly of limestones, the soils are young and are very similar to the parent rocks, and there are no mountains, streams or lakes, but only minor springs; the main geomorphological features are karstic limestone plateaux, hillsides covered with clay taluses, gently rolling limestone plains, valleys which drain runoff during the wet season, steep sea-cliffs on the south-western coasts, and gently sloping rocky shores to the Northeast. The main vegetational assemblages are maquis, garigue and steppe; minor ones include patches of woodland, coastal wetlands, sand dunes, freshwater, and rupestral communities; the latter are the most scientifically important in view of the large number of endemic species they support. Human impact is significant. Some 38% of the land area is cultivated, c. 15% is built up, and the rest is countryside. The present landscape is a result of the interaction of geology and climate, coupled with the intense human exploitation of the environment over many thousands of years, which has altered the original condition of the vegetation cover, principally through the diversion of vast tracts of land to cultivation, the construction of terraces, water catchment devices, irrigation channels and drainage ditches, the grazing of animals on uncultivated land, and the development of land for buildings and industry. The scantiness of the soil, combined with the erratic rainfall and the periodic disturbance of the vegetation cover, has resulted in extensive erosion. As a consequence it is now difficult for the original vegetation to reassert itself, affecting the landscape drastically and permanently. Much of the original native flora has been lost or marginalised and the present day non-urban landscape is now dominated by vegetation consisting mainly of ruderal and introduced species. As the population increases, and human pressure on the environment mounts, such trends are likely to continue and it is only very recently that some important initiatives have been taken to manage the environment and halt the deterioration of the landscape.

Feeding and particle selection in the echiuran worm Bonellia viridis Rolando (Echiura: Bonelliidae)

The feeding behaviour of Bonellia viridis Rolando is described. Small individual particles (⪕ 94 ± 25 μm) are taken onto the proboscis by means of cilia. Individual particles of intermediate size (150±40 μm) are taken up by means of a combination of ciliary and muscular activity. The larger particles (230 ± 60 to 290 ± 60 μm) are picked up by muscular action. When presented with large accumulations of particles, the terminal lobes pass them onto the ventral surface of the proboscis by active muscular action. The ciliary currents on the ventral surface of the proboscis have been mapped. Particles are carried from the terminal lobes of the proboscis to the mouth via the stem gutter and boluses may be formed at the neck region. Transport along the proboscis may be ciliary for the smaller particles (⪕ 290 ± 60 μm), ciliary and muscular for the intermediate sizes (230±60 to 480 ± 70 μm) or purely muscular for the largest particles (550 ±140 μm). Particles may be rejected at various levels along the proboscis, the site of rejection also being dependent on size. The density of ciliation in the various regions of the proboscis has been estimated and correlated with function. Ventrally (i.e., on the uppper surface) cilia are densest at the fringe region, the terminal lobe gutters and the lateral margins of the proboscis. The dorsal surface of the proboscis is always extremely sparsely ciliated except for a short distance from the ‘leading’ edge which has a dense covering of locomotory cilia. When presented with a monolayer of clean sand grains of different sizes, most animals only pick up particles of 150±40 m and less. When presented with a choice of clean sand grains or sand enriched with various plant and animal extracts, the terminal lobes actively pick up enriched particles but not, to any appreciable extent, clean sand. The terminal lobes are not attracted to the enriched substratum from a distance. Once on Isochrysis-enriched sand they spend significantly more time on it. With two other plant-enriched substrata (Phaeodactylum and Ulva) they again spend more time on them than on clean sand, but the difference was not statistically significant. The terminal lobes do not feed exclusively on enriched substrata but graze over all the available area.

Sex determination and larval sexual interaction in Bonellia viridis Rolando (Echiura: Bonelliidae)

Abstract Sex determination in the echiuran Bonellia viridis Rolando has classically been regarded as depending primarily on the environment of the newly settled larvae. The majority of the sexually undifferentiated larvae settling on an adult female become males; the larvae which settle away from the adult female become females in most cases. Previous work on this problem is reviewed. The behaviour, including the time-course of settlement, and the development of the indifferent larvae of B. viridis in the presence and absence of an adult female is described. Evidence is provided that even in the absence of adult females there is an interaction between newly settled larvae such that up to 20% of larvae become attached to one another in pairs with masculinization of one partner, the other developing into a female. Using larvae cultured singly, it was shown that crude extracts of adult female proboscis and trunk body-wall and the pigmented secretion of an irritated female masculinize indifferent larvae, the vast majority of which would have developed into females in pure sea water. Solutions of the purified integumentary pigment, bonellin, gave inconsistent results. Our experiments prove conclusively that sex determination is metagamic i.e., not fixed at fertilization but is the result of an interaction between genetic and environmental factors, in ≈ 83% of all larvae. The main environmental factor is a substance produced by the female. The other 17% are larvae whose sex is determined exclusively by their genetic make-up. These are syngamic males, females and intersexes. The problem of indefinitely undifferentiated larvae is discussed.

Some aspects of the ecology of the echiuran worm Bonellia viridis and associated infauna

In inshore waters around Malta, Bonellia viridis inhabit burrows with multiple exits in calcareous rocks, and are most abundant in areas bordering Posidonia oceanica meadows. The associated epiflora and fauna are typical of the Mediterranean hard-substrate infralittoral zone. The larger infaunal species associated with B. viridis include a poriferan, Cliona sp.; an unidentified nemertean worm; various polychaetes; the sipunculans Phascolosoma granulatum and Aspidosiphon muelleri; two species of molluscs-Lithophaga lithophaga and Lepidopleurus cajetanus; 8 species of decapod crustaceans; and the teleost Gobius geniporus. Four types of burrows are found in rocks containing B. viridis. The larger burrows (referred to herein as UBA burrows) contain 3 main species-the decapods Upogebia deltaura and Alpheus dentipes and B. viridis itself. Experimental evidence suggests that U. deltaura excavates the UBA burrows mechanically, although B. viridis may secondarily modify them by secretion of an acidic mucus and gentle mechanical action. There is a definite community of organisms living as commensals in the burrow of U. deltaura. The assemblage of organisms bears a striking parallelism to the assemblages inhabiting the burrows of species of Upogebia, Callianassa and the echiuran Urechis caupo of the Pacific coast of N. America. A food web for the UBA burrow community is suggested.

Low faunal diversity on Maltese sandy beaches: fact or artefact?

Eight sandy beaches on Malta and two on Gozo were sampled for macrofauna to test the hypothesis that Maltese beaches have an intrinsically low diversity. Stations distributed in the supralittoral (dry zone), mediolittoral (wet zone) and upper infralittoral (submerged zone to 1 m water depth) were sampled by sieving core samples and standardised searching during daytime, and pitfall trapping and standardised sweeping of the water column using a hand-net at night, as appropriate. Physical parameters of the sediment were measured and human occupancy of the beaches was estimated. From the supralittoral and mediolittoral, 39 species represented by 1584 individuals were collected by the combined techniques of pitfall trapping, sieving and standard searching. For Ramla beach, which had the highest diversity, 267 individuals representing 25 infaunal species were collected by sieving from a combined volume of 1.175 m 3 of sand, and 149 individuals representing 28 epifaunal species were collected by standardised searching from a combined area of 700 m 2 of sand during two winter and two summer sampling sessions between 1992 and 1993. For nine other beaches sampled during the summer of 2000, only six macrofaunal species were collected from core samples, with overall population densities ranging from 4.13 to 45.45 individuals m � 2 . Only 92 individuals belonging to 12 species were collected by hand-net from the uppermost infralittoral of five beaches sampled using this method during the summer of 2000. Taxa of gastropods, bivalves, decapods, mysids and staphylinid beetles generally abundant on Mediterranean sandy beaches, were entirely absent from the beaches sampled. Few correlations that could explain the impoverishment of Maltese sandy beaches were found between physical parameters and faunal abundances, and other factors such as inadequate sampling effort, human disturbance and marine pollution were also excluded; however, seasonally biased sampling may partly explain the results obtained. One factor that may explain why certain species are missing could be lack of recruitment, due to Malta’s geographical isolation from the European and African mainlands. 2003 Elsevier Ltd. All rights reserved.

Improving the Conservation of Mediterranean Chondrichthyans: The ELASMOMED DNA Barcode Reference Library

Cartilaginous fish are particularly vulnerable to anthropogenic stressors and environmental change because of their K-selected reproductive strategy. Accurate data from scientific surveys and landings are essential to assess conservation status and to develop robust protection and management plans. Currently available data are often incomplete or incorrect as a result of inaccurate species identifications, due to a high level of morphological stasis, especially among closely related taxa. Moreover, several diagnostic characters clearly visible in adult specimens are less evident in juveniles. Here we present results generated by the ELASMOMED Consortium, a regional network aiming to sample and DNA-barcode the Mediterranean Chondrichthyans with the ultimate goal to provide a comprehensive DNA barcode reference library. This library will support and improve the molecular taxonomy of this group and the effectiveness of management and conservation measures. We successfully barcoded 882 individuals belonging to 42 species (17 sharks, 24 batoids and one chimaera), including four endemic and several threatened ones. Morphological misidentifications were found across most orders, further confirming the need for a comprehensive DNA barcoding library as a valuable tool for the reliable identification of specimens in support of taxonomist who are reviewing current identification keys. Despite low intraspecific variation among their barcode sequences and reduced samples size, five species showed preliminary evidence of phylogeographic structure. Overall, the ELASMOMED initiative further emphasizes the key role accurate DNA barcoding libraries play in establishing reliable diagnostic species specific features in otherwise taxonomically problematic groups for biodiversity management and conservation actions.

Rocky Shore Biotic Assemblages of the Maltese Islands (Central Mediterranean): A Conservation Perspective

Abstract Limestone rocky shores constitute ca 90.5% of the 272km coastline of the Maltese islands. Only some 40% of this rocky coastline is gently sloping and easily accessible. Such shores are heavily impacted with 96% of the accessible coastline dominated by tourist-related or by maritime activities. We characterised the biotic assemblages of lowland Maltese rocky shores and tested the popularly held view that given the scarce variation in physical characteristics, such shores form a homogenous habitat. Belt transects were laid perpendicular to the shoreline from biological zero to the adlittoral zone on seven Coralline Limestone and one Globigerina Limestone shores. Cover (for algae and encrusting species) or population density (for animals except sponges) were estimated using 0.5m X 0.05m quadrats placed contiguously for the first few metres and then at regularly spaced intervals. Overall, 19 faunal and 47 floral species, and 10 faunal and 8 floral species were recorded from the Coralline and Globigerina transects respectively, with 60.8% faunal and 25.6% floral species common to the two substrata. Hierarchical clustering showed that the Coralline and Globigerina transects harboured distinct biotic assemblages and identified an upper shore assemblage dominated by the littorinid Melarhaphe neritoides and barnacles, and a lower shore assemblage dominated by algae and molluscs; a mid-shore transition zone where certain species from both assemblages reached peaks of abundance was present in almost all Coralline and the majority of Globigerina transects. Differences in biota between the two types of shore are most likely primarily related to differences in microtopography and, to a lesser degree, to exposure. It is concluded that in spite of gross physical similarity, Maltese lowland rocky shores are biotically inhomogeneous, making conservation of individual sites much more important than previously thought.