Max Christie

The College of William and Mary, VA

Ecological interactions across a late Pliocene interval of faunal turnover: naticid cannibalism north and south of Cape Hatteras, North Carolina

Previous work has indicated that over 70% of western Atlantic molluscan species went extinct during a pulsed event across the Plio-Pleistocene boundary; however, the fundamental nature of this extinction differs north and south of the biogeographic boundary represented by Cape Hatteras, NC.  North of this boundary extinction without origination is thought to have occurred while south of the boundary, paleocommunities experienced both.  Although the evolutionary effects of this extinction have been studied , the community-level ecological effects are poorly understood.  This research focuses on confamilial ecological interactions among naticid snails, a family of shell-drilling predators, and seeks to determine the degree of cannibalism, changes in predator and confamilial prey size, and changes in naticid drill hole location across the Plio-Pleistocene and biogeographic boundaries.

This study compiled float and museum collections of the Yorktown (Moore House Member) and Duplin Formations (sampling before the extinction) and the lower Waccamaw and Chowan River Formations (sampling after the first pulse of extinction).  Measured traits include: percent cannibalism per collection (cannibalized naticids/total naticids), naticid size, prey size (drilled specimens), predator size (drill hole size), and drill hole location.  Naticid size was calculated using geometric mean of length and height.  Drill hole location was determined using an eight sector grid.  To date, eight samples (4 pre-first pulse, 4 post-first pulse; 5 north of Cape Hatteras, 3 south of Cape Hatteras) have been processed totaling 447 specimens and representing nine species. 

Preliminary results suggest a non-significant decrease in percent cannibalism and a non-significant increase in average size and maximum size across the first pulse of extinction.  These results suggest that there was no difference in the confamilial ecological interactions of naticids across the first pulse of the Plio-Pleistocene extinction.  Results were also non-significant when comparing naticids within 10mm size classes.  This research is significant because it will lead to an understanding of how ecological interactions change across an extinction event and will define a baseline for naticid predator interactions across the Plio-Pleistocene boundary.

Stephen Durham

Dartmouth College
NH

Observations of Fasciolaria feeding behavior as a baseline for interpreting the trace fossil record of predation

Shells of the tulip snail Fasciolaria lilium hunteria often exhibit small (<1 cm), scallop-shaped repair scars parallel to the growth lines of the convex edge of the apertural lip. These scars have been assumed to be induced by the predators’ own feeding behavior. Wells (1958) observed that, in attacking bivalves, Fasciolaria inserts the edge of its apertural lip between the prey’s valves in order to wedge them apart, which may result in breakage to the predator’s shell.  This wedging behavior is not used in feeding on gastropods.  Wells also observed a preference for gastropod prey over bivalves, perhaps because of the risk involved in attacking bivalves. In this study, we document whether the repair scars commonly found on the shell of tulip snails result from shell wedging and reevaluate Wells’ observation of a preference for gastropod prey. 

We collected 10 live F. lilium hunteria from a shelly, sandy-mudflat with patchily distributed oyster clumps in Masonboro Sound, NC. Two tulip snails, ranging from 41 - 88 mm, were placed in five 38-liter aquaria.  Observations were conducted throughout July, 2008.  Two prey treatments were used: an oyster only treatment that contained a large clump of Crassostrea virginica (between 5 and 15 individuals, 10 – 110 mm in size); and a similar oyster clump plus an alternative gastropod prey, Urosalpinx cinerea (5 individuals, 10 – 25 mm in size).  Prey consumed were replaced daily.  We observed that repair scars on the convex edge of the apertural lip of Fasciolaria are induced by attempts to wedge open tightly closing bivalves such as oysters.  In contrast to Wells (1958), a feeding preference for gastropod prey by F. lilium hunteria was not supported.  Preliminary analysis shows the mean daily number of oysters consumed in the absence of alternative gastropod prey (0.6 oysters/day) was not significantly different from when Urosalpinx was present (0.5 oysters/day; ANOVA; F. 1.28; p= 0.26).

Our observations confirm a feeding-induced origin for the scars on shells of F. lilium hunteria, which will enable interpretation of the history of the behavior in the fossil record. However, the lack of a clear preference for gastropod prey over bivalve prey complicates assessments of the importance of these components of Fasciolaria’s diet from an analysis of the trace fossil record of feeding-induced scars alone.

Elizabeth A. Graybill

Lafayette College
PA

Age of the Duplin and Waccamaw Formations, Cape Fear River Basin, North Carolina

The age of Plio-Pleistocene stratigraphic units in southern North Carolina has been greatly debated.  Dating these strata is important for correlation to other coastal plain strata.  Two formations were studied, the Duplin and Waccamaw.  DuBar (1980) dated these units as Pliocene and early to mid Pleistocene, respectively.  Cronin et al. (1984) dated the Duplin as late Pliocene and the Waccamaw as latest Pliocene to earliest Pleistocene.  Using microfossils, Swain (1968) dated the Waccamaw as late Miocene to early Pliocene.  Campbell & Campbell (1992) claimed the youngest Waccamaw sediments are older than 2.0 Ma.  Blackwelder (1981) considered all Waccamaw deposits as early Pleistocene.  Based on the presence of molluscan taxa also found in the late Pliocene Caloosahatchee Formation of Florida, Lyons (1991) considered at least the lower Waccamaw Formation (?) as late Pliocene.

Strontium isotopic ratios were used to assign dates to the units.  Strontium is uniformly distributed in seawater and, having chemical properties similar to calcium, is incorporated in shells of organisms during growth.  Although strontium ratios in seawater have fluctuated during the Phanerozoic, specific ratios can suggest time of shell origin.

Macrofauna were hand-picked from bulk samples of the Duplin Formation collected on the Lumber River near Lumberton in Robeson Co. and the Waccamaw Formation at the Register Quarry in Columbus Co., NC.  The Waccamaw Formation? was also collected from Walkers Bluff in Bladen Co. and Sykes Landing in Columbus Co.  Nine specimens that showed minimal diagenesis and evidence of reworking were analyzed, including barnacles, Ostrea, Plicatula, and Macrocallista.  XRD analyses indicate that all Duplin samples were calcite and all Waccamaw samples aragonite.  Samples were washed in an ultrasonic bath and 0.1N HCl to remove foreign material and analyzed using a mass spectrometer at UNC Chapel Hill.

Duplin samples were dated at 2.30, 2.40, and 2.80 Ma (mid to late Pliocene).  Lower Waccamaw samples from the Register Quarry were dated at 1.55, 1.80, and 2.00 Ma (latest Pliocene to earliest Pleistocene).  As the 1.55 Ma date from the Register Quarry is not concordant with the other dates, it may be too young.  Upper Waccamaw samples from Walker’s Bluff yielded dates of 1.60 Ma (two samples) and one sample from Syke’s Landing 1.50 Ma.

Jessica Gail Lambert

SUNY College of Envir Sci & Forestry
NY

Anthropogenic Influence on the Health of Oyster Reef Ecosystems in southeastern North Carolina as Evidenced by rank-order abundance of Living and Death Assemblages

Over the last half-century Crassostrea virginica oyster reef systems subjected to eutrophication and sedimentation have shown marked decline. Estuaries subjected to a high degree of anthropogenic eutrophication had poorer live-dead assemblage fidelity than those less impacted by human influence (Kidwell 2007). We examine the impact of long-term human influences such as sedimentation and eutrophication on the health of Crassostrea virginica oyster reefs located along the southeastern coast of North Carolina.We analyze the fidelity of the death assemblages to the living community in terms of rank order abundance and species richness, employing the methodology of Kidwell (2007).  We hypothesize that increased nutrient deposition and accretion due to sedimentation from runoff has led to a significant change in rank order abundance of C. virginica and associated oyster reef fauna in these reefs. We also hypothesize that live-dead assemblage data will be more concordant in less disturbed environments than in environments with greater anthropogenic influence.

Live and dead assemblages were sampled from oyster reefs in five locations in New Hanover and Pender counties in southeastern North Carolina. All samples were gathered from environments not previously subjected to restoration efforts, shell plantings or recent commercial harvesting.  Two sites at Masonboro Island are located within a currently protected area and are defined here as “less impacted.”  Sites with greater anthropogenic influence were located on Hewletts, Pages, and Howe Creek. Three samples were taken within the oyster reefs at each site by excavating a 1m2 area to 30 cm depth. Rank order abundance and species richness data were gathered from the living and death assemblages by identifying and counting specimens gathered from each site.  Data collection has focused to date on samples from the uppermost centimeters of sediment at each site (>6100 specimens counted thus far).  All samples are dominated by Crassostrea virginica, with minor amounts of Mercenaria, barnacles, Crepidula, and mussels, and no significant differences between live and death assemblages were observed in rank order abundance of taxa at any site.  We anticipate that live and death assemblages will diverge in composition as samples from deeper within the substrate are examined.

Pedro Monarrez

California State University, Fullerton
CA

Isotope Sclerochronology of Late Pliocene Turritellidae (Gastropoda) from Southeastern North Carolina

The hypothesis that a decline in nutrients led to evolutionary turnover among mollusks during the Plio-Pleistocene in the western Atlantic has been widely proposed (e.g., Allmon et al. 1993, Todd et al. 2001). Attempts to test this hypothesis have been made in other regions of the western Atlantic, such as Florida (Lavarreda et. al. 2007), but little has been done to test this hypothesis farther north, where temperature changes may have been more important. Turritelline gastropods (Family Turritellidae) are mainly suspension feeders, and so would be expected to be particularly sensitive to changes in primary productivity. We are thus examining growth rates in Plio-Pleistocene turritellines from North Carolina to compare to results of similar studies in Florida and elsewhere. Specimens were collected from two units: the Duplin Formation (pre-turnover) and lower Waccamaw Formation (following the first turnover pulse). The Duplin consists of shelly sand, sandy marl, limestone and is of mid-late Pliocene age. The Waccamaw consists of fossiliferous clay and fine-grained sand and is latest Pliocene. We performed δ18O isotope analysis of samples of shell material from each whorl of each specimen and used whorl number vs. age plots to estimate growth rate (a conservative estimate that is somewhat independent of body size). The species analyzed were the largest complete turritelline specimens from their respective formations. Analyses were conducted on one Turritella etiwanensis? specimen (length ~ 70 mm) collected from the Duplin near Lumberton, NC. Results revealed that this specimen had a lifespan of ~ 2.5 years and a growth rate slightly lower than two turritelline species from the mid-late Pliocene Pinecrest Sand of Florida, but slightly higher than that of a recent turritelline in high-nutrient areas of the Gulf of California. Preliminary analysis of several specimens of Turritella subannulata? (average length 28 mm), collected from the lower Waccamaw near Old Dock, NC, was inconclusive, but suggestive of average lifespans of 3-3.5 years and lower growth rates. If these results are confirmed, they would indicate that turritellines from the Duplin grew faster than those from the Waccamaw, consistent with the hypothesis of a drop in productivity as a leading cause of Plio-Pleistocene molluscan turnover.

Brad Parnell

University of North Carolina Pembroke
NC

Paleoenvironments of the Duplin (Pliocene) and Waccamaw (Plio-Pleistocene?) Formations in Southeastern North Carolina inferred from microfossils

The Plio-Pleistocene was a time of faunal decimation in the western Atlantic, but the causes, timing, and magnitude of extinction and the extent of the recovery are poorly understood in the Carolinas.  In comparing macrofaunal assemblages during this time, it is necessary to investigate any fluctuations in paleoenvironments.  Microfossils such as ostracods and foraminifers can be helpful in reconstructing paleoenvironments.

Bulk samples were collected from the Pliocene Duplin Formation on the Lumber River near Lumberton, NC, and from the Waccamaw Formation at Register Quarry near Old Dock, NC (following a late Pliocene phase of faunal turnover).  In total, 900 ostracodes were picked for paleoenvironmental analysis, 300 from the Duplin and 300 from each of two beds of the lower Waccamaw at Register Quarry.  Previous work by Howard (1974) suggests that Duplin microfauna should indicate an intermediate to inner-shelfal environment; Waccamaw microfossils should represent brackish to open-marine environments.  Ostracod genera previously documented in these assemblages include Aurila, Brachycythere, Cushmanidea, Cytherura, Hulingsina, Murrayina, and Paracytheridea.  Expected foraminifers comprise Ammonia, Amphistegina, Angulogerina, Cassidulina, Cibicides, Elphidium, Globigerina, Hanzawaia, Ruessella, and Rosalina (Howard, 1974).

Sorting of ostracods is in progress, but the most common species in each sample is Aurila conradi conradi.  Other identified species include Haplocytheridea bradyi, Mutilus confragosa, Neocaudites triplistriata, and Paracytheridea vandenboldi.  Preliminary work on foraminifers indicates the presence of globigerinids, miliolids, and textulariids.  Howard (1974) distinguished 3 biofacies in both the Duplin and Waccamaw formations; future work shall compare the composition of microfossils in our samples to these biofacies.  The results should aid in understanding Plio-Pleistocene extinctions by constraining paleoenvironmental variation.

Joshua J. Poole

Elizabeth City State University
NC

Unexpected absence of cannibalism for the naticid gastropod Neverita duplicata under experimental conditions of high competition

Previous work argued that cannibalism by shell-drilling naticid gastropods is a predictable result of prey selection to maximize energy gain per foraging time (Kelley 1991).  Based on cost-benefit analysis, Kitchell et al. (1981) reported that naticids should be favored prey items; laboratory experiments indicated that cannibalism readily occurred and depended on the size ratio of two naticids, not on availability of bivalve prey.  Evidence of cannibalism is also common in the fossil record (Kelley & Hansen 2007).  Because cannibalism may be beneficial by reducing competition, we hypothesized that naticids will more likely cannibalize in high competition environments than in low competition environments in the presence of bivalve prey.

To test our hypothesis, three replicates of a high competition (HC) environment for naticids were set up with 5 Neverita duplicata (3 large, 2 small).  Three replicates of a low competition (LC) environment contained 2 Neverita (1 large, 1 small).  Small and large naticids ranged from 16-30 mm and 31-40 mm, respectively.  The following bivalve prey were provided in each tank: 10 Donax, 3 Mercenaria, and 5 Chione.  Bivalve prey ranged from ~1-3 cm, well within the limits of predator manipulation, and varied from relatively thin, smooth-shelled Donax to thick-shelled Mercenaria and Chione.  Consumed bivalves were replaced weekly with a similar prey item for the first month of study.  After this period, prey were not replaced, and prey abundance was gradually reduced by naticid predators in both the HC and LC treatments until only a few prey remained at the termination of the experiment.  The experiment was run from early July - mid November.

As predicted, no evidence of cannibalism occurred in any low competition tank.  Despite the fact that predator density was much greater than under normal conditions and that small naticids were available as prey, no evidence of cannibalism was observed in high competition tanks until October, when bivalve prey were greatly depleted.  One naticid was consumed in each HC tank, but without drilling; two naticids in one HC tank exhibited incomplete drillholes.  The lack of cannibalism in the presence of abundant bivalves and the long duration until the onset of cannibalism were unexpected and indicate that prevailing ideas about naticid cannibalism should be reevaluated.

Jennifer Toledo Rivera

University of Puerto Rico

Anthropogenic influence on the rank-order abundance of living and death assemblages of sea grass associated mollusks at Chadwick Bay, Onslow County, NC

Ecosystems are constantly undergoing changes due to human influences.  High input of nutrients, changes in salinity, dredging and drilling, boat traffic, runoff, and clamming are examples of factors that degrade water quality, causing loss of habitat (Kidwell 2007).  Such disturbances have both indirect and direct effects on sea grass ecosystems, causing habitat fragmentation (Bell et al. 2002) and influencing communities that live in them. Changes in growth rate and survival may cause discordance between the local living community of mollusks and the dead assemblages in taxonomic composition (Hubt & Nuttle 2007).  Because sea grass habitats in Chadwick Bay, Onslow Co., North Carolina, have been disturbed by these factors, we hypothesize that living and death assemblages of the molluscan community will show discordance in taxon rank-order abundance and diversity due to anthropogenic influence.

To test this hypothesis, samples of live-dead mollusk assemblages were taken from four different transects within Chadwick Bay.  Each transect was divided into three 5-m2 sampling areas, distributed from 20 meters to 100 meters from the Intracoastal Waterway.  The upper 30 cm of sediment was sampled using 25 randomly located hand cores within each sampling area.  Mollusks were extracted using a 2-mm sieve.  Following Kidwell’s approach, this research was based on a minimum of 20 live and 20 dead specimens per sample (the minimum number of specimens needed to compare rank-order abundance between live and death assemblages using Spearman’s rho). These data were collected to obtain a baseline for measuring the impact of disturbance and see if there is fidelity in taxonomic composition between the live and death assemblages. Preliminary results indicate that the death assemblages are far more diverse than the living assemblages, although a few taxa (particularly Solemya, an organic-loving species found in areas of anthropogenic eutrophication; Kidwell 2008) are found only in the live assemblages.  Two samples for which analyses are complete (Transect 3, samples B and C) yielded total diversities of 28 and 26 genera and Spearman’s rank correlation of 0.2453 and 0.3390, both nonsignificant.  This discordance in rank ordering between live and dead assemblages supports our hypothesis of significant anthropogenic impact in this area.

Kevin Selders

University of North Carolina Chapel Hill
NC

Using latitudinal differences in growth increment formation in Mercenaria mercenaria to locate paleobiogeographic boundaries in the western North Atlantic

Previous studies identified an extinction event in the western Atlantic during the late Pliocene to early Pleistocene. A latitudinal difference in the recovery of molluscan diversity followed this extinction event in Florida and Virginia (Allmon et al., 1996). Today, a latitudinal change in fauna occurs at Cape Hatteras, North Carolina. Determining the past location of this provincial boundary, which we assume was farther north during the Pliocene, is necessary to ensure the validity of latitudinal shifts in biogeographic boundaries based on comparisons of pre- and post-extinction faunas in Virginia and the Carolinas.

A latitudinal difference in the formation of seasonal growth increments in Mercenaria mercenaria shells has been documented in previous studies. Water temperature is likely the primary factor accounting for this pattern. We hypothesized that the latitudinal pattern of summer versus winter growth cessation could be used to reconstruct provincial boundaries in the fossil record. First, however, we tested whether such a method faithfully captures the known biogeographic boundary occurring at Cape Hatteras today. Oxygen isotope time series spanning at least 3 years of growth were used to identify the timing of seasonal growth cessation in M. mercenaria shells from 14 coastal localities from North Carolina (34°13’N) to Rhode Island (41°20'N). We predicted that shells north of Cape Hatteras produced winter cessation increments and those to the south formed summer cessation increments.

Of the 13 specimens for which we have results, 12 of them – spanning the full latitudinal range of our study – have δ18O values indicating periods of slower growth during the summer season. The specimen collected from Barnegat Bay, New Jersey (39°45’N), was the only shell with δ18O values indicating periods of slower growth during the winter season. These data do not support the hypothesis that a latitudinal change in M. mercenaria growth patterns occurs at the modern biogeographic boundary at Cape Hatteras. Our findings suggest that for M. mercenaria this method is not suitable for identifying the past location of provincial boundaries.