Small mammal bone assemblages from the scats of three families of mammalian carnivores are described here as a means of assessing paleontological bone assemblages. These are the Viverridae, Canidae, and Mustelidae. Marking behavior by the carnivores or their use of latrine areas can produce concentrations of bone comparable with those produced from owl pellets, but the nature of the bone assemblages is very different. Skeletal element preservation has a characteristic pattern related to the relative strength of the individual elements, so that the strongest elements are the ones that best survive the physical abrasion of the predator's teeth. Bone breakage is greatest in canids and least in some viverrids. Digestion of the bone occurs to a certain extent with all carnivores, with moderate rounding but little corrosion of either bones or teeth in viverrid-derived bone assemblages, very great rounding and severe corrosion of bone and tooth enamel (but little corrosion of tooth dentine) in canids, and moderate rounding and slight to moderate corrosion of bone and tooth dentine (but less of enamel) in mustelid-derived bone assemblages. Tooth marks are generally rare except in canid-derived bone assemblages. These patterns may enable the identification of predators responsible for the accumulation of fossil predator assemblages, and a consideration of the hunting and other behavior of that type of predator then permit the identification of possible biases in the fossil fauna. In the viverrids, for instance, the mongoose selected for large prey size and against cranial elements, while the genet took all prey sizes available to it and all body parts. Larger canids such as foxes and coyotes took both large and small species, but proportionally more of the most common prey species were present. Skulls were underrepresented. The small canid, the bat-eared fox, selected for the smallest prey species available to it but took all body parts. Mustelids took all sizes of prey, although not necessarily in proportion to their abundance in the habitat, and large species were underrepresented. If the predator type is known, therefore, for a fossil assemblage, these biases can be taken into account in assessing the paleoecological significance of the fauna. An example from the Olduvai Pleistocene deposits is given.

Tiering is the vertical distribution of organisms within the benthic boundary layer. Primary tierers are suspension-feeding organisms with a body or burrow that intersects the seafloor. Secondary tierers are suspension-feeders that maintain positions above or below the sediment-water interface as either epizoans on primary tierers and plants or by living in the burrows of primary tierers. Different primary tierers from soft substrata, nonreef, shallow subtidal shelf and epicontinental sea settings have had different tiering histories, resulting largely from contrasting constructional and phylogenetic constraints. Primary colonial tierers generally occupied lower epifaunal tiers during the Paleozoic and Mesozoic, but since the Cretaceous they have been dominant in the highest tier (+ 20 to +50 cm). Primary echinoderm tierers have been almost exclusively epifaunal, and from the Paleozoic through the Jurassic they were present throughout the epifaunal tiered structure. Although primary bivalve tierers have been both epifaunal and infaunal, they have occupied only lower epifaunal tiers, whereas they have adapted to all levels of the infaunal tiering structure, particularly from the late Paleozoic through the Recent. Brachiopods have lived primarily in tiers directly above or below the water-sediment interface and have not contributed significantly to tiering complexity.

Of the numerous physical and biotic processes and constraints that affect shallow marine benthos, a few have contributed more significantly to changes in tiering patterns. Trends for increasing body size could have accounted for most of the development of tiering complexity up to +50 cm and down to –12 cm. Development of tiering above +50 cm could have been due to processes which would have yielded greater feeding capability, such as competitive interactions for a place from which to feed or adaptations to velocity gradients in the hydrodynamic boundary layer. The most significant process for development of infauanl tiering below –12 cm appears to have been as an adaptive response for predator avoidance.

Unlike infaunal tiering, which never declined after it developed, epifaunal tiering has undergone a general reduction twice. Reduction in epifaunal tiering at the end of the Paleozoic appears to have been the result of the mass extinction at this time, whereas long-term biotic processes seem to have been more important for the tiering decline at the end of the Mesozoic. Tiering structure through the Phanerozoic was thus produced through interactions of a number of physical and biotic factors, tempered by constructional and phylogenetic constraints of each primary tierer group.

When the isthmian land bridge triggered the Great American Interchange, a large majority of land-mammal families crossed reciprocally between North and South America at about 2.5 Ma (i.e., Late Pliocene). Initially land-mammal dynamics proceeded as predicted by equilibrium theory, with roughly equal reciprocal mingling on both continents. Also as predicted, the impact of the interchange faded in North America after about 1 m.y. In South America, contrary to such predictions, the interchange became decidedly unbalanced: during the Pleistocene, groups of North American origin continued to diversify at exponential rates. Whereas only about 10% of North American genera are derived from southern immigrants, more than half of the modern mammalian fauna of South America, measured at the generic level, stems from northern immigrants. In addition, extinctions more severely decimated interchange taxa in North America, where six families were lost, than in South America, where only two immigrant families became extinct.

This paper presents a two-phase ecogeographic model to explain the asymmetrical results of the land-mammal interchange. During the humid interglacial phase, the tropics were dominated by rain forests, and the principal biotic movement was from Amazonia to Central America and southern Mexico. During the more arid glacial phase, savanna habitats extended broadly right through tropical latitudes. Because the source area in the temperate north was six times as large as that in the south, immigrants from the north outnumbered those from the south. One prediction of this hypothesis is that immigrants from the north generally should reach higher latitudes in South America than the opposing contingent of land-mammal taxa in North America. Another prediction is that successful interchange families from the north should experience much of their phylogenetic diversification in low latitudes of North America before the interchange. Insofar as these predictions can be tested, they appear to be upheld.

Studies of ontogenetic processes are fundamentally dependent on hypotheses of phylogeny. The model of Alberch et al. (1979) is reformulated in terms of phylogenetics and used to describe how heterochronic ontogenetic processes can be detected in nature. Heterochronic processes producing paedomorphosis can result in morphologies which resemble primitive (retained ancestral) traits; the conditions under which paedomorphic and primitive features can and cannot be distinguished are described. The utility of ontogeny for determination of evolutionary character transformations and character polarity and for detection of convergence and parallelism are considered. The ontogenetic criterion for assessing polarity is independent of hypotheses of phylogeny and may be as effective as outgroup comparison. Ontogenetic analysis may aid in the detection of convergence but not in the detection of parallelism.

The impact of environment, interspecific competition, and, to a lesser extent history, on the structure of the guild of large predatory mammals is explored in one fossil and four Recent communities. Two aspects are emphasized: (1) the number of species within each guild and (2) the extent of locomotor convergence as inferred from morphology among the constituent species. Locomotor behavior reflects habitat choice, hunting mode, and escape strategy, all of which appear to be important avenues of adaptive divergence among coexisting predators.

Locomotor behavior in extinct and extant predators is determined from body weight and five measured characteristics of the postcranial skeleton, including ungual shape, elbow shape, and limb proportions. Results indicate that levels of morphologic and inferred ecologic similarity between large predators are higher in the tropical grassland guild of East Africa than in the equivalent guilds of either tropical or temperate forest. This may be due to the great density and diversity of terrestrial herbivores in the productive grasslands. The fossil guild, from the Late Chadron–Orellan (Oligocene) of North America, appears most similar to the tropical forest guilds, but the predators seem to have been slower and more robust than their modern counterparts. Since the Orellan represents an early stage in the evolution of large, fissiped carnivores, both ancestry and time could have influenced Orellan guild structure.

Approaches to patterns of diversification based on counting taxa at a given rank can be misleading, even when all taxa are monophyletic. Such “rank-based” approaches are unable to reflect a hierarchy of evolutionary events because taxa of the same rank cannot be nested within one another. Phylogenetic trees specify an order of origination of characters and clades and can therefore be used in some cases to test hypotheses on causal relationships between characters and changes in diversity. “Tree-thinking” also clarifies discussions of the age of groups, by distinguishing between splitting of the stem-lineage from its sister group and splitting of the crown-group into extant clades.

Cladistic evidence that Pentoxylon, Bennettitales, and Gnetales are the sister group of angiosperms implies that the angiosperm line (angiophytes) existed by the Late Triassic. The presence of primitive members of five basic angiosperm clades indicates that the crown-group (angiosperms) had begun to diversify by the mid-Early Cretaceous (Barremian-Aptian), but not necessarily much earlier. The greatest unresolved issue raised by cladistic analyses concerns the fact that the angiosperm tree can be rooted in two almost equally parsimonious positions. Trees rooted near Magnoliales (among “woody magnoliids”) suggest that the angiosperm radiation may have been triggered by the origin of intrinsic traits, e.g., a fast-growing, rhizomatous habit in the paleoherb and eudicot subgroup. However, trees rooted among paleoherbs, which are favored by rRNA data, imply that these traits are basic for angiosperms as a whole. This could mean that the crown-group originated not long before its radiation, or, if it did originate earlier, that its radiation was delayed due to extrinsic factors. Such factors could be a trend from environmental homogeneity and stability in the Jurassic to renewed tectonic activity and disturbance in the Early Cretaceous. Potentially relevant pre-Cretaceous fossils cannot be placed with confidence, but may be located along the stem-lineage (stem angiophytes); their generally paleoherb-like features favor the paleoherb rooting. The history of angiophytes may parallel that of Gnetales: some diversification of the stem-lineage in the Late Triassic, near disappearance in the Jurassic, and vigorous radiation of the crown-group in the Early Cretaceous.

Cohort analysis provides an effective method of analysing taxonomic survivorship in the fossil record where large data sets are available. An analysis of the stratigraphic ranges of about 8,500 fossil genera and subgenera shows that survivorship patterns are substantially the same throughout the Phanerozoic. These patterns are used to calculate an average value for mean species duration among fossil invertebrates (11.1 Myr.). Also, the extra extinctions near the Permo-Triassic boundary are shown to be equivalent to about 85 Myr of normal, background extinction.

Global taxonomic richness is affected by variation in three components: within-community, or alpha, diversity; between-community, or beta, diversity; and between-region, or gamma, diversity. A data set consisting of 505 faunal lists distributed among 40 stratigraphic intervals and six environmental zones was used to investigate how variation in alpha and beta diversity influenced global diversity through the Paleozoic, and especially during the Ordovician radiations. As first shown by Bambach (1977), alpha diversity increased by 50 to 70 percent in offshore marine environments during the Ordovician and then remained essentially constant for the remainder of the Paleozoic. The increase is insufficient, however, to account for the 300 percent rise observed in global generic diversity. It is shown that beta diversity among level, soft-bottom communities also increased significantly during the early Paleozoic. This change is related to enhanced habitat selection, and presumably increased overall specialization, among diversifying taxa during the Ordovician radiations. Combined with alpha diversity, the measured change in beta diversity still accounts for only about half of the increase in global diversity. Other sources of increase are probably not related to variation in gamma diversity but rather to appearance and/or expansion of organic reefs, hardground communities, bryozoan thickets, and crinoid gardens during the Ordovician.

Endothermic tetrapods differ dramatically from ectothermic tetrapods in having a great capacity to sustain vigorous locomotion. I suggest that this difference reflects alternative adaptive responses to a mechanical constraint that was an inherent consequence of the vertebrate transition from aquatic to terrestrial modes of locomotion and respiration. The earliest tetrapods may not have been able to walk and breathe at the same time. Their sprawling gait and lateral vertebral bending would have required unilateral contractions of the thoracic musculature that may have interfered with the bilateral movements necessary for breathing. Modern lizards, whose locomotor and respiratory anatomy resembles that of the early tetrapods, provide support for this hypothesis because their breathing is greatly reduced during locomotor activity. Tetrapod lineages that gave rise to modern ectotherms apparently retained the constraint, becoming either highly specialized for burst activity based on anaerobic metabolism or specialized in passive mechanisms of defense against predators. The lineages from which birds and mammals are derived have undergone morphological changes that enable simultaneous running and breathing. In modern tetrapods upright posture is correlated with endothermic metabolism. This correlation may have arisen to circumvent ancestral constraints on locomotor stamina.

Evolutionary changes in herbivore abundance, diversity, and ability to excavate calcareous substrata occurred independently in three major herbivore groups: mollusks, urchins, and fish. These changes in herbivory have escalated dramatically since the mid-Mesozoic Era to the extent that herbivory today is more intense than at any time in the past.

Today, calcareous red algal crusts (i.e., corallines) are a conspicuous and frequently dominant element of shallow marine communities. Corallines and their (assumed) parent taxon, solenopores, probably always required grazing to remain free of epiphytes. The large scale evolutionary trend toward increased grazing intensity may have given the corallines a distinct advantage over the more herbivore-susceptible solenopores. Four anatomical characters unique to corallines may have facilitated their rapid expansion into shallow reef environments and subsequently (more than 100 million years later) allowed them to withstand the more intense levels of grazing that evolved in those environments. Today corallines are the only algal form to thrive under, and often even require, intensive herbivory. The extinction of solenopores and the adaptive radiation of corallines parallel a gradient of escalating herbivory over time.

Herbivores capable of excavating calcareous substrate are also frequently capable of denuding, and preventing the establishment of, stands of large leathery (kelp-like) macrophytes. It is possible that large-scale increases in herbivory have caused large-scale reduction of algal biomass in benthic communities. The presence of algal forests especially in the Paleozoic and early Mesozoic Eras may be a new variable to consider in reconstructing paleoenvironments.

It has long been suspected that trends in global marine biodiversity calibrated for the Phanerozoic may be affected by sampling problems. However, this possibility has not been evaluated definitively, and raw diversity trends are generally accepted at face value in macroevolutionary investigations. Here, we analyze a global-scale sample of fossil occurrences that allows us to determine directly the effects of sample size on the calibration of what is generally thought to be among the most significant global biodiversity increases in the history of life: the Ordovician Radiation. Utilizing a composite database that includes trilobites, brachiopods, and three classes of molluscs, we conduct rarefaction analyses to demonstrate that the diversification trajectory for the Radiation was considerably different than suggested by raw diversity time-series. Our analyses suggest that a substantial portion of the increase recognized in raw diversity depictions for the last three Ordovician epochs (the Llandeilian, Caradocian, and Ashgillian) is a consequence of increased sample size of the preserved and catalogued fossil record. We also use biometric data for a global sample of Ordovician trilobites, along with methods of measuring morphological diversity that are not biased by sample size, to show that morphological diversification in this major clade had leveled off by the Llanvirnian. The discordance between raw diversity depictions and more robust taxonomic and morphological diversity metrics suggests that sampling effects may strongly influence our perception of biodiversity trends throughout the Phanerozoic.

Three major arguments have been raised against the crucial claim, documented by Whittington and colleagues for the Burgess Shale fauna, and so contrary to traditional views, that disparity of anatomical design reached an early maximum in the history of multicellular life: (1) the presence of many early taxa with low membership and high rank is an artifact of naming; (2) cladistic analysis of Burgess arthropods negates the claim for greater early disparity; and (3) Whittington's argument is a retrospective fallacy based on assigning high rank to differentia only by virtue of their later capacity to define major branches. I show that all these arguments are either false or illogical, and that the claim for increased early disparity is justified: (1) Taxonomic rank is an artifact, but no one has ever based a claim for greater disparity on this false criterion. (2) Cladistics can only deal with branching order, whereas disparity is a phenetic issue. These two legitimate aspects of evolutionary “relationship” are logically distinct. The rooting of a cladogram only illustrates monophyletic ancestry (which no one doubts, as we are not creationists), and cannot measure disparity. (3) The active stabilization of the differentia of Baupläne (for genetic and developmental reasons only dimly understood) provides a powerful rationale for weighting these characters in considerations of disparity; nothing had so stabilized in the Burgess fauna. If these differentia were steadily changing contingencies, rather than actively stabilized features with “deep” architectural status, then the retrospective argument would be justified. Although the three arguments are wrong, the claim for greater early disparity cannot be confidently established until we develop quantitative techniques for the characterization of morphospace and its differential filling through time. This is a dauntingly difficult problem, much harder than cladistic ordering, but not intractable.

Taphonomy plays diverse roles in paleobiology. These include assessing sample quality relevant to ecologic, biogeographic, and evolutionary questions, diagnosing the roles of various taphonomic agents, processes and circumstances in generating the sedimentary and fossil records, and reconstructing the dynamics of organic recycling over time as a part of Earth history. Major advances over the past 15 years have occurred in understanding (1) the controls on preservation, especially the ecology and biogeochemistry of soft-tissue preservation, and the dominance of biological versus physical agents in the destruction of remains from all major taxonomic groups (plants, invertebrates, vertebrates); (2) scales of spatial and temporal resolution, particularly the relatively minor role of out-of-habitat transport contrasted with the major effects of time-averaging; (3) quantitative compositional fidelity; that is, the degree to which different types of assemblages reflect the species composition and abundance of source faunas and floras; and (4) large-scale variations through time in preservational regimes (megabiases), caused by the evolution of new bodyplans and behavioral capabilities, and by broad-scale changes in climate, tectonics, and geochemistry of Earth surface systems. Paleobiological questions regarding major trends in biodiversity, major extinctions and recoveries, timing of cladogenesis and rates of evolution, and the role of environmental forcing in evolution all entail issues appropriate for taphonomic analysis, and a wide range of strategies are being developed to minimize the impact of sample incompleteness and bias. These include taphonomically robust metrics of paleontologic patterns, gap analysis, equalizing samples via rarefaction, inferences about preservation probability, isotaphonomic comparisons, taphonomic control taxa, and modeling of artificial fossil assemblages based on modern analogues. All of this work is yielding a more quantitative assessment of both the positive and negative aspects of paleobiological samples. Comparisons and syntheses of patterns across major groups and over a wider range of temporal and spatial scales present a challenging and exciting agenda for taphonomy in the coming decades.

Our assessment of morphological diversity is influenced by morphological extremes and therefore depends on sample size (taxonomic richness). Rarefaction predicts the morphological diversity that would probably be observed in a sample of reduced size, thereby allowing both compensation for differences in sample size that may be strictly preservational, and analysis of diversity structure, that is, the relationship between morphological and taxonomic diversity. Middle and Late Cambrian trilobites exhibit a diversity structure characterized by many variations on a few morphological themes. In contrast, Middle and Late Ordovician trilobites occupy a larger range of morphospace per unit of species richness. Diversity structure in the Devonian is similar to that in the Middle and Late Ordovician, but the magnitude of morphological diversity is lower in the Devonian, as many fewer species are observed. For blastoids, different aspects of morphological diversity (range of morphospace occupied, number of character states possessed, and number of different regions in morphospace occupied) exhibit different relationships to taxonomic richness. In all cases Permian blastoids are characterized by a diversity structure in which morphological diversity per unit of taxonomic richness is greater than for Devonian blastoids. Changes in morphological diversity in fissiculate blastoids appear to reflect evolution of continuous variation in thecal morphology more than changes in the number of character states. Saunders and Swan's data on Namurian ammonoids illustrate some significant differences in diversity structure among stratigraphic levels, but many apparent differences in morphological diversity are consistent with the possibility that they reflect the sampling of different numbers of species from the same underlying diversity structure. Rarefaction curves are also presented for idealized increases and decreases in diversity, and these are compared to some of the observed changes in trilobites, blastoids, and ammonoids.

Ediacaran fossils are taphonomically similar to impressions of fossil plants common in quartz sandstones, and the relief of the fossils suggests that they were as resistant to compaction during burial as some kinds of Pennsylvanian tree trunks. Fossils of jellyfish are known from siderite nodules and fine-grained limestone, and even in these compaction-resistant media were more compressed during burial than were the Vendobionta. Vendobionta were constructed of materials that responded to burial compaction in a way intermediate between conifer and lycopsid logs. This comparative taphonomic study thus falsifies the concept of Vendobionta as thin soft-bodied creatures such as worms and jellyfish.

Lichens, with their structural chitin, present a viable model for the observed preservational style of Vendobionta, as well as for a variety of other features that now can be reassessed from this new perspective. The diversity of Ediacaran body plans can be compared with the variety of form in fungi, algae, and lichens. The large size (ca. 1 m) of some Ediacaran fossils is reasonable for sessile photosynthetic symbioses, and much bigger than associated burrows of metazoans not preserved. Microscopic tubular structures and darkly pigmented cells in permineralized late Precambrian fossils from Namibia and China are also compatible with interpretation as lichens. The presumed marine habitat of Ediacaran fossils is not crucial to interpretation as lichens, because fungi and lichens live in the sea as well as on land.

Elongate canines evolved independently at least four times among mammalian carnivores, and each time skulls were modified in similar ways. We have compared the cranial morphology of sabertooths to that of their non-sabertoothed relatives, living and extinct, and applied simple biomechanical models to elucidate the functional significance of the morphological differences. Our analysis suggests that (1) sabertooth morphology represents modification for wider gape with retention of a powerful bite force at the carnassial; (2) sabertooths probably used a throat or ventral neck slash to kill prey; and (3) elongate canines and retractile claws may have facilitated the exploitation of relatively larger prey by sabertooths compared to non-sabertooth carnivores.

Two intervals of the Phanerozoic stand out as times of biosphere-scale revolution in the sense that biogeochemical cycles came under increased control by organisms. These are the early Paleozoic (extending from just before the Cambrian to the Middle Ordovician, a duration of about 100 m.y.), characterized by the appearance of predators, burrowers, and mineralized skeletons, and by the subsequent diversification of planktonic animals and suspension-feeders; and the later Mesozoic (latest Triassic to mid-Cretaceous, a duration of somewhat more than 100 m.y.), marked by a great diversification of predators and burrowers and by the rise of mineralized planktonic protists. This paper explores the economic conditions that make such revolutions possible.

I argue that opportunities for innovation and diversification are enhanced when raw materials and energy are supplied at increasing rates, or when organisms gain greater access to these commodities through rising temperatures and higher metabolic rates. Greater per capita availability of resources enables populations to grow; lessens or alters ecological constraints on functional improvement; makes possible the evolution of high metabolic rates (large incomes), which in turn permit improvement in each of several otherwise incompatible functions; and favors the establishment and spread of daughter species arising through founder speciation. Reductions in productivity reinforce adaptational constraints and may bring about extinctions.

Massive submarine volcanism, together with its associated phenomena of warming, sea-level rise, and widening of warm-weather zones, is proposed to be the chief extrinsic trigger for the Phanerozoic revolutions. The later Mesozoic was characterized by continental rifting, which accompanied massive submarine volcanic eruptions that produced large quantities of nutrients and carbon dioxide. This activity began in the Late Triassic and peaked in the mid- to Late Cretaceous. The Early Cambrian was also a time of rifting and may likewise have been marked by large-scale submarine volcanism. Continental and explosive volcanism, weathering, and upwelling are other potential means for increasing evolutionary opportunity, but their effects are either local or linked directly or indirectly with cooling. Intense chemical weathering in the Early Cambrian, however, may have contributed to the early Paleozoic revolution.

The extrinsic stimulus was greatly amplified through positive feedback by the evolution of higher metabolic rates and other means for acquiring, trading, retaining, and recycling resources more rapidly and from a wider range of environments. Because these novelties usually require a high and predictable supply of resources, their evolution is more likely when extrinsically controlled supplies increase rather than when per capita availability is low.

In the view adopted here, the microevolutionary and microeconomic market forces of competition and natural selection operate against a backdrop of macroeconomic supply and demand. Resources are under both extrinsic and intrinsic control. Positive and negative feedbacks link processes at the micro- and macroeconomic levels. This view complements the genealogical and hierarchical conception of evolution by emphasizing that the pattern of descent is influenced by resources and by market forces operating at all scales of space and time.

For almost 30 years, paleontologists have analyzed evolutionary sequences in terms of simple null models, most commonly random walks. Despite this long history, there has been little discussion of how model parameters may be estimated from real paleontological data. In this paper, I outline a likelihood-based framework for fitting and comparing models of phyletic evolution. Because of its usefulness and historical importance, I focus on a general form of the random walk model. The long-term dynamics of this model depend on just two parameters: the mean (μstep) and variance (σ2step) of the distribution of evolutionary transitions (or “steps”). The value of μstep determines the directionality of a sequence, and σ2step governs its volatility. Simulations show that these two parameters can be inferred reliably from paleontological data regardless of how completely the evolving lineage is sampled.

In addition to random walk models, suitable modification of the likelihood function permits consideration of a wide range of alternative evolutionary models. Candidate evolutionary models may be compared on equal footing using information statistics such as the Akaike Information Criterion (AIC). Two extensions to this method are developed: modeling stasis as an evolutionary mode, and assessing the homogeneity of dynamics across multiple evolutionary sequences. Within this framework, I reanalyze two well-known published data sets: tooth measurements from the Eocene mammal Cantius, and shell shape in the planktonic foraminifera Contusotruncana. These analyses support previous interpretations about evolutionary mode in size and shape variables in Cantius, and confirm the significantly directional nature of shell shape evolution in Contusotruncana. In addition, this model-fitting approach leads to a further insight about the geographic structure of evolutionary change in this foraminiferan lineage.

Interpretations of vertebrate paleoecology depend on knowledge of taphonomical processes which alter the composition of the preserved fossil assemblage from that of the original community. Study of the potential fossil record of a recent mammal community in Amboseli National Park, Kenya, shows the effects of some of these biasing processes and demonstrates how a bone assemblage on a modern land surface can be a source of past and present ecological information. In the bone assemblage, species presence or absence and relative abundance differ from recorded living species occurrences and population sizes: only 74% of the extant species in the basin are identified in the bone sample, and carcass abundances vary significantly from known population sizes of the major herbivore species. Both biases appear to be strongly correlated to body size, and this results from greater destruction of bones of smaller animals within the weight range from about 1-1000 kg. This size-biasing against small species appears to be due primarily to the greater susceptibility of small bones to destruction by carnivore mastication, breakage through bioturbation (trampling), and physical and chemical processes of weathering. Size-biasing resulting from such primary processes can thus be inherited by buried bone assemblages whatever their final mode of deposition. The bone assemblage also provides information on the spatial distributions of the major herbivore species over six major habitats. Patterns of strong habitat specificity are accurately represented in the bone assemblage. However, the record for certain species is affected by their seasonal and diurnal habitat shifts so that their bone distributions do not match live census data. The Amboseli bone assemblage provides a modern analogue for taphonomical processes which may have affected fossil assemblages derived from paleo-land surfaces prior to fluvial transport. It also helps to define limits of resolution in interpreting paleoecological information from such fossil assemblages.

Body shape and mode of swimming were major factors that affected the swimming capabilities of Mesozoic marine reptiles. By estimating the total drag and the amount of energy available through metabolism, the maximum sustained swimming speed was calculated for 115 marine reptile specimens. Calculated sustained swimming speeds range from 1.8 to 2.7 m/sec, but are probably too high by as much as a factor of two. Mesozoic marine reptiles were probably much slower than modern toothed whales. The diversification of fast, agile teleost fish in the Cretaceous may have therefore contributed to the decline of the marine reptiles.

Long-bodied reptiles appear to have had slower sustained swimming speeds than deep-bodied forms of the same length. For a given length, ichthyosaurs were probably faster sustained swimmers than plesiosaurs, and plesiosaurs were probably faster sustained swimmers than crocodiles and mosasaurs. This suggests that the long-bodied forms probably used an ambush technique to capture prey, to maximize the range of possible prey and to minimize competition with the faster pursuit predators.