Unvegetated Areas Research Articles

The relationship between the soil seed bank and above‐ground vegetation of a coastal barrier island

Abstract. The germinable soil seed bank is described from a coastal barrier island off the northwest coast of Florida, USA. Soil samples collected from seven vegetation types, recently deposited dredge spoil and unvegetated areas in autumn 1990 and spring 1991 were placed out in greenhouse trays. 110 taxa germinated from the samples with the largest number (41) being C3 perennial dicots. The largest number of taxa germinated from dry (57) and wet (54) swales, the fewest (one species: Heterotheca subaxillaris) from strand. Similarity of seed bank densities to above‐ground species cover was low (Jaccard's Index = 0.36), not different between vegetation types, but higher in the autumn than in the following spring. Compositional gradients in the seed bank and above‐ground vegetation determined using DCA ordination were highly correlated and related to distance from mean high water, and plot elevation. At the landscape scale, the seed bank provided an equally clear delineation of vegetation types to that based upon the above‐ground vegetation. The seed bank of low disturbance, late‐succession vegetation types (wooded dunes, swales, marshes) was well developed (high species richness, emergent density, and percentage annual species) with the exception that the large‐seeded woody species (i.e. Quercus spp.) were absent from the wooded dune seed bank. By contrast, a poorly developed and transient seed bank occurred in more frequently disturbed (extensive sand movement, salt spray), early successional dredge spoil, unvegetated areas and strand. These contrasts support a general pattern of increasing seed bank development and a persistent rather than transient seed bank with decreasing disturbance frequency, increasing time since disturbance and successional maturity.

Fish Communities in a Tropical Bay (Gazi Bay, Kenya): Seagrass Beds Vs. Unvegetated Areas

The fish fauna of seagrass beds and unvegetated areas in Gazi Bay (Kenya) was sampled in 9 stations with a beach seine. A total of 3601 fishes (> 95 % juveniles) was caught, comprising 75 species (40 families). The spatial patterns in the fish communities were investigated with multivariate statistical techniques. A first community occurred in the downstream part of a major river-fed creek and was characterised by a low density and diversity. These were sandy bottom stations with sparse seagrass vegetation. The dominant species of this community were Leiognathus elongatus and Bothus myriaster. A second community occurred in the upstream part of the same creek, and was characterised by a high density and diversity. Gerres acinaces and Atherinomorus duodecimalis were the dominant species. A third community occurred in the stations of the shallow part of the bay and was characterised by a high diversity but a lower density. The dominant species were Apogon thermalis and G. acinaces. Both latter communities occurred in stations with dense seagrass beds.

Macroinfauna of a Southern New England salt marsh: seasonal dynamics and production

The animal-habitat relationships and seasonal dynamics of the benthic macroinfauna were investigated from November 1986 to October 1988 in the Great Sippe-wissett salt marsh (Massachusetts, USA). Total macrofaunal abundance varied seasonally, displaying a peak in late spring and early summer, then declining sharply during late summer and recovering briefly in fall before collapsing in winter. Three macroinfaunal assemblages were found in the marsh, distributed along gradients of environmental factors. These included a sandy non-organic sediment assemblage, a sandy organic sediment assemblage and a muddy sediment assemblage. The species groups characteristic of unstable sandy non-organic sediments included the polychaetes Leitoscoloplos fragilis, Aricidea jefreyssi, Magelona rosea and Streptosyllis verrilli, the oligochaete Paranais litoralis, and the crustacean Acanthohaustorius millsi. Sandy organic sediments were characterized by the polychaetes Marenzelleria viridis, Capitella capitata, Neanthes succinea, N. arenaceodonta, Polydora ligni and Heteromastus filiformis, the oligochaete Lumbricillus sp., and the mollusc Gemma gemma. In muddy sites, the polychaete Streblospio benedicti and the oligochaetes Paranais litoralis and Monopylephorus evertus were the dominant species. Secondary production of benthic macroinfauna in each of these habitats was estimated. The highest values of biomass and production were recorded in the sandy organic sediments. Secondary production was estimated to be 1850 kJ m-2 yr-1 in sandy organic areas, but only 281 kJ m-2 yr-1 in sandy non-organic areas and 113 kJ m-2 yr-1 in muddy areas. This results in an area-weighted average production of 505 kJ m-2 yr-1 for the unvegetated areas of the marsh. The Great Sippewissett salt marsh has an area of 483800 m2, the total secondary production of the macroinfauna for the whole unvegetated area of the marsh was estimated as 4651 kg dry wt yr-1, expressed as somatic growth. This production value seems consistent with production data obtained for other intertidal North Atlantic environments.

Effects of removal of seagrass canopy on assemblages of small, motile invertebrates

MEPS Marine Ecology Progress Series Contact the journal Facebook Twitter RSS Mailing List Subscribe to our mailing list via Mailchimp HomeLatest VolumeAbout the JournalEditorsTheme Sections MEPS 118:129-137 (1995) - doi:10.3354/meps118129 Effects of removal of seagrass canopy on assemblages of small, motile invertebrates Connolly, R. M. To test the importance of seagrass canopy to epifaunal invertebrates in a southern Australian estuary, patches of the short, fine-leaved seagrass Zostera muelleri Irmisch ex Aschers. were cleared of canopy. All other factors were known to be consistent with seagrass presence, and a procedural control was used to measure any effects of the method used to remove seagrass. Effects on epifauna were measured as changes in abundance and biomass of key taxa and in total production, and as differences amongst assemblages, tested using an analysis of similarity (ANOSIM) randomisation routine. Removal of seagrass canopy had a weak but detectable effect on epifauna over and above the slight effect caused by the disturbance concomitant with seagrass removal. Epifauna associated with habitat from which seagrass had been removed did not, however, match that from areas unvegetated prior to the experiment. The epifauna from these previously unvegetated areas were characterised by low abundance and biomass of several key taxa, apart from 1 group, cumaceans, which were far more common in this habitat. The results suggest that the overriding importance of Z. muelleri to epifauna is not simply the presence of seagrass canopy, and explanations of the higher abundance of epifaunal invertebrates in vegetated compared to unvegetated habitats based merely on the presence of seagrass canopy are not supported. Zostera . Macrofauna . Crustacea . Predation . ANOSIM Full text in pdf format PreviousNextExport citation RSS - Facebook - Tweet - linkedIn Cited by Published in MEPS Vol. 118. Publication date: March 09, 1995 Print ISSN: 0171-8630; Online ISSN: 1616-1599 Copyright © 1995 Inter-Research.

Removal of seagrass canopy: effects on small fish and their prey

In an experiment in a southern Australian estuary, patches of seagrass canopy were removed to test the importance of the canopy to fish in areas where all other factors were known to be consistent with seagrass presence. The total number of fish was slightly lower in patches cleared of seagrass than in patches of undisturbed seagrass, but was not as low as in unvegetated patches. The benthic habitat was expected to be especially important to non-pelagic species, yet their numbers, and those of the most important commercial species, Sillaginodes punctata, were not lower in patches cleared of seagrass, despite being lower in unvegetated patches. The disturbance associated with removing seagrass was simulated and was not found to affect fish numbers. The diet of all fish caught consists mainly of invertebrates associated with the seagrass canopy and sediment surface (epifauna). Epifaunal abundance and production were highest in seagrass patches, lowest in unvegetated patches and intermediate in patches cleared of seagrass. Patterns of fish abundance did not provide evidence of the importance of seagrass canopy in attracting increased fish abundances compared with unvegetated areas but were consistent with a model stressing the importance of prey availability in the role seagrass plays as habitat for small fish.

Comparisons of Cove Rotenone and Electrofishing Catch-Depletion Estimates to Determine Abundance of Age-0 Largemouth Bass in Unvegetated and Vegetated Areas

ABSTRACT We compared cove rotenone sampling and Leslie catch-depletion estimates of age-0 largemouth bass (Micropterus salmoides) abundance in vegetated and unvegetated areas of Lake Guntersville, Alabama, in July—August 1992. With the Leslie catch-depletion method, bass abundance was estimated in 0.05 to 0.32-hectare areas enclosed with a block net. A 4.3-meter direct-current electrofishing boat was able to completely fish each enclosure with five to seven consecutive sampling runs and constant effort. Size-selectivity was not evident as no significant correlations existed between mean length and mean weight at capture for each sampling interval. In addition, catchability remained constant over time. In a cove that contained Eurasian watermilfoil (Myriophyllum spicatum), we estimated 1,088 age-0 largemouth bass/hectare from the two electrofishing catch-depletion samples compared to 1,386 bass/hectare by subsequent rotenone sampling. Cove rotenone density and biomass estimates in four unvegetated coves (1...

A comparison of fish assemblages from seagrass and unvegetated areas of a southern Australian estuary

Assemblages of small fish from eelgrass (mainly Zostera muelleri) and unvegetated patches in a shallow, marine-dominated estuary were compared over one year as a preliminary step towards finding the consequences of eelgrass loss to small fish. There were more species and more individuals at eelgrass sites than at unvegetated sites at every sampling period. Multivariate analysis (MDS ordination) of assemblages showed distinct grouping of eelgrass and unvegetated sites. The statistical significance of groupings was tested by using an analysis of similarities (ANOSIM) randomization routine. The cryptic syngnathid Stigmatopora nigra and juvenile whiting, Sillaginodes punctata, the species of greatest economic importance in the estuary, were predominantly over eelgrass, whereas the flounder Rhombosolea tapirina was usually caught at unvegetated sites. Atherinosoma microstoma, the most abundant species, was more common over eelgrass at two dates but had similar abundances over both habitats at other dates. The limitations of survey work caused by possible associations between the presence of vegetation and environmental factors have been partially offset by interspersion of sites and by measurement of water temperature and salinity. Secondarily to habitat differences, fish assemblages were weakly grouped according to distance of sites from open water.

Effects of an Introduced Aquatic Plant, Hydrilla verticillata, on Benthic Communities in the Upper Chesapeake Bay

The introduced submersed aquatic plant, Hydrilla verticillata, is currently spreading in the Chesapeake Bay, and is colonizing previously unvegetated areas and displacing native flora. To examine the effects of this invasion, we compared faunal densities between Hydrilla beds and unvegetated sand areas. Transplants of Hydrilla and sampling from natural Hydrilla and unvegetated patches indicate that this plant enhances the abundances of a variety of benthic fauna. Among two clam species transplanted into Hydrilla patches and adjacent unvegetated areas, one demonstrated enhanced survival within Hydrilla patches while the other exhibited reduced tissue growth. Similarly, the growth form of Hydrilla may have been affected by clam presence. Dissolved oxygen, sediment characteristics, and sediment resuspension also varied between Hydrilla and unvegetated areas. The spread of Hydrilla in the Chesapeake Bay clearly has significant, though complex, effects on the bottom community. Such effects emphasize the potential importance of introduced plants in altering community characteristics.

Comparing benthic macrofaunal assemblages of creekbank beds of the spikerushEleocharis parvula (R&S) link and adjacent unvegetated areas in a Mississippi brackish marsh

The benthic macrofaunal assemblages inhabiting creekbank beds of the dwarf spikerushEleocharis parvula and adjacent unvegetated areas were compared in a brackish marsh in Mississippi to determine the importance of diminutive plants as invertebrate habitat. Benthic samples were collected over a 12-month period.Eleocharis beds contained 3.6 times higher densities of organisms than adjacent unvegetated areas. Fifteen of 35 taxa collected in beds accounted for 94.7% of the total assemblage (13,266 organisms per m2). Fifteen of 37 taxa collected in bare areas represented 92.4% of the total assemblage (3,625 organisms per m2). Dominant species in beds were the oligochacteMonopylephorus parvus, the crustaceansHargeria rapax andCorophium louisianum, and the insectHydrellia americana (a leaf miner ofEleocharis). Dominant species in bare areas were the polychaetesStreblospio benedicti andHobsonia florida and the crustaceanH. rapax. Overall, the greatest proportion of organisms (60–100%) were distributed within the upper 2 cm of substrate in both areas. Although several factors may explain observed differences in diversity and abundance between beds and bare areas, organisms inEleocharis beds probably benefit most from the relative stability provided by the physical structure of beds. Plants provide attachment sites for tube-building species and may limit predation and disturbance from large aquatic organisms.

Downstream Effects of Timber Harvesting on Channel Morphology in Elk River Basin, Oregon

AbstractDownstream effects, a type of cumulative watershed effect, were identified using changes in the width and distribution of open riparian canopies measured from aerial photography taken between 1956 and 1979 in Elk River basin, southwest Oregon. Open canopies appear on aerial photographs of densely forested basins as unvegetated areas bordering stream channels. Opening occurs when large disturbances, such as landslides, debris flows, large floods, and excessive sedimentation, disrupt the vegetation in the riparian corridor. Downstream changes in channel morphology, inferred by the changing pattern of open reaches were linked to upslope forestry activities; a causal link was assumed where: (i) open reaches extended continuously downstream from clearcuts and roads or (ii) the timing and pattern of opening downstream varied in direct relation to the intensity of upslope forestry activities. Open riparian canopies were observed in first‐ through fifth‐order channels, though only 11% of open reaches in low‐order channels were spatially connected to open reaches in higher order channels. Open reaches on low‐order tributaries were attributed to landslides and surface erosion generated from clearcuts and roads; the total length of open reaches in low‐order channels increased 30‐fold during the study period. Open reaches occurred on higher‐order channels throughout the study period but did not increase in size or change location in relation to upslope harvest activities. Instead, open canopies were restricted mainly to wide and low gradient channel reaches, which comprised approximately one‐third of the length of higher‐order channels. Limited downstream change in riparian canopies associated with upslope forestry activity during the study period, which included a 100‐yr storm, was attributed to three physical factors: (i) lack of debris flows in most parts of the basin; (ii) channels constrained by competent hillslopes limiting the potential for opening; and (iii) low harvest levels over much of the basin at the time of the 100‐yr storm. While air photo interpretation proved useful in deciphering the gross disturbance history of the basin and in distinguishing the general processes which generate downstream effects, sedimentation processes that do not disturb the riparian canopy may also be active in Elk River basin but were not detected due to the coarseness of the techniques used.

Seagrass habitat complexity and macroinvertebrate abundance in Lakshadweep coral reef lagoons, Arabian Sea

Macrofauna of seagrass community in the five Lakshadweep atolls were studied and compared. The associated epifaunal and infaunal taxa comprising nine major taxonomic groups, showed significant differences in the total number of individuals (1041–8411 m−2) among sites and habitats. The density of macrofauna was directly related to mean macrophytic biomass (405–895 g wet wt. m−2). The fauna was dominated by epifaunal polychaetes, amphipods and isopods in the vegetated areas. When compared with the density of nearby unvegetated areas\(\left( {\bar x = 815{\text{m }}^{ - 2} } \right)\), seagrass meadows harbour a denser and richer macroinvertebrate assemblage\(\left( {\bar x = 4023{\text{m }}^{ - 2} } \right)\).

Fishes and Decapod Crustaceans of Cape Cod Eelgrass Meadows: Species Composition, Seasonal Abundance Patterns and Comparison with Unvegetated Substrates

Bimonthly trawl samples from eelgrass and nearby unvegetated areas on Cape Cod, Massachusetts, showed greater species richness in eelgrass meadows relative to unvegetated areas, and greater summer abundance in vegetation for decapod crustaceans and fishes. The composition of eelgrass-associated decapods and fishes was dominated by cold-water taxa and was strikingly different from that of the better studied eelgrass meadows of the mid-Atlantic coast. Four of the eight decapod species collected, including the second and third most abundant taxa, do not even appear in collections reported from Chesapeake Bay eelgrass meadows. Similarly, 10 of the 22 fish species taken, including the first and sixth most abundant species, are not reported from Chesapeake Bay eelgrass samples. Cape Cod eelgrass beds seem to play a nursery role for several commercially important fish species, although the nursery function is less obvious than in previously studied mid-Atlantic eelgrass meadows.

Occupation of submerged aquatic vegetation by fishes: testing the roles of food and refuge.

We conducted a series of field experiments to examine the roles of refuge and food availability in explaining the distribution and abundance of fish in tidal freshwater marsh creeks. Two hypotheses were tested: (1) relative predation pressure is less in SAV than in unvegetated areas and (2) fish food availability is greater in SAV than in nearby unvegetated areas. Tethering experiments using mummichogs (Fundulus heteroclitus) in vegetated and unvegetated areas revealed that relative predation pressure was significantly less in areas with SAV. Banded killifish (Fundulus diaphanus) maintained in vegetated enclosures consumed prey associated with SAV, whereas those held in unvegetated pens had empty stomachs. No differences were found in the number of prey eaten by bluespotted sunfish (Enneacanthus gloriosus) or mummichogs when confined in vegetated or unvegetated enclosures. However, larger prey were consumed by bluespotted sunfish and mummichogs maintained in vegetated enclosures. These data suggest that foraging profitability is significantly enhanced by feeding in the SAV. Submerged plant beds in tidal freshwater marsh creeks not only afford protection from predators, but also provide a rich foraging habitat. By foraging in SAV, fish consume larger prey and may have higher growth rates, lower mortality, and higher fecundity.

Community Changes Associated with the Spread of an Introduced Seagrass, Zostera Japonica

Species introductions have provided a valuable source of information for understanding the factors that regulate community composition. However, the effect of such introductions has often been obscured by a lack of information on distribution and abundance patterns before or during an invasion event. I examined the changes in a benthic community associated with the ongoing spread of an introduced seagrass, Zostera japonica, by sampling transplanted seagrass plots and established Z. japonica patches of known ages. The sedimentary and faunal changes associated with Z. japonica were similar to those observed with native seagrass species. Mean sediment grain size declined and sediment volatile organics increased within Z. japonica patches. Faunal richness was higher within Z. japonica patches compared with adjacent unvegetated areas, and many numerically dominant species were positively associated with this introduced seagrass. However, the effects of Z. japonica on faunal abundance varied with both the age of a seagrass patch and site location. The introduction of this seagrass has thus changed the physical habitat as well as the richness and densities of resident fauna. Many studies of introduced species have concentrated on direct interactions between introduced and native organisms. In contrast, the community changes associated with the introduction of Z. japonica emphasize the potential importance of indirect and system—level effects of introduced species on community composition.

Micro-arthropods associated withWelwitschia mirabilisin the Namib Desert

Significantly more micro-arthropods were found under Welwltschla plants than in unvegetated areas. Densities varied considerably with locality of plant ranging from a mean ± SE of 1710 ± 720 micro-arthropods m-2 in the Northern Najnib 10 145 700 ± 85 930 micro-arthropods m-2 in the southernmost locality. The same taxa were found at all sites and these are also common to North American and Australian deserts. The structure of the micro-arthropod community varied considerably with locality and sex of the plant. Tarsonemld mites were numerically dominant at some sites while Collembofa were dominant at others. Prostlgrnatids were the most abundant group of mites, and mites were generally the most abundant micro-arthropod.

Fish and macrocrustacean use of submerged plant beds in tidal freshwater marsh creeks

Fishes and macrocrustaceans (nekton) were sampled biweekly from mid-June through October 1985 from submerged aquatic vegetation (SAV) and during September 1985 from unvegetated areas of tidal freshwater marsh creeks. Nineteen species of fishes (6918 individuals, 3.068 kg preserved wet weight) from 9 famhes, and 3 species of invertebrates (12036 individuals, 1.577 kg preserved wet weight) were collected from SAV The most abundant species were grass shrimp Palaemonetes pugio, banded killifish Fundulus diaphanus, mummlchogs F. heteroclitus, bluespotted sunfish Enneacanthus gloriosus, and mud crabs Rhithropanopeus harrisj:. Recruitment of small juvenile fishes of several species coincided with the period of greatest SAV area1 cover (late June through September). Average densities of fauna were significantly greater in SAV than over nearby unvegetated creek bottoms in September. Fauna1 abundance was not significantly related to SAV biomass, perhaps because the structural complexity of the SAV in our study area was high and nekton were abundant when SAV was present at low densities (i.e. low biomass). The numerically dominant species that occurred in SAV at low tide and on the marsh surface at high tide were similar Grass shrimp, banded lullifish, mummichogs and bluespotted sunfish accounted for > 90 % of the total number of organisms collected in each of the 2 habitats, SAV and marsh surface. The SAV of tidal freshwater marsh creeks is probably most important as habitat for forage fishes. Young of recreational species such as bluegill Lepomis macrochirus and pumpkinseed L. gibbosus also occupy this habitat.

Exposure, vegetation and sediment as primary factors for mobile epibenthic faunal community structure and production in shallow marine soft bottom areas

Mobile epibenthic fauna was quantitatively assessed in 22 shallow-water (0–1.5 m), soft-bottom areas on the Swedish west coast. Relationships between faunal structure, density, biomass and annual production on the one hand, and physical factors such as exposure, vegetation cover and sediment structure on the other, were investigated. The areas were grouped according to their physical characteristics into 4 categories: exposed, semi-exposed and sheltered with little or no vegetation and vegetated. Species number, density, biomass and annual production were shown to be significantly higher in vegetated areas ( Zostera marina L.) than in mainly unvegetated ones. The epibenthic production in Z. marina meadows was 6 g (AFDW)-m −2·y −1. Species composition was also different and Carcinus maenas L. was the only dominant species occurring in both vegetated and unvegetated areas. In all unvegetated areas the number and composition of species were about the same, regardless of exposure, whereas biomass and production varied with the degree of exposure. Highest production (3 to 5.5 g (AFDW)-m −2·y −1) in these areas was found in semi-exposed areas. In exposed and sheltered areas production was 1.1 to 3.3 g and 0.6 to 1.7 g (AFDW)-m −2·y −1, respectively. Degree of exposure, vegetation cover and sediment structure are suggested as the primary factors affecting the composition and quantity of mobile epibenthic fauna in the investigated areas. Biotic factors e.g. predation and competition are generally considered to be subordinate to these physical factors.

The influence of waves and seagrass communities on suspended particulates in an estuarine embayment

Resuspension of bottom sediments by waves, corresponding changes in suspended particulate material (SPM) concentrations in the overlying water column, and transport pathways of SPM were investigated in a shallow estuarine embayment colonized by seagrass communities in Chesapeake Bay. In shallow (<2 m), unvegetated regions significant resuspension of bottom sediments was evident when northerly winds exceeded 25 km h −1, increasing SPM concentrations up to 10-fold. High concentrations of SPM generated by resuspension dissipated rapidly (within 24 h) after winds became calm. Patterns of SPM along the embayment's depth gradient suggest that part of this resuspended material was transported offshore into deeper reaches of the estuary. In areas of the embayment colonized by seagrasses, wave energy was attenuated by the vegetation, suppressing resuspension and enhancing deposition. As a result, SPM concentrations were significantly lower inside seagrass beds than in adjacent unvegetated areas. During periods of high winds when wave induced resuspension occurred in unvegetated areas, SPM concentrations remained unchanged inside the bed at normal water levels. However, when water levels were elevated by spring tides or storm surges, plants were less effective at attenuating wave energy, and SPM concentrations increased inside the seagrass bed due to resuspension and advective processes. Calculations based on the results of this study indicate that sedimentation rates are substantially higher in seagrass communities than in unvegetated areas.

The response of interstitial ammonium in eelgrass ( Zostera marina L.) beds to environmental perturbations

Natural and human-induced perturbations of eelgrass ( Zostera marina L.) beds were used to examine the interaction between the sediment interstitial ammonium pool and nitrogen uptake by the plants. Eelgrass colonization of unvegetated areas was accompanied by a substantial decrease in the interstitial ammonium pool over a 4-yr period. The changes in interstitial ammonium and shoot density during colonization support an already determined relationship between shoot density and ammonium pool measurements. In field perturbation experiments, removing eelgrass leaves and sealing the sediment surface altered the flux of ammonium from the interstitial ammonium pool, and resulted in a rapid increase in interstitial ammonium concentrations. Measurements of ammonium accumulation under the various perturbation conditions and a control permitted calculation of the sediment ammonium flux. These estimates include uptake by eelgrass roots, regeneration in the root zone, and diffusion from the sediments. Nitrogen limitation was observed in some eelgrass beds.

Refuges for infaunal bivalves from blue crab, Callinectes sapidus (Rathbun), predation in Chesapeake Bay

Monthly field sampling in the Choptank River, Chesapeake Bay, revealed rapid increases in population densities of infaunal bivalves in winter and spring followed by a decline in summer, presumably as a result of increased predation. Spat settlement of Mya arenaria L. occurred predominately in late fall and winter, declining in early spring. Macoma balthica (L.) spat settled primarily in spring. Spat of Macoma mitchelli Dall and Mulinia lateralis (Say) were found throughout most of the year. Field studies revealed a tendency for larger Mya arenaria, Macoma balthica, and M. mitchelli to be found deeper in the sediment. Laboratory studies tested the effectiveness of possible refuges within the sediment in protecting the soft clam, Mya arenaria, from blue crab ( Callinectes sapidus Rathbun) predation. Mya arenaria living below 10 cm were foraged for by blue crabs less effectively than those living closer to the surface. In the field, soft clams which settled in the fall were found living below 10 cm by the following May, reaching their maximum burial depth of 25 cm in June–July. An artificial plant structure also afforded some degree of protection from blue crabs in laboratory experiments: sparse and dense assemblages of artificial plants allowed higher survival of M. arenaria compared with control treatments. The protection afforded infaunal bivalves by these spatial or structural refuges is not absolute, although living deep within the sediment or within vegetation beds reduces the risk of predation in comparison to living near the sediment surface or in unvegetated areas.