PaleoBios, Volume 14, Supplement to Number 1 1992 CALIFORNIA PALEONTOLOGY CONFERENCE ABSTRACTS LIVE SIMULATION OF AN ARCHEAN ECOSYSTEM Hartley, Julie K., Department of Earth and Space Sciences, University of California Los Angeles, Los Angeles, CA 90024 Three kinds of direct evidence contribute to current understanding of Archean life: stromatolites, microfossils, and kerogens. These data indicate that stromatolitic prokaryotic ecosystems (microbial mats) had become established by 3500 Ma ago, but because of the scarcity of ancient fossils, detailed interpretation of Archean ecology is difficult. A fourth type of evidence is available to us: reconstruction of Archean-type ecosystems in the laboratory using modern, primitive microorganisms. Such models can aid in interpreting the geologic record and thus help to decipher early evolutionary history. I am culturing a small-scale model of an Archean microbial mat in order to observe geologically preservable chemical signatures, particularly carbon isotopes, and the short-term preservation potential of the component microorganisms. The types of microorganisms chosen to make up the mat are those that may be plausibly interpreted to have been present in Archean stromatolites. All are prokaryotic and grow in modern stromatolites, such as those in Laguna Mormona, Mexico. Additionally, they have primitive metabolisms and 16S rRNA phytogenies indicate that they originated early in life's history. Evidence from all three lines of fossil data suggests that Archean life may have been evolving in an environmental regime different from that of later geologic time. Average d13C values taken from the literature are approximately the same for Archean kerogens and for kerogens of later geologic times (d13Corg = -25 %o); however, the lowest values are much lower. For example, the Tumbiana Formation in Australia contains extremely light organic matter (average d:3CDrg = -48.9 %o, n=28). Two basic kinds of environmental and ecological hypotheses have been proposed to explain these negative d13C values. First photosynthetic organisms fractionate carbon more strongly under conditions of elevated C02 partial pressure. Thus, the Archean stromatolites may have been growing in a C02-rich atmosphere. However, the maximal effect of elevated pC02 in a microbial mat is unknown. Second, methane released by methanogenic archaebacteria may be depleted in 13C by 50 %o or more relative to the carbon source. The very light isotopic values of Archean carbon may be due to recycling of biogenic methane. However, the effect of methane recycling by methylotrophic bacteria is poorly studied. Third, a combination of the above factors may have been operating in the Archean. The laboratory simulation will allow measurement of the relative contributions of pC02 and methane recycling to d13CQrg- The taphonomy of prokaryotic organisms is a matter of great uncertainty. There is widespread belief that the sheaths of filamentous prokaryotes are preferentially preserved because they are more resistant to biological degradation than are cells. I will test this hypothesis by monitoring microbial degradation at various stages of mat maturation, and at various depths in the mat. The "ecosystem" response to other environmental factors, such as UV flux, pC»2, and nitrogen starvation can also be tested, and perhaps can help improve our understanding of Archean biota. ESTIMATING ABSOLUTE SEASONAL SEA SURFACE TEMPERATURE RANGES FOR THE PAST 150,000 YEARS Billups, Katharina, and Howard J. Spero, Department of Geology, University of California, Davis, CA 95616 The stable isotope geochemistry of fossil foraminifera constitutes a widely accepted tool for the reconstruction of paleoceanographic and paleoclimatic events. The oxygen isotopic composition (d180) of fossil planktonic foraminiferal shells can be used to infer the size of continental ice-sheets and to estimate paleosea surface temperatures. Seasonal paleotemperature ranges can be determined by isotopically analyzing individual planktonic foraminifera of the same species from different levels in a deep sea core. This intraspecific isotopic variability can yield absolute summer and winter temperatures. Page 2 1992 California Paleontology Conference Abstracts Glacial and interglacial intervals from two equatorial Atlantic cores have been studied. The intervals correspond to isotope stages 1, 2, 5e, and 6 which span the last 150,000 years during the development and deglaciation of the Laurentide ice-sheet. Isotopic results from analyses of the surface dwelling foraminifera Orbulina universa, from one eastern and western equatorial Atlantic core will be presented. Differences in the oceanic regime between the two sites will be shown. Initial results show that the intraspecific oxygen isotopic variability for the western equatorial Atlantic is 1.25%o, which translates into a seasonal temperature range of 5°C. This result is consistent with results obtained from studies using faunal transfer functions. Application of this methodology promises accurate reconstructions of the paleoceanographic conditions at the selected core sites and time intervals. CHEMOSTRATIGRAPHY OF NEOPROTEROZOIC-CAMBRIAN UNITS, WHITE-INYO REGION, EASTERN CALIFORNIA: NEW CORRELATIONS BETWEEN CHINA, SIBERIA, AND THE GREAT BASIN Corsetti, Frank A., and Kaufman, Alan J., Department of Geological Sciences/Preston Cloud Research Laboratory, University of California, Santa Barbara, CA, 93106, Botanical Museum, Harvard University, 26 Oxford SL, Cambridge, MA, 02138 The first chemostratigraphic framework of a well known North American Neoproterozoic- Cambrian transition section is presented for the White-Inyo region of eastern California, showing that the transition can be constrained between the Lower and Middle Deep Spring Formation, stratigraphically higher than previously thought. A reinterpretation of the global Neoproterozoic-Cambrian carbon isotope abundance curves, coupled with the trace fossil assemblages of Crimes (1987), demonstrates that "Precambrian-Cambrian boundary" biomarker horizons in China may be younger than either the Deep Spring fauna or basal Tommotian faunas in Siberia. Therefore, the diachroneity of the small shelly assemblages should be kept in mind when making biostratigraphic correlations between Precambrian-Cambrian boundary sections. A NEARLY COMPLETE GOMPHOTHERIUM SPECIMEN FROM THE TRUCKEE FORMATION IN WEST-CENTRAL NEVADA Fitzgerald, Jennifer L., Department of Geology, University of California, Davis, CA 95616 A nearly complete specimen of the Miocene mastodont Gomphotherium has been discovered in deposits from the Truckee Formation of west- central Nevada. The specimen, a single mature individual, consists of nearly all postcranial elements and dentition, and is most closely related to Gomphotherium simpsoni from the Black Hawk Ranch fauna at Mt. Diablo, California. This conclusion is based on the stratigraphic position and overall morphological characteristics of the Truckee Formation specimen, compared with other gomphotheres; genera known to be late Clarendonian (8.5 - 12 m.y.a.) in age, commonly found with the Black Hawk Ranch fauna, are found in the Truckee Formation. This specimen is extraordinarily complete and, therefore, should be useful in future studies of the morphology and evolution of early North American proboscideans. MIDDLE PLEISTOCENE FISH FOSSILS FROM THE LAHONTAN BASIN Gobalet, Kenneth W., Department of Biology, and Robert M. Negrini, Department of Physics and Geology, California State University, Bakersfield, CA 93311 Fossils of five species of bony fishes have been recovered from Middle Pleistocene lacustrine sediments of the Humboldt River Canyon 150 km northeast of Reno, Nevada. The species have been tentatively identified as cutthroat trout (Oncorhynchus clarki; family Salmonidae), cui- ui (Chasmistes cujus; family Catostomidae), Tahoe sucker (Catostomus tahoensis; family Catostomidae), tui chub (Gila bicolor; family Cyprinidae), and Paiute sculpin (Cottus beldingi; family Cottidae). The abundant remains of vertebrae of the cutthroat trout suggest large individuals up to a meter in standard length. The sediments are dated at ages between 610,000 and 730,000 yr B.P. as evidenced by their position below the Lava Creek volcanic ash and above the Matuyama/Brunhes magnetic polarity transition. Every one of these species is currently found in the Lahontan Basin suggesting consistency of form for at least 610,000 years despite repeated 1992 California Paleontology Conference Abstracts Page 3 expansions and regressions of pluvial Lake Lahontan. This implication contradicts previous ideas published by ichthyologists who suggested that racial and specific differences of living fishes of the Great Basin have been the result of speciation since the end of the last glacial age 10,000-15,000 years before present. NEW DATA AND INTERPRETATIONS OF AN ENIGMATIC FOSSIL FROM THE MONTEREY FORMATION Hedley, John, A., University of California at Berkeley, Museum of Paleontology, Berkeley, CA 94720 New data and new specimens from the Monterey Formation near the type area have begun to shed light on the origin of the conical foraminifera- coated objects described by Hedley (1991). Although a single class of objects, all composed of the tests of the planispiral benthic foraminifer Pseudononion, was reported initially, a new locality in Carmel Valley yields the original planispiral class as well as a second class using only biserial foraminiferal tests. It is unclear whether the two classes represent two taxa of objects or simply two morphotypes of a single taxon. Measurements of sizes for both the objects and the foraminifera of which they are composed show no evidence of accretionary growth nor any correlation between length of the cone and size of the foraminifera. A strong correlation is indicated, however, between the maximum width of the cone and foraminiferal size. The overall shape of the objects also shows a correlation between maximum width and maximum length, although the number of perfect specimens from which all of the measurements can be made is small enough to suggest caution when drawing conclusions from these observations. Conclusive evidence of the nature of the organism making these objects is still lacking, but a search of candidates has been narrowed considerably. Although a species of living mollusc does use foraminiferal tests to coat its eggs, better preserved fossils show no evidence of closure at the widest end and effectively eliminate the possibility of the objects being the remains of a molluscan egg case. A better candidate may be found within the anellida, because some species of pectinariid and sabellid polychaetes are known to build tubes from foraminiferal tests and other small sedimentary particles, often demonstrating strong selectivity for particle size and shape. References: Hedley, J. A., 1991, An Enigmatic Fossil from the Monterey Formation, Carmel Valley, California (abs.), PaleoBios v.13, n.50 (supplement). THE FOSSIL HISTORY OF EPIBIONT COMMUNITIES Lescinsky, Halard, Department of Geology, University of California, Davis, CA 95616 Encrusting organisms are commonly preserved attached to the shells of other organisms. These epibionts form a simple and well-defined paleocommunity that can be traced from the Cambrian to the Recent and is suitable for studies of evolutionary patterns. I have used preliminary observations of Paleozoic and Recent epibionts and an extensive literature review to reconstruct the evolutionary history of important epibiont groups. The earliest known epibiont is a small tubular fossil of uncertain affinity from the basal Cambrian. Other Cambrian epibionts include inarticulate brachiopods and unscleratized colonial hemichordates. By the Late Ordovician, a wide variety of epibiontic organisms had evolved including encrusting edrioasteroids and other echinoderms, problematic cornulitids and a diverse bryozoan fauna. The first calcareous annelid epibionts occur during the Silurian. Articulate brachiopods occur as cemented epibionts on shells during the Carboniferous and reach high diversities during the Permian. Mesozoic epibionts beginning in the Triassic include many oysters, calcareous annelids and cyclostome bryozoans. The Cenozoic epibiont fauna consists largely of cheilostome bryozoans, barnacles and a wide variety of patellaform gastropods and anomiid bivalves. Epibiont groups common during particular time intervals in the past largely reflect patterns of world-wide diversity. Preliminary results also suggest that gross colony morphology in colonial organisms has remained fairly constant throughout the Phanerozoic and that few organisms have shifted their ecological preferences between cryptic and exposed shell surfaces. The growth of large epibiont organisms today may reflect an evolutionary trend toward faster growth rate on ephemeral surfaces, or the colonization of more sturdy, less quickly destroyed shells. Page 4 1992 California Paleontology Conference Abstracts THE DEVELOPMENT OF EARLY SHELL BEDS: EVIDENCE FROM THE CAMBRIAN AND ORDOVICIAN OF THE GREAT BASIN Li, Xing, Department of Earth Sciences, University of California at Riverside, Riverside, CA 92521 Shell concentrations (shell beds) have constituted an important and conspicuous part of the stratigraphic record since the Early Cambrian. The paleontological and stratigraphic significance of the shell beds is well understood, primarily from Mesozoic and Cenozoic examples. Lower Paleozoic shell concentrations, however, have not received much attention. The Cambrian and Ordovician evolutionary radiations were two of the most significant events in the history of life and established the Cambrian and Paleozoic faunas respectively. In order to determine the effect of these radiations on the development of shell accumulations, a systematic study of Early Paleozoic shell beds was conducted in the Great Basin areas of Utah, Nevada, and California. Shell beds were compared from strata deposited in similar depositional environments and in similar tectonic settings. Preliminary analysis of the shell beds from relatively pure carbonate facies and mixed carbonate and siliciclastic facies shows: 1) the shell concentrations became a significant stratigraphic feature in the late Early Cambrian; 2) the thickness and lateral extent of the shell beds increase from Early Cambrian to Middle Ordovician; 3) the abundance and internal complexity of the shell beds increase from Early Cambrian to Middle Ordovician; and 4) the Cambrian and Early Ordovician shell beds were primarily, if not exclusively, dominated by trilobites whereas the Middle Ordovician shell beds were dominated by brachiopods and ostracodes. These data show a temporal trend in the development of the Early Paleozoic shell beds. The nature of the Cambrian and Ordovician shell beds differs qualitatively and quantitatively. There is an increase in physical scale, abundance, and internal complexity through time. The thickness and abundance of the trilobite beds increase through the Cambrian. Interestingly, although trilobites were still diverse and abundant, they did not commonly generate thick trilobite beds after the Late Cambrian. The early Middle Ordovician is a critical time in the development of Early Paleozoic shell beds. A variety of monotaxic and polytaxic shell beds, including 6m thick composite beds, first appeared at this time. CONFIDENCE INTERVALS ON STRATIGRAPHIC RANGES FOR SECTIONS WITH NON-RANDOM DISTRIBUTIONS OF FOSSIL HORIZONS Maxshick, Patricia, Department of Earth and Space Sciences, UCLA, Los Angeles, CA 90024 The fossil record is incomplete. Species' true longevities are not reflected in the fossil record. However, Strauss and Sadler (1989) have presented a method for calculating confidence intervals for stratigraphic ranges. This procedure requires the observed stratigraphic range, the number of fossil horizons in which the taxa are found, and the confidence level desired. Increasing the confidence level has the effect of lengthening the confidence interval. Strauss and Sadler's procedure for calculating confidence intervals assumes random fossilization, or stated differently, that fossil recovery potential is uniform throughout a stratigraphic section. However, fossilization potential is not generally uniform. For example, sea level changes and tectonic activity affect rates of deposition, and therefore fossilization. Fortunately, it may be possible to resolve this problem numerically by determining confidence intervals with computer simulation. It would then be possible to calculate confidence intervals for stratigraphic sections that do not meet the criteria of random fossil recovery potential. Confidence intervals may be extended to problems beyond estimating true longevities of single taxa. For example, they may be used to distinguish between sudden and gradual extinctions. Sudden extinctions will appear gradual in a literal reading of the rock record due to the incompleteness of the fossil record. Springer (1990) provided an analytical method to distinguish between sudden and gradual extinctions using the confidence intervals developed by Strauss and Sadler. As a first step towards establishing the viability of applying numerical techniques to generate confidence intervals, computer simulations of extinction events were conducted, testing Springer's analytical method. The results 1992 California Paleontology Conference Abstracts Page 5 of these initial simulations verify that the two forms of extinction can be distinguished from each other. Since there are a limited number of stratigraphic sections where fossilization can be said to be random, relaxing randomness is an important goal for widespread application of confidence intervals. ACTUOPALEONTOLOGY OF RED SEA ECHINOIDS FROM THE NORTHERN BAY OF SAFAGA, EGYPT Nebelsick, James H., Institute of Paleontology, University of Vienna, Universitaetsstrasse 7. A-1010, Vienna, Austria Actuopaleontological investigations have been conducted on the regular and irregular echinoids fauna from the Northern Bay of Safaga (Red Sea) as part of a detailed study of the flora, fauna and sediments and facies of this area. The distribution of echinoids was deduced after identifications of quantification of fragments from standardized bulk sediment samples. The quantified fragments were compared using correlation and cluster analytic techniques. Their distributions were furthermore directly compared to sedimentological data. The results of the analysis show that echinoid distribution can be analyzed for both regular and irregular echinoid taxa although complete echinoid tests are uncommon. There is a close correlation between echinoid distributions and sedimentary and bottom facies as well as to grain size parameters and other environmental factors such as water energy. This distribution of sea-urchin fragments is not only a function of the living echinoid biocoenosis, but also that of the widely different taphonomic pathways of the diverse taxa as well as of the size limitations of the sampling method. Fragmentation of remains is an important factor. Predation by gastropods seems to be quite common. Transport is not seen to be that important a factor in the mixing of echinoid remains at the scale of the study area. The process of time-averaging can, in fact, help to counter the problems induced by taphonomic bias as well as the difficulties of quantifying living echinoid populations which often show cryptic or patchy distributions. Encrustation of echinoid tests is not common in the study area, though this process is well-known elsewhere for recent and fossil sea-urchins. PTEROSAURS AS OPTICAL ILLUSIONS: ALLOMETRY, ONTOGENY, AND THE MECHANICS OF BONE Padian, Kevin, Dept of Integrative Biology and Museum of Paleontology, Univ. of California, Berkeley, CA 94720; M.C.H. van der Meulen and Dennis R. Carter, Dept. of Mechanical Engineering, Stanford University, Stanford, CA 94305 At first glance, pterosaurs appear to have enormous skulls and forelimbs, and relatively thin and frail-looking hindlimbs. A common supposition is that they could not stand and walk on the hindlimbs, but this ignores three factors of functional morphology, phylogeny and the development of bone. First, pterosaur hindlimbs were structured and articulated like those of birds and other dinosaurs, which were upright, parasagittal bipeds (but secondary quadrupeds). Second, the outgroups to pterosaurs — dinosaurs and lagosuchids — were bipedal and digitigrade. And third, pterosaur skeletons are in large part optical illusions, because of constraints of bone mechanics associated with apparently very rapid growth. The legs of pterosaurs, compared to their torsos, are quite robust, and morphometric analyses show that the hindlimbs vary isometrically with wing size. Pterosaur forelimbs, in contrast, are hypertrophied for flight, but not just lengthwise. To maintain strength, an increase in length requires an increase in thickness of bone wall or diameter of bone shaft. Pterosaurs saved weight by decreasing bone wall thickness while they increased shaft diameter. This is why the limb disparity seems so great both in length and mass, but the bones weigh very little. They are highly vascularized and consist mostly of primary fibrolamellar bone that was quickly deposited with growth. On the other hand, the bones are very strong. A pterosaur humerus of the same length as a mammal's has 126% the diameter, one-fourth the bone wall thickness, and (despite weighing half as much) 83% of the torsional and bending strength. It was evidently able to withstand high bending and rotational loads, which is commensurate with what we know of the forces acting on the wing during flight. Therefore, although the hindlimbs appear small and spindly, they are quite in proportion with the torso, and mechanically quite strong enough to support the body on the ground. Page 6 1992 California Paleontology Conference Abstracts NEW MICROBE FOSSILS IN AMBER AND THEIR USE IN PALEOECOLOGY AND PHYLOGENY Poinar, George O., Jr., Department of Entomology, University of California, Berkeley, CA 94720; Benjamin J. Waggoner, Museum of Paleontology and Department of Integrative Biology, University of California, Berkeley, CA 94720 Fossil protists, bacteria and fungi are largely restricted to forms with substantial hard parts or forms preserved under unusual conditions. Most groups of microbes, including most non-marine microbes, have sparse or no fossil records. Such organisms, however, may be well preserved in amber, furnishing both clues to their paleoecology and direct fossil evidence of their phylogeny. Although the amber fossil record is temporally and spatially restricted, and not without its own biases, its potential usefulness for the phylogeny and paleoecology of microorganisms is only beginning to be appreciated. Eocene-Oligocene amber from mines in the El Mamey Formation of the Dominican Republic has previously yielded the first known fossil mushroom. More recently, this amber has yielded several protist fossils, including a myxomycete plasmodium with well-preserved cytoplasmic structure. Several possible ciliates and testate amoebae also appear in amber from this area. These protists are well known today from moist litters. The oldest certain amber has recently been found in the Raibler Sandstone, in the Carnian Stage of the late Triassic of the Bavarian Alps. No megafossils have yet been found in this amber, but it has so far yielded unicellular protists that appear to be ciliates of unknown affinities. This Triassic amber also contains filamentous organisms that most closely resemble modern filamentous iron bacteria such as Leptothrix or Clonothrix. Despite difficulties in precise identification, these organisms are again typical of modern wet terrestrial environments; nearly identical communities have been seen in inundated soil cultures. INTERPRETING MORPHOLOGICAL VARIANCE IN THE BIVALVE CHI ONE CANCELLATA, DURING THE LAST 3.5 MILLION YEARS Roopnarine, Peter D., Department of Geology, University of California, Davis, CA 95616 Descriptions of evolution in fossil lineages generally consist of qualitative or quantitative descriptions of patterns of variance among several OTUs. An advancement beyond these descriptions is the analysis of the association of the patterns with extrinsic variables, such as time and paleoenvironmental factors. One possible method for conducting these analyses can be chosen from the family of matrix association techniques now popular among numerical taxonomists and ecological geneticists. The discovery of a Late Neogene morphological "trend" in the venerid lineage Chione cancellata presents the opportunity to apply some of these techniques. Samples of C. cancellata were collected in Florida from the lower Pinecrest Beds (3.5-3.0 m.y.a.), Caloosahatchee Fm. (2.7-1.8 m.y.a.), Bermont Fm. (1.6-1.1 m.y.a.), Anastasia Fm. (0.05 m.y.a.) and Recent. Shell morphology was quantified with eleven morphometric variables. Selection of these variables was based on their predicted biomechanic sensitivity to changes in shell shape. Inter-sample variation was analyzed with size-adjusted canonical variates analysis. Analyses were performed at both the between- and within-formation levels. The most obvious result is a significant morphological shift between Pliocene and post- Pliocene samples. The time-based significance of the shift was tested using Mantel's (Z) test. Within-formation geographic variability was examined by analyzing the relationship between morphological dissimilarity and geographic distance using the Mantel test. Results indicate that there was a gradual development of distance-associated morphological variance during 3.5-1.1 m.y.a. The implications of this are uncertain, but may be a result of changing oceanographic conditions of the Florida peninsula during this period. 1992 California Paleontology Conference Abstracts Page 7 ECOPHENOTYPIC VARIATION IN THE SAND DOLLAR DENDRASTER EXCENTRICUS Ryan, Dallas A., Department of Geology, University of California, Davis, CA 95616 Dendraster excentricus (Echinoidea: Clypeasteroidea) is a common sand dollar found along the Pacific coast of North America from Baja, California, to southeastern Alaska. Along its geographic range, it is found in four distinct habitats: bays, tidal channels, protected outer shore, and unprotected outer shore. Dendrasterid sand dollars are unique among the clypeasteroids: their apical system is posteriorly positioned along the plane of bilateral symmetry. This aberrant position is defined as 'eccentricity', and is used to define the group taxonomically. Early studies have shown that the eccentricity of dendrasterid sand dollars is a difficult morphologic trait to use when identifying fossil taxa. Raup (1956) shows that there is a wide degree of variation in the eccentricity of populations of Dendraster living in different habitats. Stanton et al. (1979) note that fossil taxa of Dendraster exhibit a similar range in degree of eccentricity, and recognize the need for further morphometric studies. The present study is designed to quantify eccentricity on the aboral surface as it relates to habitat variation. The posterior migration of the apical system through ontogeny affects the position and lengths of the petals and the angles between them. To determine how the ambulacral system varies among individuals and among populations, twelve homologous points were chosen to characterize the aboral surface. A box truss was constructed from the points, and the box truss segments were analyzed using principal components analysis. Three populations were analyzed using this method: two populations from protected environments, and a third from an unprotected outer shore. Because the populations consisted of a wide size range of individuals, the first eigenvalue of the PCA is attributed to size. Still, generalizations about shape change through ontogeny can be made. The lengths of the petals, and the angles between them vary the most with size, while the distance between the posterior paired petals and the ambitus vary the least. This would indicate that through ontogeny, as the apical system migrates posteriorly, the paired petals lengthen and spread apart, and migrate away from the anterior petal. Eccentricities of the populations are not found to be statistically different. However, Dendraster living in unprotected outer shore habitats are significantly larger than those in protected habitats, concurrent with the findings of Stanton et al. (1979). Advanced morphometric techniques such as principal components analysis will further understanding of morphologic changes through ontogeny in living populations of Dendraster excentricus, and give taxonomists new tools for defining fossil species. APPLYING CONFIDENCE INTERVALS TO COMPOSITE RANGES: THE LONGEVITY OF THE BURGESS SHALE ANIMAL ANOMALOCARIS Saltzman, Matthew R., Department of Earth and Space Sciences, UCLA, Los Angeles, CA 90024 The notion that observed stratigraphic ranges of taxa underestimate true stratigraphic ranges is central to many issues in paleontology; the fossil record is incomplete and therefore it is difficult to resolve a taxon's longevity. Confidence intervals provide a powerful method for assessing the incompleteness of the fossil record by estimating true stratigraphic ranges based on the number of horizons known and the observed stratigraphic range. The method assumes randomly distributed fossil horizons within a randomly fluctuating, conformable facies regime. This rarely occurs for local sections due to artificial range truncations that result from unconformities in the rock record. However, the integration of biostratigraphic and chronostratigraphic data allow the pasting together of local taphonomic windows (analogous to a sort of dendrochronology) which best approximate a constant facies regime. If the composite range does not violate the assumption of randomly distributed fossil horizons, then each new horizon should gradually fill out the predicted range, as established by the confidence intervals for the first few horizons. If later finds consistently fall outside the predicted range or cluster at a single horizon within the predicted range, the assumption of nonrandomness is violated. I have assembled a composite range chart for the globally identified Cambrian genus, Anomalocaris (and related types). The composite range of Anomalocaris currently includes thirteen horizons on four continents. Page g 1992 California Paleontology Conference Abstracts Each horizon containing Anomalocaris was placed within this global range chart in the order in which it was discovered. Successive finds fell well within the predicted ranges; thus the distribution appears random. The application of confidence intervals leads to the hypothesis that Anomalocaris, and thus much of the Burgess Shale fauna for which Anomalocaris seems to be a taphonomic proxy, arose in the basal Cambrian and went extinct somewhere near the Middle-Upper Cambrian boundary, perhaps at a biomere boundary. Refinement of the biostratigraphic and chronostratigraphic controls for the Cambrian system are needed in order to explore this hypothesis. A DESCRIPTIVE COMPARISON OF ENAMEL ULTRASTRUCTURE OF FOSSIL AND RECENT ARTIODACTYLS Stevens, Kimberlee, Department of Geology, University of California, Davis, CA 95616 Among herbivorous mammals, enamel ultrastructure characteristics have not been studied adequately at the ordinal level. Because structurally complex enamel is interpreted to be a derived character, I have examined and compared ultrastructure patterns within the order Artiodactyla (eVen-toed ungulates) and interpreted these patterns with respect to present interpretations of artiodactyl phylogeny. For this study, lower first molars of individuals from eight superfamilies were cut vertically through the posterior cusp to obtain a cross-section that would most accurately illustrate three- dimensional prism behavior and Hunter-Schreger bands across the enamel and around the tooth. Three suborders, Palaeodonta, Suina and Ruminantia, make up the order Artiodactyla. The Eocene superfamily Diacodectidae of the suborder Palaeodonta has long been considered to be the earliest clade of artiodactyls; this is also reflected in the primitive character of their enamel ultrastructure. Prisms are arranged in Boyde's (1965, 1969) patterns 2 and 3, grading from one to another. Hunter-Schreger bands are absent but prism bundles display slight sinuosity. Preliminary work on suborder Suina has shown an increase in complexity from Diacodectidae. Hunter-Schreger bands are present in all taxa observed, the patterns becoming more complex over geologic time. In an Oligocene entelodontid, the Hunter-Schreger bands are irregular, inconsistent and extend to only mid-enamel from the enamel-dentine boundary. The Hemphillian Tayassuidae are similar in extent of Hunter-Schreger bands but they are more regular around the tooth, save the cusp. Here, Hunter-Schreger bands are absent but prism bundles retain a slight sinuosity. Recent Suidae contain very complex Hunter-Schreger band patterns throughout the enamel layer. At the cusp, Hunter-Schreger bands take on a cone- in-cone arrangement, sharing a common axis. Overall, the suborder Ruminantia has the most highly developed and ordered ultrastructure signature of the artiodactyls. In each of the taxa examined, Hunter-Schreger bands are present throughout the prismatic portion of enamel. Variation occurs in Hunter- Schreger band width, orientation, and shape. These variations are greatest in the Oligocene and Miocene Merycoidodontidae but lessen in Pleistocene and Recent Camelidae and are nearly nonexistent in Bovidae and Cervidae. Unique to Ruminantia is the characteristic sheet-like interprismatic material. In early ruminants the sheets of interprismatic material are somewhat undulose, filling in between prisms occupying the same column. Successively geologically younger ruminants display increasingly planar sheets of interprismatic material. ON THE GREAT STRENGTH OF MOLLUSK SHELLS IN LAKE TANGANYIKA, AFRICA: FORM, FUNCTION AND IMPLICATIONS FOR COEVOLUTION West, Kelly, Department of Earth and Space Sciences, UCLA, Los Angeles, CA 90024 The first malacologists to describe the endemic freshwater gastropods of Lake Tanganyika (East Africa) placed them in marine gastropod families. This mistake is understandable, for the Tanganyikan gastropods, with their heavily calcified shells, coarse noded ribbing, spines, apertural lip thickening and repair scars, bear a striking resemblance to marine gastropods. These features deviate markedly from the typically weakly calcified, unornamented shells of most lacustrine gastropods. The convergence between Tanganyikan and marine gastropods, however, is not just surficial. Shell crushing experiments demonstrate that the Tanganyikan thiarid and viviparid shells are, on average, an order of magnitude stronger than closely related lacustrine taxa. A number of the Tanganyikan gastropod genera are able to resist a 1992 California Paleontology Conference Abstracts force of over 1.0 kiloNewtons, well within the range of tropical marine gastropod shell strengths. SEM studies reveal that the Tanganyikan shells are primarily layers of crossed-lamellar crystal architecture (that is, needle-like aragonite crystals arranged into laths which are packed into sheets such that the aragonite needles of adjacent laths are never parallel). The number of crossed-lamellar layers can vary from one to three between different Tanganyikan gastropods species. In species with two or three crossed-lamellar layers, the orientation of the lamellae is offset by approximately 90 degrees between the different layers. Moreover, the strength of the Tanganyikan shells appears to be a direct function of the number of times the orientation of the crossed-lamellar layers changes. The high frequency of repair scars (which are not generally common among freshwater gastropods) suggested that the Tanganyikan gastropods are subject to strong predation pressure. Predation experiments with the endemic gastropods and endemic crabs of Lake Tanganyika showed that when all other shell characters are approximately equal, gastropods with one or two changes in the orientation of their crossed lamellae layers suffered significantly fewer fatal attacks than gastropods that had a single crossed lamellae layer. When compared to closely related taxa, features associated with predation both among the gastropods and crabs (shell strength and sculpture and crab chelae robustness) are extremely derived among the Tanganyikan taxa. The unusual Tanganyikan gastropod morphologies probably evolved in response to predation pressure. Robust crab morphologies may represent a concomitant adaptation to the upgrading of mollusc armor. Similar coevolving complexes have been documented from the Paleozoic and Mesozoic fossil records. The Tanganyika system, with a maximum origination time of late Miocene, offers an age constraint for the time required to form such complexes.