Top-down versus bottom-up limitation in benthic infaunal communities: Direct and indirect effects

Top-down effects of predators and bottom-up effects related to resource availability can be important in determining community structure and function through both direct and indirect processes. Their relative influence may vary among habitats. We examined the effects of nutrient enhancement and predation in southeastern North Carolina to determine relative effects on benthic macrofaunal communities. Short-term nutrient additions and predator exclusions were conducted in two estuaries to examine main and interactive effects on benthic microalgae and infauna. This experimental approach was complemented by comparisons of microalgal biomass, infaunal abundance and composition, predator abundance and predator exclusion among four estuarine systems that varied in background nutrient levels. In the short-term experiments, nutrient enhancement induced increased microalgal biomass but had limited effects on abundances or sizes of infauna. Predator exclusion increased the density of sedentary and near-surface dwelling fauna, but we did not observe interactions between predation and responses to nutrient additions as might be predicted from a simple cascade model. General patterns of abundance were explained to a larger extent by interannual and amongestuary pattems. These results indicate a lack of simple trophic cascade responses for this community over a short time scale and little evidence for local interactive effects. The lack of interactive effects may reflect the opportunistic nature of the dominant infaunal species and potentially different time and spatial scales for the effects of predation and resource controls.     

A nautiloid with ventral bitemarks from the Charmouth Mudstone Formation (Lower Jurassic) near Lyme Regis,Dorset, UK

Locally within the Charmouth Mudstone Formation (Sinemurian, Lower Jurassic) of the Dorset coast fatally bitten ammonites with ventral bite marks are common. Ventral bite marks occur at the rear of the body-chamber. The predator probably severed attachment muscles, thus removing its prey more easily. We report the first ventrally bitten fossil nautiloid, which is unusual in that the large bite mark is entirely within the phragmocone. The nautiloid is juvenile with a very short body-chamber extending 46° back from the apparently undamaged aperture. The whorl cross-section is nearly circular, giving the predator little to grip when biting. Nautiloid shells are also thicker than ammonite shells. Despite these difficulties, the predator bit repeatedly and damaged both the shell and septa. Five rounded bites occur on the left side, separated by sharp points. This suggests the predator was probably another cephalopod with suckered arms and powerful jaws. The centre of the bite is 160.5° back from the aperture, close to the mean value for ammonites (184.5°), suggesting the predator located the position of the first bite opposite the victim's aperture. Thus, we think the predator attacked an incomplete nautiloid shell which explains why the damage affects only the phragmocone.     

Phytoplankton biomass, cell diameter, and species composition in the low salinity zone of Northern San Francisco Bay estuary

Phytoplankton chlorophyll a concentration, biovolume, cell diameter, and species composition differed across the narrow, low salinity zone between 0.6‰ to 4‰ and may influence copepod food availability in the northern San Francisco Bay Estuary. The highest chlorophyll a concentrations (range 3.2–12.3 μg 1^?1), widest cell diameters (>5 μm diam), highest diatom densities and highest production rates of >10 μm diam cells occurred at the landward edge of the salinity zone in April during a strong spring tide and May during a strong neap tide. Near optimum predator/prey ratios, large prey estimated spherical diameters, and high chlorophyll a concentrations suggest these phytoplankton communities provided good food quantity and quality for the most abundant copepods, Eurytemora affinis, Sinocalanus doerrii, and Pseudodiaptomus forbesi. At the center of the zone, chlorophyll a concentrations, diatom densities, and production rates of >10 μm diam cells were lower and cell diameters were smaller than upstream. Downstream transport was accompanied by accumulation of phytoplankton with depth and tide; maximum biomass occurred on spring tide. The lowest chlorophyll a concentrations (1.4–3.6 μg 1^?) and consistently high densities (3,000–4,000 cells ml^?1) of <5 μm diam cells occurred at the seaward edge of the zone, where the green alga Nannochloris spp. and the bluegreen alga Synechococcus spp. were the most abundant phytoplankton. Low chlorophyll a concentrations and production rates of >10 μm diam cells, small prey estimated spherical diameters, and high predator/prey ratios suggested the seaward edge of the zone had poor phytoplankton food for copepodids and adult copepods. The seaward decrease in phytoplankton chlorophyll a concentration and cell diameter and shift in species composition in the low salinity zone were probably a function of an estuary-wide decrease in chlorophyll a concentration, cell diameter, and diatom density since the early 1980s that was enhanced in the low salinity zone by clam herbivory after 1987. *** DIRECT SUPPORT *** A01BY090 00008     

Reduction of sea star predation by the burrowing response of the hard clam Mercenaria mercenaria (Mollusca: Bivalvia)

The hard clam Mercenaria mercenaria burrows deeper into the sediment when the predaceous sea star, Asterias forbesi is present. The supposition that this increase in burial depth represents an escape response designed to reduce predation was tested experimentally by regulating clam burial depth through manipulation of the amount of sediment available for burial. Mercenaria maintained at zero depth were eaten by Asterias at greater rates than those held at ordinary burial depths (2.5–3.0 cm). These clams in turn were eaten at greater rates than those maintained at escape depths (4.0–4.5 cm). The results unambiguously establish an anti-predator function for the burrowing response, as well as underscoring the protective function of the fossorial habit. They are not confounded by behavioral predator food preferences, inherent differences between prey species, or debilitating side effects of preventing prey from escaping. Mechanisms by which the burrowing response may reduce predation are discussed and observations on the unreported clam-digging behavior of Asterias forbesi are presented.     

Predation impact on the meiobenthic harpacticoidCanuella perplexa in a lagoon of the Po River Delta,Italy

The impact of predation by juvenile marbled gobies,Pomatoschistus marmoratus, on the meiobenthic harpacticoid copepodCanuella perplexa was investigated at a subtidal station located in a shallow-water, brackish embayment of the Po River delta, northern Italy, in summer and fall 1978. In august, predation mainly centered on adults, with a very large number of prey for each predator; the energy supplied by the harpacticoid is high enough to sustain, almost entirely, the gobies’ daily energy requirement. The situation is different in September and October, when the adult copepod population declines dramatically and predation is shifted to less energy-providing copepodites and marbled gobies sensibly widen their prey spectrum.C. perplexa showed alterations of biological parameters, such as clutch size and average brood size, to cope with such a heavy predation pressure.     

Fish Predation on Juvenile Penaeid Shrimp: Examining Relative Predator Impact and Size-Selective Predation

Predation is likely the primary source of mortality for juvenile penaeid shrimp and, therefore, a key factor driving their population dynamics. We sampled juvenile penaeids and their potential predators in a salt marsh from July to August 2012 to examine the impact of these predators and possible size-selective predation on the shrimp population. We quantified predator impact using the frequency of occurrence (FO) index and a relative predation index (RPI) that accounts for predator abundance and the number of shrimp consumed per individual predator. Size selectivity was assessed by comparing the size distribution of shrimp in the study area to the size distribution of shrimp removed from predator stomachs. Two penaeid species, white shrimp Litopenaeus setiferus and brown shrimp Farfantepenaeus aztecus, were collected, and most (86%) were juvenile white shrimp ≤?12 mm carapace length. Spotted seatrout Cynoscion nebulosus, which consumed relatively large shrimp, was the most important predator based on the FO index. Bay whiff Citharichthys spilopterus, which ate the smallest shrimp, was the most important predator based on the RPI. The size distribution of shrimp removed from predator stomachs differed from those collected in the study area; the smallest shrimp were disproportionally more abundant in predator stomachs. Using the RPI, we identified some potentially important predators (e.g., bay whiff) that may impact shrimp populations by consuming the smallest shrimp in estuarine nurseries. Our approach could be useful for examining predator impacts on other populations of juvenile penaeids and more generally for any prey consumed by fish predators.     

Predation on the rocky shores of Patagonia,Argentina

Rocky intertidal communities of Argentinean Patagonia are exposed to harsh physical conditions caused by dry, strong southern trade winds (mean speed 45 km h^-1, gusts up to 140 km h^-1) that result in intense desiccation of intertidal organisms. Predator distributions in these communities were evaluated from April 2003 to December 2004 at two exposed headlands and six protected bays in Cabo dos Bahias (44°50′S, 65°40′W). Crabs and sea stars dominated the predator assemblage, with occasional scavenger snails and fish also present. During low tide, predators were never found in the open but were always associated with shelter (primarily within mussel beds and coralline algae), suggesting a strong predator dependency on foundation species to buffer them from physical stress. Few predators (mostly crabs) emerged from shelter at high tide. Unlike the larger predators found on Chilean rocky shores, invertebrate predators in this system are diminutive, generally < 2 cm. The lone exception was the newly discovered invasive green crab,Carcinus mamas. Feeding trials, gut content analyses, and visual surveys indicated that native predators feed primarily on small, soft-bodied prey.C. mamas fed on slow-moving and sessile animals, including the super abundant musselPerumytilus purpuratus, with much greater voracity than native predators. Because native intertidal organisms are dependent on mussel beds and coralline algae for shelter from desiccation, successful invasion ofC. mamas may lead to a significant decrease in native diversity by consuming foundation species. This study represents a preliminary survey of predator distributions and feeding habits on the rocky shores of Argentinean Patagonia and provides important baseline data to evaluate trophic linkages and predatory effects on Patagonian rocky shores.     

Guess who's coming to dinner: A portrait of Tyrannosaurus as a predator

The debate over whether Tyrannosaurus was a scavenger or an active predator has raged for almost a hundred years. Biomechanical studies show that the king of the tyrant lizards lived up to its name and actively hunted its prey.     

Growth rates of juvenine pinfish (Lagodon rhomboides): Effects of habitat and predation risk

Predation is often the largest source of mortality for juvenile fish and the risk of predation can influence growth rates by either forcing young fish into suboptimal foraging habitats or reducing the amount of time spent foraging. We used field experiments to test effects of predation risk by gulf flounder (Paralichthys albigutta) on juvenile pinfish (Lagodon rhomboides) growth rates by measuring changes in length and weight in three habitats (sand, low density, and high density shoalgrass,Halodule wrightii) in Perdido Key, Florida. Benthic cores, seagrass samples, and stomach contents were also analyzed to examine differences in pinfish prey densities, grass densities and epiphyte coverage, and diet, respectively, among habitat and predator treatments. Both length and weight growth rates were determined and showed similar results. We found that pinfish inhabiting seagrass habitats, particularly low densityHalodule displayed the fastest growth rates in the beginning of the growing season (June) and those in sand had the fastest growth rates later in the season (October). These differences in growth rates did not appear to be influenced by densities of pinfish prey items since the treatment having the highest density of prey was not that in which growth rates were the greatest. This seasonal shift may be attributed to increasing pinfish size. Larger pinfish in October may have been inhibited by high density grass, reducing foraging efficiency. These results demonstrate how occupying a suboptimal foraging habitat can affect juvenile pinfish growth rates. Predation risk significantly reduced length and weight growth rates of pinfish in June, but not October. This suggests that smaller pinfish early in the season traded time spent foraging for predator avoidance, while larger pinfish were likely to have reached a size refuge from predation. This study demonstrates that nonlethal effects from predation are also important influences on juvenile pinfish.     

Elemental Composition of Mnemiopsis leidyi A. Agassiz 1865 and Its Implications for Nutrient Recycling in a Long Island Estuary

The ctenophore Mnemiopsis leidyi is an ecologically important predator in temperate coastal environments. Their populations fluctuate seasonally, serving as sinks of nutrients during periodic blooms, but as sources via excretion and during population collapse. Ctenophores were analyzed for elemental composition (C, N, and P) during 2008 and 2009 in Great South Bay, NY, USA. Salt-free weight percent C, N, and P correlated positively with ctenophore sizes and zooplankton prey abundances. Nitrogen and P were higher at the onset of blooms than during collapse when prey were substantially fewer. Ctenophores collected during average to high zooplankton densities had atomic ratios averaging C/N ~6:1 and C/P ~66:1, but became C- and P-depleted (C/N ~5:1, C/P ~128:1) with decreasing zooplankton. Incubations demonstrated rapid remineralization of ctenophore biomass (as NH₄ ⁺, HPO₄ ^2?), following first order kinetics (e.g., k ~0.1–0.4 day^?1) with enriched stoichiometric N and P fractionation relative to biomass under both oxic and anoxic conditions. Based on reported excretion rates, nutrient regeneration from excretion by active populations greatly exceeds nutrients remineralized during population crashes. To our knowledge, this is the first study documenting natural seasonal patterns in ctenophore elemental stoichiometry as a function of ctenophore size and prey availability.     

Relationship between subcellular cadmium distribution in prey and cadmium trophic transfer to a predator

We tested the hypothesis that exposure-related alterations in the subcellular Cd distribution in prey relate to changes in Cd absorption by a predator. Oligochaete worms,Limnodrilus hoffmeisteri were exposed for 1 wk or 6 wk to 0.5 μg Cd 1^?1, 47 μg Cd 1^?1, or 140 μg Cd 1^?1 (including¹⁰⁹Cd as a tracer) and relationships between oligochaete subcellular Cd distribution and Cd absorption by a predator, the grass shrimp (Palaemonetes pugio), were determined. Concentration and duration of Cd exposure had direct effects on oligochaete subcellular Cd distribution. Changes in oligochaete subcellular Cd distribution were characterized by increases in both the amount and proportion of Cd bound to the cytosolic fraction. The induction of Cd-binding proteins (e.g., metallothioneins) were suspected to be responsible for these changes. We found 1∶1 relationships between the amount and percentage of Cd in oligochaete cytosol and the amount and percentage of Cd adsorbed by shrimp. These results demonstrate that only metal bound to the soluble fraction of prey is available to higher trophic levels, and that factors influencing subcellular metal distribution in prey will directly alter metal trophic transfer to predators.     

Feeding habits of red drum (Sciaenops ocellatus) in Galveston Bay, Texas: Seasonal diet variation and predator-prey size relationships

Feeding habits, seasonal diet variation, and predator size-prey size relationships of red drum (Sciaenops ocellatus) were investigated in Galveston Bay, Texas through stomach contents analysis. A total of 598 red drum ranging from 291–763 mm total length were collected and their stomach contents analyzed during fall 1997 and spring 1998. The diet of red drum showed significant seasonal patterns, and was dominated by white shrimp (Penaeus setiferus) during fall and gulf menhaden (Brevoortia patronus) during spring. Blue crab (Callinectes sapidus) was an important component of red drum diets during both seasons. Significant differences existed between prey types consumed during fall and spring as red drum diet reflected seasonal variation in prey availability. Predictive regression equations were generated to estimate original carapace width of blue crabs from several measurements taken from carapace fragments recovered in red drum stomachs. Regressions were highly significant (r²>0.97) and increased the number of blue crabs with size information nearly three fold. Predator size-prey size relationships were determined for red drum feeding on white shrimp, gulf menhaden, and blue crab. Although regression slopes were statistically significant, prey sizes increased only slightly with increasing red drum size. Comparisons of prey sizes consumed by red drum with sizes occurring in the field indicate that red drum feed in nearshore shallow water habitats, which serve as nursery areas for many juvenile fishes and crustaceans. Our findings demonstrate that red drum feed on several prey species of commercial and recreational value and may have important effects on estuarine community structure.     

Faunal colonization of artificial seagrass plots: The importance of surface area versus space size relative to body size

An index of structural habitat complexity was devised: the average inter-structural space size within a habitat/the width of the prey organism of concern (Sp/Py). Prey survivorship should be low at Sp/Py<1 as the prey will be effectively excluded from using the habitat as refuge (they cannot maneuver through the spaces). At Sp/Py near to 1, survivorship should be high, as the spaces within the habitat are ideal for the prey and their predators are excluded (assuming they are larger than the prey). As Sp/Py increases, prey survivorship should drop rapidly until reaching a lower plateau where no predators are excluded by the structure. Sp/Py is dimensionless, and is potentially applicable across different scales and habitat types. Some of the predictions of this model were tested using artificial seagrass plots deployed in a seagrass bed in the York River, Virginia. The plots had 5 different structural treatments: control (a base with no ribbon), low, medium and high densities, as well as a heterogeneous treatment (composed of 1/3 low, medium and high density in a single treatment). The abundance of 2 mobile fauna size classes (<3.5 mm width and 3.5 to 9.5 mm width) and total species richness were compared among the different density treatments. The abundance of the smaller fauna increased with increasing density, and this response was proportional to the total surface area of the plots. The small fauna apparently did not respond to the smaller, ideal space sizes associated with the higher density plots. The larger fauna responded to the treatments as well, with the highest abundances occurring in the heterogeneous and high density treatments. The larger fauna did not respond to the structure proportional to the surface area within the plots, and its is possible that they responded to the inter-structural space sizes appropriate to their body sizes, although the results do not clearly support this conclusion. The different treatments did not affect species richness when the effect of total abundance on richness was controlled.     

Seasonal variation in food web composition and structure in a temperate tidal estuary

Seasonal variation in aquatic food web structure at Mad Island Marsh, Matagorda Bay, Texas, was examined using dietary information obtained from the analysis of gut contents from large samples of fish and crustacean specimens. Unique aspects of this study include the use of large samples of consumer gut contents (n=6,452), long-term sampling (bimonthly surveys over 18 mo), and standard methods of data collection and analysis facilitating comparisons with other aquatic food webs. Dietary data were partitioned for analysis into warm (summer) and cold (winter) seasons. Most consumers fed low in the food web, with trophic levels ranging from about 2 to 3.5 during both summer and winter. Vegetative detritus was more important in macroconsumer diets than live algae and macrophytes. Low trophic levels of consumers reflected the important role of abundant detritivores (e.g., striped mulletMugil cephalus, Gulf menhadenBrevortia patronum, and macroinvertebrates) in linking detritus to top predators via short food chains, a finding consistent with many other estuarine food web studies. Despite changes in community composition and population size structure of certain species, most food web properties revealed comparatively little seasonal variation. The summer food web had more nodes (86), more links (562), a higher density of links as indicated by connectance (0.08), and a slightly higher predator: prey ratio (0.51) compared to the winter food web (75 nodes, 394 links, connectance = 0.07, predator: prey ratio = 0.47). Proportions of top (0.06–0.07), intermediate (0.75–0.76), and basal (0.19) species did not vary significantly between seasons, but mean trophic level was higher during summer. Addition of feeding links based on information from the literature increased connectance to 0.13 during both seasons; other web parameters had values similar to those obtained for our directly estimated food webs. Seasonal variation in food web structure was influenced by changes in community composition (e.g., influxes of postlarval estuarine-dependent marine fishes during winter), availability of resources (e.g., more submerged macrophytes and amphipods during summer), and size structure and ontogenetic diet shifts of dominant consumer taxa. Our findings suggest that some basic properties of estuarine food web are resilient to seasonal changes in population and community structures and food web architecture.     

Faunal communities in seagrass beds: A review of the influence of plant structure and prey characteristics on predator-prey relationships

When compared with nearby unvergetated areas, seagrass meadows contain a dense and strikingly rich assemblage of vertebrates and invertebrates. Most recent literature has focused on evaluating the role of predation in structuring seagrass faunal communities; however, habitat complexity, abundance of food and sediment stability may also be important. This paper summarizes studies relating predator-prey relationships to different features of the seagrass system. This review suggests that the abundance of many species, both epifauna and infauna, is positively correlated with two distinct aspects of plant morphology: 1) the root-rhizome mat, and 2) the plant canopy. A scheme was developed that defines the conditions under which any particular species will be abundant or rare in a seagrass assemblage. This scheme is based on prey and predator characteristics (e.g., epifaunal vs. infaunal, tube-dweller vs. nontube dweller, burrowers vs. nonburrowers, and large vs. small as adult) and on characteristics of the seagrasses (e.g., leaf morphology, shoot density, shoot biomass, structural complexity of the meadow, and root-rhizome density and standing crop).     

Latitudinal patterns in seagrass epifauna: Do patterns exist,and can they be explained?

Seagrass-associated epifauna of several taxa constitute a major parallel element of seagrass communities over a range of latitudes. Hypotheses relating to latitudinal variation of several factors (e.g., predation, competition, primary production, habitat structure, stability and time) have been proposed to explain geographic variation in the structure of biological systems. We have summarized available information to firstly determine whether any latitudinal patterns exist for the seagrass-associated epifauna, and secondly to examine hypotheses which might explain observed patterns. Diversity and density of various seagrass epifaunal groups showed inconsistent latitudinal patterns. Diversity of decapod and amphipod crustaceans increased significantly with decreasing latitude, while diversity of isopods and fishes showed nonsignificant trends with latitude. Amphipod diversity was highly correlated with seagrass biomass over the range of latitudes. However, density of amphipods showed no pattern with either latitude or seagrass biomass; large within-site differences may have overwhelmed latitudinal patterns. For seagrass-associated amphipods, two parameters presumably related to predation intensity showed contradictory patterns. Size of individuals generally decreased toward the tropics, yet susceptibility to predators showed weak trends of increasing in the tropics. We found little support for the assumed gradients of those hypotheses proposed to explain latitudinal patterns in biota. Only a weak trend of increasing seagrass production toward the tropics was found; there was no relationship between latitude and seagrass biomass (=habitat complexity in part), epiphyte species richness (=habitat heterogeneity in part), or predator abundance. No data were available comparing actual intensity of predation on seagrass fauna or on proposed gradients of stability and competition. The patterns found were not consistent with the predictions of any single hypothesis. Contrary to evidence from other biological systems, it appears that latitude is, in general, an inconsistent predictor of differences in structure of the epifaunal component of seagrass communities. Although we did find some latitudinal patterns of increasing diversity and susceptibility to predators and decreasing size of individual amphipods toward the tropics, we were astonished by the lack of data supporting the assumptions of hypotheses concerning primary productivity, stability, time, competition, predation and habitat heterogeneity and complexity.     

Interactive effects of nutrient additions and predation on infaunal communities

Nutrient additions represent an important anthropogenic stress on coastal ecosystems. At moderate levels, increased nutrients may lead to increased primary production and, possibly, to increased biomass of consumers although complex trophic interactions may modify or mask these effects. We examined the influence of nutrient additions and interactive effects of trophic interactions (predation) on benthic infaunal composition and abundances through small-scale field experiments in 2 estuaries that differed in ambient nutrient conditions. A blocked experimental design was used that allowed an assessment of direct nutrient effects in the presence and absence of predation by epibenthic predators as well as an assessment of the independent effects of predation. Benthic microalgal, production increased with experimental nutrient additions and was greater when infaunal abundances were lower, but there were no significant interactions between these factors. Increased abundances of one infaunal taxa,Laeonereis culveri, as well as the grazer feeding guild were observed with nutrient additions and a number of taxa exhibited higher abundances with predator exclusion. In contrast to results from freshwater systems there were no significant interactive effects between nutrient additions and predator exclusion as was predicted. The infaunal responses observed here emphasize the importance of both bottom-up (nutrient addition and primary producer driven) and top-down (predation) controls in structuring benthic communities. These processes may work at different spatial and temporal scales, and affect different taxa, making observation of potential interactive effects difficult.     

Food Habits of Large Nektonic Fishes: Trophic Linkages in Delaware Bay and the Adjacent Ocean

Fish diets play a critical role in our understanding of aquatic trophic dynamics and are an important component in developing ecosystem-based approaches to fisheries management. Although large nektonic fishes exert top-down predator effects on the food web and typically support viable commercial and recreational fisheries, little is known about the diet of this guild. We evaluated the diets (6327 stomachs) of four nektonic predatory fishes (Pomatomus saltatrix [78–395 mm], Cynoscion regalis [91–520 mm], Morone americana [156–361 mm], and Morone saxatilis [82–785 mm]) in Delaware Bay and in the adjacent ocean. To assess ontogenetic, geographic, and interspecific variation in diets, observations from individual fish stomachs were clustered into species-size class groups, and dietary overlap was estimated using multivariate analyses. A shift in diet composition, as well as diversity, occurred along the estuarine gradient and into the adjacent ocean. Some prey were shared by most predators, including some crustaceans (dominated by Callinectes sapidus, mysids, and Palaemonetes spp.), fundulids (dominated by Fundulus heteroclitus), engraulids (dominated by Anchoa mitchilli), and clupeids (dominated by Brevoortia tyrannus). However, inter- and intra-specific variation in diet was observed as well. In particular, M. americana consumed fewer engraulids and clupeids, and many more and diverse types of invertebrates, while P. saltatrix consumed more clupeids and less invertebrates. The lack of overlap in diet between the four predators evaluated, and between size groups for each predator, supports previous evidence that these groups feed in trophic guilds defined by species and by size within a species. The highly variable diets for these predators suggest high resolution spatial data are necessary in order to quantify their most important prey and their role in coastal ecosystems.     

Hesperornis escapes plesiosaur attack

Evidence from the Pierre Shale (Late Cretaceous) of South Dakota is presented for an attack on a juvenile Hesperornis by a polycotylid plesiosaur. The wound healed and the Hesperornis grew to maturity. Evidence of survival provides our best information about predator prey interactions in the fossil record but are rare for birds where survival is an unlikely outcome.     

Seagrass patch characteristics alter direct and indirect interactions in a tritrophic estuarine food web

We used a mesocosm approach to examine how_ratch characteristics influenced predation and habitat selection in a tritrophic food web. Our experiments included juvenile red drum (Sciaenops ocellatus; RD), juvenile pinfish (Lagodon rhomboides; PF), and grass shrimp (Palaemonetes sp.; GS), members of a food web common in seagrass meadows of the northern Gulf of Mexico. We added an additional level of complexity to the experiment by including a predator that could feed at two different trophic levels. RD were top predators, PF were both prey items for RD and predators of GS, and GS were prey for RD and PF. We used 4 different artificial seagrass habitats that varied by size (0.049 and 0.203 m²) and shape (circular and stellate) to control for covariation between patch size and seagrass density. Predation on GS was measured in each habitat when PF, RD, and PF+RD were present, and predation on PF was measured when RD and RD+GS were present. Habitat selection by each of these 3 species was measured individually and in the presence of every other combination of the 3 species. Neither predation nor habitat selection were consistently influenced by patch characteristics (size, shape, or perimeter: area ratios) or the number of trophic levels. For GS, there was a significant negative relationship between patch size and predation rates in the GS+PF+RD treatment. Habitat selection by GS without the threat of predation suggested a preference for smaller habitats, but when in the presences of RD or RD+PF, GS preferred larger habitats. In predation experiments, PF predation by RD showed no significant relationships with patch characteristics or trophic structure. For our habitat selection experiments, PF preference was for larger habitats in the PF only and GS+PF+RD treatments. There were no significant relationships between patch size, shape, or trophic structure and RD habitat selection.