NAPC 2001

June 26 - July 1 2001 Berkeley, California

Abstracts, Sto - Tw

Jump to: Stockey | Strömberg | Sumida | Sumrall | Syvanen | Tang | Taylor | Terry | Terzopoulos | Teusch | Theodor | Thiel | Tiffney | Tobin | Tremaine | Tripati | Trueman | Twitchett

Return to Main Abstracts Page


STOCKEY, Ruth A., Dept. of Biological Sciences, University of Alberta, Edmonton, AB, Canada

The Princeton chert in southern British Columbia, Canada contains one of the best preserved and most diverse Eocene biotas known to date. The plants and animals found here are permineralized, many, buried in growth position. Specimens are studied using a modification of the cellulose acetate peel technique with hydrofluoric acid. The excellent preservation of delicate plant tissues has allowed the reconstruction of several plants based on organ attachments and comparative anatomy. Diversity at the site includes taxodiaceous and pinaceous conifers, large numbers of dicotyledonous angiosperms, several ferns, numerous fungi, and large numbers of monocotyledons (rare in most fossil floras). Evidence that this locality has preserved aquatics from the edges of a small lake or pond includes the presence of soft shelled turtle bones in the chert. The plants preserved here have affinities to known vascular plant families and show numerous anatomical modifications to life in water. Among these are aerenchyma tissues, protoxylem lacunae, reduced thickening on vascular tissues, and spongy phellem. In addition to the plants themselves, the chert preserves a record of mychorrizal, saprophytic and parasitic fungi associated with these plants, which adds to our knowledge of this ancient biota. Potential for the study of plant/animal interactions and whole plant reconstructions is outstanding, and while much of this material has been described in the past 20 years, there is still enormous potential for future work.


STRÖMBERG, Caroline A.E., Museum of Paleontology, University of California, Berkeley, CA, USA

In the late early Miocene (18­16 Ma) of North America, there was a radiation of equid taxa possessing hypsodont cheek teeth and limb modifications for increased cursoriality. These morphological traits were traditionally interpreted as adaptations to an early Miocene spread of open, grass-dominated habitats. In contrast, paleosol data indicate the presence of grass-dominated environments in the early Oligocene. Due to taphonomic biases in late Tertiary deposits, palynomorph and plant macrofossil data for this interval provide little independent evidence for the proposed transition from forest/woodland to savanna grasslands in the continental interior. However, siliceous plant microfossils (phytoliths), preserved in late Tertiary mammal-bearing deposits from the Great Plains, offer an opportunity to reconstruct changes in plant community structure, correlated with mammalian communities. Phytolith assemblage analysis is commonly used in Pleistocene-Holocene paleoecology for reconstructing vegetation types (forest versus savanna versus grassland). This study uses phytolith assemblage data to track vegetation changes from the late Eocene to early Miocene of northwestern Nebraska. Phytoliths were extracted using heavy liquid separation from samples of the following sedimentary units: (1) the late Eocene Big Cottonwood Creek Member of the Chadron Formation, the early Oligocene Orella Member, the ?early to ?middle Oligocene Whitney Member and middle Oligocene brown siltstone beds of the Brule Formation (White River Group); (2) the late Oligocene-early Miocene Monroe Creek Formation, the early Miocene Harrison Formation, and the early Miocene Upper Harrison beds (Arikaree Group); (3) the early Miocene Runningwater Formation, and the Dawes Clay Member of the Box Butte Formation (Hemingford Group). Results indicate that open, grass-dominated habitats have been present in northwestern Nebraska since the late Eocene, 15 Ma before the emergence of hypsodont equids.


SUMIDA, Stuart S., California State University, San Bernardino, CA, USA; David S. Berman and Amy C. Henrici, Carnegie Musuem of Natural History, Pittsburgh, PA, USA; and Thomas Martens, Museum der Natur, Gotha, Germany

A comparison of Early Permian assemblages from North America and Europe have, until recently, suggested that North American assemblages reflected aquatic to semiterrestrial to terrestrial environmental assemblages, whereas European assemblages reflected predominantly aquatic environments. More recent work at the central European Bromacker locality has demonstrated that when redbed type deposits are compared, assemblages have certain commonalties between North American and European assemblages. However, the Bromacker appears to be exclusively terrestrial, allowing the characterization of the strictly terrestrial assemblage component. Significantly, that exclusively terrestrial component of Early Permian ecosystems is dominated by the high-fiber herbivore Diadectes, with rare pelycosaurian-grade representatives. One specimen of Dimetrodon is now reported from the Bromacker, with the possibility of the presence of a varanopseid pelycosaur, and possibly another sphenacodontid. As in modern terrestrial ecosystems, top-level carnivores are rare, and herbivores are much more abundant.


SUMRALL, Colin D., Dept. of Geoscience, University of Iowa, Iowa City, IA, USA

Estimated minimum times of divergence are easily calculated from a phylogeny and known earliest stratigraphic occurrence of sister taxa. In Echinodermata, such estimates indicate that many of major clades were already distinct by the Middle to Late Cambrian. But the fossil record of these lineages begins in the Early to Middle Ordovician, begging the question, "Why the discrepancy?" Four possibilities come to mind. (1) Present fieldwork is focussing on inappropriate facies. Early and Middle Ordovician echinoderm lineages may not have inhabited the same paleoenvironments as their Middle and Late Cambrian predecessors. Indeed, in the Early Ordovician of the Great Basin much echinoderm diversity has been in found in flat pebble conglomerate, a facies that would seem unlikely to produce fossil echinoderms. (2) There is a strong taphonomic bias hiding the early diversification of echinoderms. Most early echinoderms, especially pelmatozoans, cannot be distinguished based on isolated plates. Thus, we may only know early members of lineages in the form of echinoderm debris. (3) Lineages are not distinct at the time of divergence from their sister taxon. Two clades may split at a point of cladogenesis, but with minimal morphological divergence in one or both lineages, which retain plesiomorphic morphology. This is inherently difficult to test for, as the argument would come down to a lack of evidence. (4) Present sampling is geographically biased. Much of the thorough Middle and Late Cambrian biostratigraphic work to date has concentrated on sampling trilobites. Given the quantity of research these workers have done, any echinoderms present would have been found; but field paleontologists generally find what they are looking for and little else. Despite much productive recent fieldwork, most of the Middle and Late Cambrian of the world has yet to be adequately sampled by echinoderm workers. Systematic sampling for echinoderms on a world scale and in all available facies is suggested.


SYVANEN, Michael, Dept. of Microbiology, Unversity of California, Davis, CA, USA

The current work has determined the divergence times between major eukaryotic clades based on an analysis of 18S ribosomal RNA. A trifurcation rate test is employed which renders it unnecessary to assume that the molecular clocks in the different lineages under comparison are the same. This test suggests divergence times between some of the major clades that are consistently earlier than would be suggested by the fossil record. For example, the trifurcation rate test suggests a molecular divergence time for monocots and dicots at about 175­205 MYA, while the fossil record shows that the angiosperm radiation occurred 110 MYA. Similar discrepancies are seen between molecular and paleontological estimates of divergence times when the lineages being compared include angiosperms, gymnosperms, bryophytes and some of the metazoan phyla. This discrepancy between molecular time estimates and paleontological estimates is not as extensive with the metazoa. The protosome/deuterosome split is estimated at 430 MYA or,given the accuracy of the method, a time consistent with the Cambrian radiation. There are two unexpected early divergences; the Cnidaria and Porifera diverged from the lineage leading to other metazoan phyla about 900 MYA. In addition, the clade containing Porifera also contains chytrids and choanoflagellates. The calibration point for these data assumes a trifucation of plants, fungi and metazoa of 1.05 BYA.


TANG, Carol M., California Academy of Sciences, San Francisco, CA, USA; Jeff Agnew, Arizona State University, Tempe, AZ, USA; and Laurie C. Anderson, Louisiana State University, Baton Rouge, LA, USA

The Neogene Dominican Republic marine fossils have served as the basis for many significant papers in the fields of paleobiology and evolutionary biology. Most studies, however, are focused on single taxonomic groups, either single species or assemblages of related taxa. Few studies have focused on the paleoecology of entire assemblages across class and familial lines. In this talk, we will report on preliminary taphonomic and paleoecological observations made from field work conducted in May­June 1999 and from bulk fossil samples collected at that time. Multi-taxonomic paleoecological analyses of the Neogene Dominican Republican assemblages are highly time-consuming due to the collection of bulk fossil samples. In addition, these studies are difficult due to the need for accurate taxonomic work. However, the effort is necessary for understanding the paleocommunity context for evolutionary change in individual taxonomic lineages. The sedimentology, faunal distribution, and taphonomy of the molluscs and other invertebrate fauna are evidence for changing paleoenvironmental conditions through the Gurabo section, which must be taken into account when interpreting change and stability in selected taxa. Although bivalves and gastropods are found sporadically throughout the central and southern Gurabo River outcrops, they are often associated with different dominant groups such as hermatypic and non-hermatypic corals, scaphopods, serpulids, benthic foraminifera, callianasid shrimp and brachyuran crabs. Since the bivalves and gastropods are not always even the most abundant organisms and certainly may not be the keystone species, study of the non-molluscan fraction of the assemblage will yield important imformation regarding the context of morphological and evolutionary trends documented in the section.


TAYLOR, Rod S., Dept. of Earth Sciences, University of Cambridge, Cambridge, UK

In 1912, Charles D. Walcott erected the Family Waptiidae to accommodate the new genus and species Waptia fieldensis, a bivalved arthropod from the Middle Cambrian Burgess Shale of British Columbia, Canada (the author is currently working on a detailed redescription of W. fieldensis, the first since its initial but all too brief description). Several other bivalved arthropods, similar in general appearance to Waptia, have since been discovered such as the Lower Cambrian taxa Pauloterminus spinodorsalis (nomen nudum) from the Sirius Passet fauna of North Greenland and Chuandianella ovata from the Chengjiang fauna of southwest China. Despite their overall waptiid-like appearance, however, each of these animals possesses features which suggest their apparent similarity to Waptia fieldensis may be superficial. Considerable variability in segment number, limb number and limb type between these taxa, for example, suggests that the similarities noted in these animals may in fact be convergent; these animals may not in fact be closely related.

Several other non-waptiid arthropods also possess bivalved carapaces, including the Burgess Shale taxa Canadaspis perfecta and Plenocaris plena, the Devonian animal Nahecaris stuertzi, and many others both ancient and Recent. This indicates two alternative evolutionary scenarios. First, the relatively common occurrence of bivalved carapaces may indicate a stem-group clade of bivalved arthropods in the Cambrian, united (at least) by their possession of this type of carapace, that has survived to modern times. A second, perhaps more likely possibility is the occurrence of evolutionary convergence, resulting in bivalved carapaces in multiple unrelated (or distantly related) Cambrian arthropod taxa (and in many taxa since then).


TERRY, Mark C., Science Dept., Northwest School, Seattle, WA, USA

Making evolution real to high school students challenges the resources of most schools. Contact with fossils helps, but contact with the work of paleontology by paleontologists in the field and lab is more effective.

Northwest School's senior Primate Biology course, to supplement its study of the evolution of the primates, has been using the John Day Fossil Beds National Monument as a field resource for over a decade. A week long annual trip takes students from Seattle into the region, where they visit three major Monument Units and begin to appreciate the extent of time and diversity of environments in the Cenozoic. Monument paleontology staff have made the discipline come alive for the students by demonstrating the work of the preparation lab and by training students in basic prospecting techniques.

In 1999 the Monument invited Northwest School to complete the field extraction, preparation and study of a small bone bed found by one of the students. Work on this material continues at the School, attracting the attention of younger students and supplying seniors with a potential project focus.

The John Day trips have had a lasting impact on some 200 students, and in a few cases the impact has been life changing. While many trip veterans continue in the life sciences, at least four have gone on to serious paleontology work in their undergraduate and graduate years.


TERZOPOULOS, Demetri, Dept. of Computer Science, New York University

We have created physics-based virtual worlds inhabited by biomimetic "artificial animals." Our synthetic fauna—e.g., artificial fishes in a virtual marine environment—possess biomechanically simulated bodies, sensory organs, and brains with motor, perception, and behavior centers based on ethological principles. These lifelike autonomous agents also foster a computationally oriented understanding of biological information processing, including vision, learning and cognition. For example, learning algorithms enable artificial animals to master muscle-actuated locomotion and acquire other motor skills guided by sensory perception.


TEUSCH, Kristin P., Field of Zoology, Cornell University, Ithaca, NY, USA; Warren D. Allmon, Paleontological Research Institution, Ithaca, NY, USA; and Robert Guralnick, University of Colorado at Boulder, Boulder, CO, USA

Understanding the evolution of size and shape in a lineage depends in part on knowledge of the relationship between food availability and morphological variation. This relationship may be difficult to define in the fossil record for at least two reasons: (1) Food availability is often problematic or impossible to reconstruct; and (2) morphological change, especially shape-change, may be difficult to quantify. We present an approach to evaluating the effects of major, long-term change in food supply on the shell morphology of a marine gastropod. Preliminary surveys of modern turritellids have shown that local abundance is frequently positively correlated with marine primary productivity. It is thus reasonable to predict that growth might be similarly correlated with this environmental variable. Since turritellids are suspension feeders, their growth should be dependent upon primary productivity to the degree that food is a limiting factor.

To test this hypothesis, we used methods and data from three related fields. First, in order to quantify shell morphology, we applied both linear and eigenshape methods to turritellids from the Pliocene to Recent of southern California. Next, we reconstructed productivity for the past 2.5 my based on five proxies from the paleoceanographic literature. Comparison with the morphometric analysis indicates that times of moderate to high productivity in this region are associated with larger shells and wider whorls. Periods of lower productivity are associated with smaller, narrower shells. Finally, we surveyed modern productivity conditions in several regions where turritellids occur, using satellite data, and compared a qualitative assessment of these species' morphology to our predictions for areas of high productivity. Our results indicate that food limitation is an important influence on the evolution of turritellid shell morphology.


THEODOR, Jessica M., Dept. of Organismic Biology, Ecology and Evolution, University of California, Los Angeles, CA, USA

Molecular clock estimates of divergence times for artiodactyls vary widely in their agreement with the fossil record. Recent estimates indicate that the divergence of whales from artiodactyls occurred 60 MYA, a date which compares well with the first appearances of fossil whales around 58 MYA, and artiodactyls at 56 MYA. Other estimates imply significant gaps in the fossil record. A date of 65 MYA for the divergence of Suidae and Ruminantia predates the appearance of Ruminantia by over 10 million years, and an estimate of 58 MYA for the divergence of Suidae from Cetacea implies a gap of over 20 million years. Further, although a molecular clock estimate has not been reported, the hypothesis that hippos are the closest living relatives of the whales implies a ghost lineage for hippos of over 45 million years.

There are only two living species of hippos, and their fossil record is sparse, while cetaceans and other artiodactyls are speciose and have rich fossil records. A 45 million year gap in the fossil record of hippos could be explained by several possibilities: inadequate biogeographic sampling, taphonomic biases, or undifferentiated primitive morphology. Similarly, a number of possible problems may exist in the molecular data: rate variation in the genes sampled, the low numbers of genes examined, and insufficient age calibrations. In addition, there are potential problems in molecular phylogeny estimation, such as long branch attraction and inappropriate taxonomic sampling.

Additional estimates of divergence times among living taxa should provide a broader framework for comparison with the fossil record and provide information to identify which of these factors are causing conflict.


THIEL, Diana L., and Mary L. Droser, Dept. of Earth Sciences, University of California, Riverside, CA, USA

The Ordovician radiation was a time of dramatic taxonomic and ecologic change for benthic marine faunas. Studies of fossil concentrations through this interval reveal an increase in physical dimensions and shifts in taxonomic composition. However, the details of these changes have not been previously examined and potentially provide insight into community dynamics.

The Middle Ordovician Kanosh and Lehman formations (Ibex, Utah) are particularly rich in fossil concentrations. The Lehman represents deposition in a very shallow, marine carbonate environment while the Kanosh represents deposition in a restricted basin situated on a rimmed carbonate platform. Similar taxonomic groups, namely orthid brachiopods, ostracods, echinoderms, and uncommon molluscs and trilobites, are found throughout both formations.

Although approximately 12% of the beds are truly polytaxic, nearly two thirds of the beds in both units are exclusively brachiopod (> 95% brachiopods) or brachiopod dominated (>70% brachiopods). Those that are dominantly brachiopod beds include only two or three faunal elements. Both formations are characterized by brachiopod­ostracod shell beds. While the species types vary between the units, fossil concentrations have low brachiopod diversity within beds, and typically contain only one species of brachiopod (others may occur, but are rare) with either palaeocopid or leperditocopid ostracods. While gastropods occur rarely in the brachiopod/ostracod beds, most commonly, gastropods form their own monotaxic beds. Bivalves similarly form monotaxic beds within the Lehman. Commonly, within a shell bed, a single species exhibits a wide range of sizes. This along with the lack of abrasion and other indicators of transport suggest that most of these beds represent relatively little taphonomic modification of original skeletonized assemblages.

Thus, unlike Modern and more recent counterparts, these shell beds are remarkably low in taxonomic diversity despite the dramatic increase in overall marine diversity and the relatively high formational diversity.


TIFFNEY, Bruce H., and Julie Broughton, Dept. of Geological Sciences, University of California, Santa Barbara, CA, USA

The generally indeterminate growth mode of angiosperms means that their fossil record is dominated by serially shed organs, rather than by whole organisms. Unfortunately, these various organs have not received equal research attention throughout the world. By historical accident, the Tertiary record of angiosperms was largely approached through leaves in North America, while fruits and seeds received greater attention in Europe and Asia. This has created a false impression that fruit and seed fossils are lacking in North America, which is not the case. Further, a focus on leaves alone can over-emphasize deciduous taxa and those that grow in moist locales. While fruits and seeds possess their own biases, they complement the leaf record and frequently indicate the presence of unsuspected evergreen angiosperms. We review existing North American fruit/seed literature and point the way to other known and possible locales for their discovery and description. With time and general recognition of the taphonomic circumstances that favor the preservation of fruits and seeds, carpological fossils could become as commonly described as leaves, and the possibility of linking these organs into "larger plant" if not "whole plant" reconstructions would be enhanced.


TOBIN, R., and David B. Scott, Centre for Marine Geology, Dalhousie University, Halifax, NS, Canada; and J.M. Latimer, U.S. Environmental Protection Agency, Narragansett, RI, USA

New Bedford Harbor, Massachusetts has been a major manufacturing center and fishing port over the last 250 years and as a result of this vigorous activity, the ecology and marine resources of the harbor have been severely impacted or altered. In fact, the area is so impacted that it has been classified as an EPA Superfund Site. The industrial activities surrounding New Bedford Harbor have been well documented which makes this area an attractive site to study. With these increasing pressures placed on coastal areas, new approaches to coastal zone monitoring are required. Benthic foraminifera are useful biological indicators for assessing and characterizing coastal environments because they live on and in the substrate, in contact with the surrounding water mass. It is for these reasons that benthic foraminifera were used as proxies in this study. Surface samples collected throughout the harbor were examined and the foraminiferal response shows remediation in the upper and lower harbor areas. Also there was a zone completely barren of foraminifera in and around Clarke's Outfall because of effluent being discharged. Several cores in this area show varying foraminiferal responses to different types of activities and stressors. Benthic foraminifera have responded to the construction of a hurricane barrier, a highway, and to contaminants that have been dumped into the system. One core contains a high percentage of deformed specimens that appear to correlate with the period of high PCB use in this area. These results provide one of the first records of biologic change vs. changes in chemical pollutants through time in a highly polluted site and may provide a template for cost-effective techniques to measure the biological response using foraminiferal assemblages to evaluate pollution impact responses after the fact at other sites.


TREMAINE, Robyn J., and Brian R. Pratt, Dept. of Geological Sciences, University of Saskatchewan, Saskatoon, SK, Canada; Paul A. Johnston, Royal Tyrrell Museum of Palaeontology, Drumheller, AB, Canada; and Christopher J. Collom, Mount Royal College, Calgary, AB, Canada

Faunal composition of the Middle Cambrian Burgess Shale at "The Monarch" strongly contrasts with that of the type localities at mounts Field and Stephen to the northwest. At The Monarch, shallow-water dolomites of the Cathedral Formation are cut by a megatruncation surface known as the Cathedral Escarpment, a feature continuous for nearly 100 km. A thick megabreccia and planar laminated limestones at the foot of the Escarpment are overlain by the Burgess Shale, which consists of 100 m of shaly siltstones with synsedimentary folds. An abundant fauna is present adjacent to the Cathedral Escarpment. It includes trilobites, linguliform brachiopods, hyolithids, phosphatic worm tubes, sponges and priapulid worms (Ottoia). However, this faunal assemblage is much less diverse than that of the classic Burgess Shale. A prominent difference is the absence of non-trilobite arthropods (e.g., Marrella and Anomalocaris) and many other soft-bodied elements. Abundant Ottoia at The Monarch show that appropriate taphonomic conditions for soft-bodied preservation existed at this locality and that the low diversity of Burgess Shale-type fossils there is a primary phenomenon. Furthermore, the clustering of both shelly fossils and Ottoia close to the Escarpment, and the sporadic presence of trilobites farther away from it, indicate that this faunal distribution is also primary rather than an artefact of later diagenetic overprinting.

The narrowly restricted distribution and compositional variation of soft-bodied animals in the Burgess Shale argue that this biota is not a "normal" photosynthesis-based community. Instead, the remarkable biota may have developed along the Cathedral Escarpment in response to fault-related hydrothermal seeps, and is therefore, at least in part, a chemosynthesis-based community. The observed faunal differences can then be ascribed to the variable activity of the seeps. If substantiated, this hypothesis has major implications for understanding the "Cambrian Explosion" of metazoan life.


TRIPATI, Aradhna K., and James C. Zachos, Dept. of Earth Sciences, University of California, Santa Cruz, CA, USA; Warren D. Allmon, Paleontological Research Institution, Ithaca, NY, USA; and Stephen A. Schellenberg, Institute of Marine Sciences, University of California, Santa Cruz, CA, USA

The long-term chemical evolution of seawater reflects variations in the rates of different tectonic and biological processes. Importantly, these processes also mediate atmospheric carbon dioxide levels. We have developed a method for quantifying seawater strontium/calcium (Sr/Ca) ratios in seawater using turritellid gastropod shells (Tripati et al., in prep; Tripati and Zachos, 2000). Controls on seawater Sr/Ca ratios include riverine input, hydrothermal circulation, and calcium carbonate production/dissolution.

Turritellid gastropods are abundant in the Cenozoic marine record. In addition, these organisms precipitate aragonitic shells, making diagenesis (relatively) easy to detect. As such, fossil turritellids are a unique archive of paleo-seawater chemistry and climate. We are using Sr/Ca ratios of fossil turritellid gastropods to reconstruct the long-term evolution of seawater Sr/Ca ratios through the Paleogene. In addition, we estimate paleo-seasonality of seawater temperature using the amplitude of shell Sr/Ca and oxygen isotope profiles. Paleocene and Eocene turritellids used in this study are from the Moodys Branch Formation, the Clayton Formation, the Lisbon Formation, and the Gosport Sand. In order to minimize inter-species offsets, we are using members of one lineage that spans the Cenozoic. The temperature dependence of the partition coefficient has been established using (modern) specimens of Turritella exoleta (Tripati et al., in prep.).

Using Sr/Ca data for a Middle Eocene shell, we estimate a paleo-seasonality of 9°C ± 2°C. This is consistent with independent proxy reconstructions for the Gulf Coast (Andreasson and Schmitz, 2000; Ivany et al., 2000; Green wood and Wing, 1995). We interpret high shell Sr/Ca values to indicate Middle Eocene seawater Sr/Ca ratios were much higher than modern, implying that substantial changes in oceanic strontium and calcium cycling have occurred over the past ~45 million years.


TRUEMAN, C.N., and A.K. Behrensmeyer, Dept. of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington, DC, USA; R. Potts, Dept. of Anthropology, National Museum of Natural History, Smithsonian Institution, Washington, DC, USA; and N. Tuross, Smithsonian Center for Materials Research and Education, Maryland, MD, USA

Modern ecosystems can be viewed as collections of ecologically discrete microenvironments that differ from one another with subtle changes in many parameters including hydrology, organic content, Eh/pH conditions, and microbial ecology. The diversity of these microenvironments is a fundamental character of an ecosystem affecting such processes as ecological responses to environmental change. Unfortunately, it is difficult to reconstruct meter-scale environmental variability across ancient landscape surfaces because few paleoenvironmental proxies have been developed with sufficient resolution.

In this study, we use the post-mortem enrichment of trace elements (REE, U, Th) in bone as a proxy for local environmental conditions. The trace element composition of a fossil bone reflects local hydrological and chemical factors. In a sample of >400 bones from two landscape surfaces from the Pleistocene succession at Olorgesailie, Kenya, we demonstrate meter-scale heterogeneity in fossil bone chemistry that can be related directly to differences in local hydrological conditions. In addition to reconstructing chemical variability across landscapes, these data can be used to interpret the taphonomic history ofl vertebrate accumulations.

The implications of the study are: (1) diagenetic trace element signals in fossil bones can be used as a measure of high resolution landscape variability, (2) temporal or spatial differences in absolute chemistry or chemical heterogenity may be related to ecological or taphonomic factors of particular paleolandscape surfaces and, (3) paleoenvironmental or paleoclimate reconstructions based on low sample sizes may not be representative of "average" conditions, as microenvironments may develop a discrete chemistry and may retain this chemistry for long periods of time.

In summary, high resolution, large sample studies across paleolandscapes provide much more detailed paleoenvironmental information than is currently used, and thus allow a more holistic picture of ancient landscape ecology and chemical taphonomic signatures in fossil bones.


TWITCHETT, Richard J., Dept. of Earth Sciences, University of Southern California, Los Angeles, CA, USA The diversity of fossil taxa fell sharply during the Permian-Triassic (P-Tr) and Triassic-Jurassic intervals, with a moderate recovery in between. These periods of low diversity are characterized by increasing numbers of Lazarus taxa, and hence a decrease in the completeness of the fossil record. For example, a mere 10% of sponge families known to have been present in the Early Triassic are represented by actual fossils, compared to 50­70% in the Late Permian. This pattern applies to all benthic groups so far studied. These data suggest that our perception of the magnitude of these events is certainly affected by the quality of the fossil record.

The periods of low apparent diversity are also characterized by a decrease in the size of the fossil organisms (the Lilliput Effect). Size decrease is most likely a response to low levels of primary production. During the P-Tr interval, global warming led to oceanic stagnation and stratification and hence a reduction in upwelling, nutrient recycling and primary production. When food supply is dramatically reduced, animal biomass must also be reduced. Individual taxa can achieve this by reducing population size or reducing body size. In the former case, such taxa will be more prone to real extinction and will be less likely to enter the fossil record (apparently becoming extinct). In the latter case, numbers are maintained and the taxa escape both real and apparent extinction. Lazarus taxa are therefore those taxa that reduced population size, and managed to escape real extinction, until productivity levels improved and numbers could increase once more. Thus, changes in climate, ocean circulation and productivity can affect the quality of the fossil record and the size of fossil organisms preserved. This model is testable and can be applied to terrestrial as well as marine events.