Because predation by drilling gastropods is uniquely preservable in the fossil record, it represents important evidence for the study of coevolution. Previous studies of drilling gastropod predation have been largely descriptive and sometimes contradictory. We formulate and test a model of prey selection by naticid drilling gastropods. The model adequately predicts both prey species selection and prey size selection. Prey preferences parallel prey profitabilities, determined by calculating prey species-specific and predator size-specific cost-benefit functions. The model also specifically suggests the evolution of potential refugia from predation and the evolution of potential predatory attributes. Application of the model to several Miocene and Pliocene assemblages studied by Thomas (1976) corroborates the feasibility and utility of this approach in examining the evolutionary record of naticid predation, which extends from the Late Mesozoic. Apparent evolutionary stasis and convergent morphological trends among prey species may be consistent with continuous selection pressures against predation.

The presumed geometry of clam and brachiopod clades (brachiopod declines matched closely by clam increases) has long served as primary data for the classic case of gradual replacement by competition in geological time. Agassiz invoked the geometric argument to assert the general superiority of clams, and it remains the standard textbook illustration today. Yet, like so many classic stories, it is not true. The supposed replacement of brachiopods by clams is not gradual and sequential. It is a product of one event: the Permian extinction (which affected brachiopods profoundly and clams relatively little). When Paleozoic and post-Paleozoic times are plotted separately, numbers of clam and brachiopod genera are positively correlated in each phase. Each group pursues its characteristic and different history in each phase—clams increasing, brachiopods holding their own. The Permian extinction simply reset the initial diversities. The two groups seem to track each other in each phase and a plot of brachiopod vs. clam residuals (each from their own within-phase regressions against time) yields significantly positive association. Some of this tracking may be an artifact of available rock volumes; we could, however, detect no effect of stage lengths. Passive extrapolation of microevolutionary theory into the vastness of geological time has often led paleontologists astray. Competitive interaction may rule in local populations, but differential response to mass extinctions (surely not a matter of conventional competition) may set the relative histories of large groups through geological time. Similarly, adaptive superiority in design cannot, in the usual sense of optimal engineering, have much to do with the macroevolutionary success of clams. The interesting question lies one step further back: what in the inherited Bauplan of a clam permits flexibility in design and why are other groups, however successful in their own domain, unable to alter their basic design.

Mass extinctions are important to macroevolution not only because they involve a sharp increase in extinction intensity over “background” levels, but also because they bring a change in extinction selectivity, and these quantitative and qualitative shifts set the stage for evolutionary recoveries. The set of extinction intensities for all stratigraphic stages appears to fall into a single right-skewed distribution, but this apparent continuity may derive from failure to factor out the well-known secular trend in background extinction: high early Paleozoic rates fill in the gap between later background extinction and the major mass extinctions. In any case, the failure of many organism-, species-, and clade-level traits to predict survivorship during mass extinctions is a more important challenge to the extrapolationist premise that all macroevolutionary processes are simply smooth extensions of microevolution. Although a variety of factors have been found to correlate with taxon survivorship for particular extinction events, the most pervasive effect involves geographic range at the clade level, an emergent property independent of the range sizes of constituent species. Such differential extinction would impose “nonconstructive selectivity,” in which survivorship is unrelated to many organismic traits but is not strictly random. It also implies that correlations among taxon attributes may obscure causation, and even the focal level of selection, in the survival of a trait or clade, for example when widespread taxa within a major group tend to have particular body sizes, trophic habits, or metabolic rates. Survivorship patterns will also be sensitive to the inexact correlations of taxonomic, morphological, and functional diversity, to phylogenetically nonrandom extinction, and to the topology of evolutionary trees. Evolutionary recoveries may be as important as the extinction events themselves in shaping the long-term trajectories of individual clades and permitting once-marginal groups to diversify, but we know little about sorting processes during recovery intervals. However, both empirical extrapolationism (where outcomes can be predicted from observation of pre- or post-extinction patterns) and theoretical extrapolationism (where mechanisms reside exclusively at the level of organisms within populations) evidently fail during mass extinctions and their evolutionary aftermath. This does not mean that conventional natural selection was inoperative during mass extinctions, but that many features that promoted survivorship during background times were superseded as predictive factors by higher-level attributes. Many intriguing issues remain, including the generality of survivorship rules across extinction events; the potential for gradational changes in selectivity patterns with extinction intensity or the volatility of target clades; the heritability of clade-level traits; the macroevolutionary consequences of the inexact correlations between taxonomic, morphological, and functional diversity; the factors governing the dynamics and outcome of recoveries; and the spatial fabric of extinctions and recoveries. The detection of general survivorship rules—including the disappearance of many patterns evident during background times—demonstrates that studies of mass extinctions and recovery can contribute substantially to evolutionary theory.

Actualistic experiments have quantified rate of anaerobic decay and associated mineralization around proteinaceous macro-organisms. Carcasses of the polychaete worm Nereis and the eumalacostracans Nephrops and Palaemon were buried in airtight glass jars filled with sediment and water from marine, brackish, and lacustrine environments. Over a period of 25 weeks the contents were examined to determine the state of decay and were chemically analyzed to monitor early diagenetic mineralization (two methods for such analysis are reviewed). Decay processes were active in the experimental conditions despite anoxia and had virtually destroyed the carcasses within 25 weeks. However, decay-rate in the sulfate-reducing marine system was greater than in the methanogenic freshwater environments. Petrological and geochemical analyses of the organic remains identified discrete layers of authigenic iron monosulfide (a pyrite precursor) on the surface of the decaying Nephrops cuticle within weeks of initiating the experiment. Chemical analysis of decomposing flesh showed a marked increase in pore-water calcium content with time.

The results clearly show that anoxia is ineffective as a long-term conservation medium in the preservation of soft-bodied fossils. However, decay-induced mineralization can be very rapid so that even a slight reduction in decay rate can lead to improved levels of fossil preservation. Traditionally, stagnation and rapid burial are considered to be the main prerequisites for the preservation of soft-bodied fossils and the formation of Konservat-Lagerstätten. Clearly these factors are only important in that they promote early diagenetic mineralization. This is the only way to halt information loss through decay.

It has been argued that many clades originating in the early Paleozoic filled their design space rapidly while still at low taxonomic diversity. Standardization of morphology for analytical purposes facilitates testing of this claim. Here I document evolutionary patterns of morphological disparity in Ordovician-Devonian crinoids, using a set of 75 discrete characters covering the principal features of the crinoid stem, cup, tegmen, and arms. Disparity is measured as the average dissimilarity among species, the range of morphospace occupied, and the number of realized character-state combinations. Comparison with generic richness reveals that the full range of form was essentially attained by the early part of the Caradocian, long before the time of maximal taxonomic diversity. Despite subsequent taxonomic diversification, the variety of crinoid form did not expand appreciably; increased diversity was accommodated by the evolution of variations upon the spectrum of designs established earlier. The data discussed here do not definitively imply specific sources of constraint, but the effective stasis in disparity supports previous arguments that some morphological limits were reached early in crinoid history.

Convergent evolution of hypercarnivorous adaptations in canids has occurred a number of times in the last 40 m.y. among distantly related taxa. The adaptations include an increase in carnassial blade length, reduction or loss of post-carnassial molars, and transformation of the talonid of the lower first molar from a basinlike depression into a trenchant, bladelike cusp. Although the diversity of these specialized canids is typically low in past and present communities, it was unusually high during the Late Oligocene of North America and the Pleistocene of South America. These two comparable events provide an opportunity for exploring possible causes of the evolution of hypercarnivory in canids. Plots of generic diversity against time for North American predators reveal a roughly inverse relationship between the number of hypercarnivorous canid taxa and the numbers of other hypercarnivores, such as creodonts, nimravids, mustelids, and amphicyonids. Similarly, the radiation of hypercarnivorous canids in South America occurred at a time of relatively low diversity of other hypercarnivores. Analysis of trophic diversity within the North American carnivore paleoguild before, during, and after the Late Oligocene reveals considerable taxonomic turnover among carnivores because of immigration and speciation. Late Oligocene hypercarnivorous canids appear to have been replaced first by amphicyonids and large mustelids, and then by felids.

Despite the repeated tendency of canids to evolve adaptations for hypercarnivory, a canid has yet to appear that is completely catlike, that is, without any post-carnassial molars. This possible constraint on morphological evolution in canids is argued to have resulted, paradoxically, in increased flexibility over evolutionary time and a great potential for rapid diversification and clade survivorship. Finally, it is suggested that the iterative pattern of specialization of the lower molars for meat-slicing that is seen in all families of carnivores, past and present, is probably a result of intraspecific competition for food, perhaps among littermates. This intraspecific selective force is countered by competition among species, since there are limits on the number of sympatric hypercarnivorous species within a single community.

Four basic types of skeletal concentrations are modeled in terms of changes in sedimentation rate alone. The model categorizes fossil concentrations on the relatively objective basis of their bed contacts, and uses this criterion to infer directional shifts in net sedimentation. This radical simplification of accumulation histories, in which hardpart input is held constant, yields a surprisingly powerful model capable of predicting a broad spectrum of taphonomic and paleobiologic phenomena. Type I concentrations grade from less fossiliferous sediments and terminate in omission surfaces; if hardpart supply is held constant, they record a slowdown from positive to zero net sedimentation. Type II concentrations are the same as Type I but terminate in erosion surfaces (slowdown to negative net sedimentation), and Type III and IV concentrations are characterized by basal erosion or omission surfaces, respectively, grade upward into less fossiliferous sediments, and record increases in net sedimentation from negative or zero rates to positive rates. According to the model, samples collected from successive horizons within any of these shell beds will differ in the degree and type of post-mortem bias owing to differing histories of hardpart exposure at the depositional interface. Moreover, because rates of sediment accumulation govern the abundance of hardparts at the depositional interface and thus many of the physical characteristics of the benthic habitat, the dynamics of fossil accumulation have direct consequences for the structure of benthic communities (taphonomic feedback) and for ecologically controlled species morphometry.

The model is highly robust to fluctuations in hardpart input, as judged by its ability to correctly infer modes of formation of concentrations in synthetic stratigraphic sections. In addition, field examples of Type I–IV concentrations show independent evidence of formation during intervals of reduced net sedimentation, and many exhibit trends in taphonomic and paleobiologic features expected from the postulated changes in net sedimentation. The model thus provides a testable working hypothesis for the accumulation of fossil material in a wide range of environments, and should be applicable to concentrations of any taxonomic composition, state of preservation, or geologic age. The power and robustness of this heuristic model in fact argue that fossil-rich and fossil-poor strata provide fundamentally different records of past conditions, and that sedimentation rather than hardpart input is the primary control on the nature of the fossil record.

The evolutionary pattern of speciation and extinction in any biologic group may be described by a variety of mathematical models. These models provide a framework for describing the history of taxonomic diversity (clade shape) and other aspects of larger evolutionary patterns. The simplest model assumes time homogeneity: that is, speciation and extinction probabilities are constant through time and within taxonomic groups. In some cases the homogeneous model provides a good fit to real world paleontological data, but in other cases the model serves only as a null hypothesis that must be rejected before more complex models can be applied. In cases where the homogeneous model does not fit the data, time-inhomogeneous models can be formulated that specify change, regular or episodic, in speciation and extinction probabilities. An appendix provides a list of the most useful equations based on the homogeneous model.

The Quaternary ice ages were paced by astronomical cycles with periodicities of 20–100 k.y. (Milankovitch cycles). These cycles have been present throughout earth history. The Quaternary fossil record, marine and terrestrial, near to and remote from centers of glaciation, shows that communities of plants and animals are temporary, lasting only a few thousand years at the most. Response of populations to the climatic changes of Quaternary Milankovitch cycles can be taken as typical of the way populations have behaved throughout earth history. Milankovitch cycles thus force an instability of climate and other aspects of the biotic and abiotic environment on time scales much less than typical species durations (1–30 m.y.). Any microevolutionary change that accumulates on a time scale of thousands of years is likely to be lost as communities are reorganized following climatic changes. A four-tier hierarchy of time scales for evolutionary processes can be constructed as follows: ecological time (thousands of years), Milankovitch cycles (20–100 k.y.), geological time (millions of years), mass extinctions (approximately 26 m.y.). “Ecological time” and “geological time” are defined temporally as the intervals between events of the second and fourth tiers, respectively. Gould's (1985) “paradox of the first tier” can be resolved, at least in part, through the undoing of Darwinian natural selection at the first tier by Milankovitch cycles at the second tier.

In several increasingly realistic steps, a model of the stratigraphic distribution of fossils is presented. The first and simplest step assumes that if a taxon was extant it will have been preserved. The second step admits that if a taxon was extant, there is some probability less than one that it will have been preserved. This step produces randomly distributed gaps, and fossil ranges that are somewhat shorter than the span of time in which a taxon actually lived. The third step assumes facies-controlled taxa and parasequence-style cyclicity. This produces randomly and nonrandomly distributed gaps, including the anomalously long gaps recognized in confidence limit studies. The final model incorporates depositional sequences and indicates that first and last occurrences will cluster at sequence boundaries and at flooding surfaces in the transgressive systems tract. Across-shelf gradients in diversity, taphonomy, or eurytopy can control the strength, but not the stratigraphic position of these peaks. Comparison of the model to data from the Upper Ordovician suggests that these modeled features are present in the fossil record. Many previously studied paleobiologic patterns may be, at least in part, an artifact of facies control and sequence architecture. Such patterns include gradual or stepwise mass extinction, punctuated morphologic and taxonomic change, iterative evolution, and the replacement of shallow water faunas by deeper water faunas at biomere boundaries.

Trophic diversity within guilds of terrestrial predators is explored in three modern and two ancient communities. The modern communities span a range of environments including savannah, rainforest, and temperate forest. The paleocommunities are North American, Orellan (31–29 Ma), and late Hemphillian (7–6 Ma), respectively. The predator guilds are compared in terms of: 1) species richness; 2) the array of feeding types; and 3) the extent of morphological divergence among sympatric species. Feeding type is determined from dental measurements that reflect the proportion of meat, bone, and non-vertebrate foods in the diet. Measurements include estimates of canine shape, tooth size, cutting blade length, and grinding molar area. Morphological divergence among sympatric predators is measured by calculating Euclidean distances among species in a six-dimensional morphospace. Results indicate that the number of predator and prey species are roughly correlated in both ancient and modern communities. Two of the predator guilds, the late Hemphillian and modern Yellowstone, contain relatively few species and appear to be the result of extinction without replacement. Despite differences in history, age, and environment, the extent of morphological divergence within guilds does not differ significantly for the sampled communities. It is clear that the basic pattern of adaptive diversity in dental morphology among coexisting carnivores was established at least 32 million years ago. It appears that interspecific competition for food has acted similarly to produce adaptive divergence among sympatric predators in communities that differ widely in time, space, and taxonomic composition.

Delayed maturation and growth to large sizes are only advantageous under stable environmental conditions where food resources are limited. Specialization to algal symbiosis is also highly advantageous under those conditions if sunlight is available. The coevolution of these two characteristics has occurred many times in many foraminiferal lineages. These traits are sometimes associated with increased embryon size and suppression of sexual reproduction, which are also characteristics most advantageous under stable environmental conditions. Specialization for these traits, ensuring success in warm, shallow, stable, oligotrophic environments, often dooms the species or lineage to extinction when conditions change.

Twelve different bones from the skeleton of the holotype specimen of the hadrosaurian dinosaur Hypacrosaurus stebingeri were thin-sectioned to evaluate the significance of lines of arrested growth (LAGs) in age assessments. The presence of an external fundamental system (EFS) at the external surface of the cortex and mature epiphyses indicate that the Hypacrosaurus specimen had reached adulthood and growth had slowed considerably from earlier stages. The number of LAGs varied from none in the pedal phalanx to as many as eight in the tibia and femur. Most elements had experienced considerable Haversian reconstruction that had most likely obliterated many LAGs. The tibia was found to have experienced the least amount of reconstruction, but was still not optimal for skeletochronology because the LAGs were difficult to count near the periosteal surface. Additionally, the numbers of LAGs within the EFS vary considerably around the circumference of a single element and among elements. Counting LAGs from a single bone to assess skeletochronology appears to be unreliable, particularly when a fundamental system exists.

Because LAGs are plesiomorphic for tetrapods, and because they are present in over a dozen orders of mammals, they have no particular physiological meaning that can be generalized to particular amniote groups without independent physiological evidence. Descriptions of dinosaur physiology as “intermediate” between the physiology of living reptiles and that of living birds and mammals may or may not be valid, but cannot be based reliably on the presence of LAGs.

Modern ruminant species have traditionally been placed in three broad dietary categories (browsers, grazers and intermediates) based on their observed feeding habits. Shape analysis of premaxillary outlines of 31 species of ruminants shows that their premaxillae differ according to their dietary category. Browsers have pointed premaxillae and grazers square ones. Intermediate feeders have intermediate outlines.

The Miocene giraffid Samotherium boissieri has always been viewed as a specialized browser similar to the modern okapi, Okapia johnstoni. However, the premaxillary shape of S. boissieri falls very close to the mean of the grazers and is most similar to that of the African buffalo (Syncerus caffer), a committed grazer.

Quantitative analyses of the microscopic wear patterns on the molars reveal significant differences between three modern species from the three dietary groups. S. boissieri has more microscopic scratches on its teeth than either the modern giraffe Giraffa camelopardalis (a browser) or Grant's gazelle Gazella granti (an intermediate feeder). In this respect, it is indistinguishable from the wildebeest Connochaetes taurinus which is a committed grazer.

Both of these analyses suggest that this extinct giraffid was a grazer, although we cannot rule out the possibility that it was an intermediate feeder. It was definitely not a specialized browser as are both living members of the Giraffidae.

Organic preservation of non-mineralizing animals constitutes an important part of the paleontological record, yet the processes involved have not been investigated in detail. Organic-walled fossils are generally explicable as a coincidence of original, relatively recalcitrant, extra-cellular materials and more or less anti-biotic depositional circumstances. One of the most pervasive natural inhibitors of biodegradation results from substrate and enzyme adsorption onto, and within, clay minerals; such interactions are likely responsible for many of the organic-walled fossils preserved in clastic sediments. Close examination of the fossil Lagerstätte of the Burgess Shale (Middle Cambrian, British Columbia) reveals that most of its so-called soft-bodied fossils are composed of primary (although kerogenized) organic carbon. Their preservation can be attributed to pervasive clay-organic interactions as the organisms were transported in a moving sediment cloud and buried with all cavities and spaces permeated with fine grained clays. The organic-walled Burgess Shale fossils were studied both in petrographic thin section and isolated from the rock matrix, following careful acid maceration. Isotopic analysis of bulk organic and carbonate carbon yielded values consistent with a normal marine paleoenvironment. Anatomical and histological consideration of the enigmatic Burgess worm Amiskwia suggest that it may in fact be a chaetognath, while the putative chordate Pikaia appears not to be related to modern cephalochordates.

Immature specimens of the Late Cretaceous pterosaur Pteranodon were identified using three size-independent criteria: (1) fusion of various cranial and postcranial elements; (2) degree of epiphyseal ossification; and (3) bone grain or degree of ossification of limb-bone shafts. Immature individuals make up 15% of available specimens of Pteranodon and do not differ significantly in size from mature individuals. This and the extensive fusion of the mature skeleton suggest that Pteranodon had determinate growth. The bone of limb-bone shafts of immature individuals is fibro-lamellar bone, which suggests that they grew rapidly to adult size. The size-independent criteria can also be used to identify immature and mature individuals of other pterosaur taxa, and other large pterodactyloids also probably exhibited rapid determinate growth.

The number of cell types required for the construction of a metazoan body plan can serve as an index of morphological (or anatomical) complexity; living metazoans range from four (placozoans) to over 200 (hominids) somatic cell types. A plot of the times of origin of body plans against their cell type numbers suggests that the upper bound of complexity has increased more or less steadily from the earliest metazoans until today, at an average rate of about one cell type per 3 m.y. (when nerve cell types are lumped). Computer models in which increase or decrease in cell type number was random were used to investigate the behavior of the upper bound of cell type number in evolving clades. The models are Markovian; variance in cell type number increases linearly through time. Scaled to the fossil record of the upper bound of cell type numbers, the models suggest that early rates of increase in maximum complexity were relatively high. The models and the data are mutually consistent and suggest that the Metazoa originated near 600 Ma, that the metazoan “explosion” near the Precambrian/Cambrian transition was not associated with any important increase in complexity of body plans, and that important decreases in the upper bound of complexity are unlikely to have occurred.

Nature's discontinuities occur both in the hierarchical structuring of genealogical individuals and in the distinct processes operating at different scales of time, here called tiers. Conventional evolutionary theory denies this structuring and attempts to render the larger scales as simple extrapolation from (or reduction to) the familiar and immediate—the struggle among organisms at ecological moments (conventional individuals at the first tier). I propose that we consider distinct processes at three separable tiers of time: ecological moments, normal geological time (trends during millions of years), and periodic mass extinctions.

I designate as “the paradox of the first tier” our failure to find progress in life's history, when conventional theory (first tier processes acting on organisms) expects it as a consequence of competition under Darwin's metaphor of the wedge. I suggest a resolution of the paradox: whatever accumulates at the first tier is sufficiently reversed, undone, or overriden by processes of the higher tiers. In particular, punctuated equilibrium at the second tier produces trends for suites of reasons unrelated to the adaptive benefits of organisms (conventional progress). Mass extinction at the third tier, a recurring process now recognized as more frequent, more rapid, more intense, and more different than we had imagined, works by different rules and may undo whatever the lower tiers had accumulated.

Most modern marine ecology is ultimately based on unicellular phytoplankton, yet most large animals are unable to graze directly on even relatively large net phytoplankton; the repackaging effected by herbivorous mesozooplankton thus represents a key link in marine metazoan food chains. Despite the deep taphonomic biases affecting plankton fossilization, there is a clear record of phytoplankton from at least 1800 m.y ago. Proterozoic plankton are represented by small-to medium-sized sphaeromorphic acritarchs and probably do not include many/most of the unusually large acritarchs that characterize the Neoproterozoic. The first significant shift in phytoplankton diversity was therefore the rapid radiation of small acanthomorphic acritarchs in the Early Cambrian. The coincidence of phytoplankton diversification with the Cambrian radiation of large animals points compellingly to an ecological linkage between the two, particularly in light of recently discovered filter-feeding mesozooplankton in the Early Cambrian. The introduction of planktic filter feeders would have established the second tier of the Eltonian pyramid, potentially setting off the “self-propagating mutual feedback system of diversification” now recognized as the Cambrian explosion (Stanley 1973, 1976).

By consuming significant percentages of net phytoplankton and suspending it as animal biomass and non-aggregating fecal pellets, mesozooplankton cause a net reduction in export production; a general introduction of zooplankton would therefore have reduced carbon burial and moderated the bloom and bust cycle that must have characterized Proterozoic populations of net phytoplankton. The effect of added trophic levels in Early Cambrian ecosystems can be viewed as a serial application of the trophic cascade process observed in modern lakes, whereby the introduction of higher trophic levels determines the accumulation of plant biomass at the base of the system. As such, the major biogeochemical perturbations that mark the onset of the Phanerozoic might be considered a consequence, rather than a cause, of the Cambrian explosion; reduced C export due to zooplankton expansion explains the otherwise anomalous drop in δ13C at the base of the Tommotian.

Cambrian acanthomorphic acritarchs likely derived from planktic leiosphaerids exposed to mesozooplanktic grazing pressure, the ornamentation effectively increasing vesicle size without compromising buoyancy or surface-area:volume ratios. Alternatively, they may represent an escape into the plankton through a miniaturization of the much larger Neoproterozoic acanthomorphs. An invasion of small benthic herbivores into the water column to exploit the phytoplankton accounts for the origin of the mesozooplankton and may have been the key innovation in the Cambrian explosion.

The history of life is replete with apparent order. Much of this order may reflect the deterministic causes conventionally invoked, but we cannot be sure until we measure and subtract the order that arises in simple random systems. Consequently, we have constructed a random model that builds evolutionary trees by allowing lineages to branch and become extinct at equal probabilities. We proceed by dividing our simulated tree into clades and by comparing their sizes and shapes with the patterns exhibited by “real” clades as recorded by fossils.

We regard the similarity of real and random clades as the outstanding result of this comparison. In both real and random systems, extinct clades arising after an “ecological barrel” had been filled have their maximum diversity at the midpoint of their duration; clades arising during the initial “filling” reach an earlier climax during this preequilibrial period of rapid diversification. However, some potential differences also emerge. Clades still living are much larger than extinct clades. We may attribute this to the morphological superiority of survivors, but we can also simulate it in a model that chooses the originators of clades at random. Real clades undergo greater fluctuations in diversity than do random clades, but the effect is not marked.