The number of regenerated arms was counted on specimens of two distinct phenotypes of the stalked crinoid Endoxocrinus parrae (Gervais) from a wide bathymetric range in the Caribbean (178-723 m). In one phenotype, the sample was divided into two groups, one from shallower (< 500 m) depths, the other from deeper (≥ 500 m); in the other phenotype the group divided at 550 m. In both phenotypes, the frequency of regenerated arms is significantly higher in specimens from shallower water than in those from deeper water. If the regenerated arms in Endoxocrinus parrae were the result of sublethal predation, as previously suggested, then predation intensity is higher in shallow water than deep water. These results are consistent with the idea of the late Mesozoic marine revolution—that there has been stronger predation on various invertebrates in shallow-water environments since the late Mesozoic. The stalked crinoids may have been unable to cope with increased predation in shelf environments, and they migrated to offshore environments.

Evaluating the relative importance of biotic versus abiotic factors in governing macroevolutionary patterns is a central question of paleobiology. Here, we analyzed patterns of changes in global relative abundances and diversity of ecological groups to infer the role of biological interactions as driving evolutionary forces in mid-Mesozoic macrobenthic marine ecosystems. Specifically, we tested the hypothesis of escalation, which states that macroevolutionary patterns were controlled by an increasing pressure exerted by enemies on their victims. Associated with evidence of increasing levels of predation and biogenic sediment reworking (bulldozing) is an increasing representation of predation- and disturbance-resistant groups in the fossil record. In particular, we observe increasing proportions of mobile organisms; a decline of vulnerable epifauna living freely on the substrate; and a trend toward infaunalization of the benthos. These trends were most pronounced in the paleotropics, i.e., the region where biological activity is thought to have been highest. The observation that these changes affected several biotic traits and occurred within independent clades argues against the overriding role of a single key adaptive innovation in causing shifts in ecological abundance. Also, changes in the abiotic environment cannot explain these faunal patterns because of lacking cross-correlations with physico-chemical parameters such as global sea level, climate, and seawater chemistry. We conclude that in marine benthic ecosystems of the mid Mesozoic, enemy-driven evolution, or escalation, was a plausible and important factor.

The problem of how accurately paraphyletic taxa versus monophyletic (i.e., holophyletic) groups (clades) capture underlying species patterns of diversity and extinction is explored with Monte Carlo simulations. Phylogenies are modeled as stochastic trees. Paraphyletic taxa are defined in an arbitrary manner by randomly choosing progenitors and clustering all descendants not belonging to other taxa. These taxa are then examined to determine which are clades, and the remaining paraphyletic groups are dissected to discover monophyletic subgroups. Comparisons of diversity patterns and extinction rates between modeled taxa and lineages indicate that paraphyletic groups can adequately capture lineage information under a variety of conditions of diversification and mass extinction. This suggests that these groups constitute more than mere “taxonomic noise” in this context. But, strictly monophyletic groups perform somewhat better, especially with regard to mass extinctions. However, when low levels of paleontologic sampling are simulated, the veracity of clades deteriorates, especially with respect to diversity, and modeled paraphyletic taxa often capture more information about underlying lineages. Thus, for studies of diversity and taxic evolution in the fossil record, traditional paleontologic genera and families need not be rejected in favor of cladistically-defined taxa.

In evaluating the biotic effects of transgressions and regression, care must be taken to ensure that observed patterns are not simply an artifact of the location of available stratigraphic sections along an ancient onshore-offshore gradient of adaptive types. Analysis of Recent bivalves suggests that very nearshore benthic assemblages are dominated by species that are geographically more widespread, are more eurytopic, and more often have planktotrophic larvae than species in offshore assemblages (Jackson 1974; Jablonski and Valentine 1980); this pattern may serve as a null hypothesis for paleobiogeographic analysis. Late Cretaceous bivalve and gastropod faunas of the Gulf and Atlantic Coastal Plain exhibit decreasing levels of endemism and increasing mean geographic range over the course of regression and return to high levels of endemism and low mean geographic range with the succeeding transgression. In addition, species with longer durations are more frequent at peak regression, while geologically shorter-lived species are more prevalent at peak transgression. As seen in Recent examples, the molluscan assemblages from nearshore facies have a higher proportion of species with broad environmental tolerances and planktotrophic larvae, and thus more extensive geographic and stratigraphic ranges, than do the more offshore shelf assemblages. Because late regressive phases are represented only by very nearshore facies, these directional changes in biogeographic and evolutionary characteristics are most parsimoniously interpreted as a reflection of the nature of the facies available for sampling rather than biotic effects of transgression and regression.

The incidences of selected features of dicotyledonous wood that are believed to be of ecologic and/or phylogenetic significance (distinct growth rings, narrow and wide vessel diameter, high and low vessel frequencies, scalariform perforations, tangential vessel arrangement, ring porosity, and helical wall thickenings) were plotted through time (Cretaceous–Recent). There are marked differences between the Cretaceous and Tertiary in the frequency of all wood anatomical features. Incidences of features that are associated with markedly seasonal climates in extant floras do not approach modern levels until the Neogene. Correlations of wood anatomical features with ecology do not appear to have been constant through time, because in the Cretaceous different features provide conflicting information about the climate. Throughout the Tertiary the southern hemisphere/tropical and the northern hemisphere/temperate regions differed in the incidences of ecologically significant features and these differences are similar to those in the Recent flora. Possibilities for reliably using dicotyledonous wood for climatic reconstructions appear restricted to the Tertiary and Quaternary. However, at present the fossil wood record for most epochs and regions is too limited to permit detailed reconstructions of their past climate.

A major aim of the analysis of faunal remains from archaeological sites is the reconstruction of the palaeoecology of early man. Since taphonomic work must precede palaeoecological work, theories concerning the palaeoecology of early man would benefit from firmly based taphonomic studies of the archaeological deposits. One of the numerous taphonomic factors to be considered is the bone fragmentation produced by early man.

The degree of bone fragmentation in a deposit uninfluenced by human factors is an expression of the resistance of the various bones to mechanical and chemical decomposition before and during burial. Thus, the degree of fragmentation where human factors are predominant should provide an illustration of man as a taphonomic factor. In both cases the degree of fragmentation may be used as an indicator of the degree of taphonomic overprint.

Examination of bone material from four West European Mesolithic sites (all bog deposits), Star Carr 7200 ± 120 B.C., Kongemosen 6600 ± 100 B.C., Præstelyngen 3200 ± 100 B.C., and Muldbjerg I 2900 ± 80 B.C., reveals differences in marrow fracturing techniques. Furthermore, the different techniques result in different numbers of fragments for the same type of bone. Comparison of faunas from different sites based on the number of bone fragments must therefore be preceded by analysis of the marrow fracturing techniques used at the various sites. The number of fragments per estimated minimum number of individuals should indicate the degree of taphonomic loss.

The various types of marrow fracturing found seem to belong to cultures at different levels of technical development. Thus, it may be possible within a limited area such as northern Europe to establish a chronology using marrow fracturing type as one of the ecostratigraphic tools.

Much new empirical evidence on the levels of Phanerozoic paleoprovinciality and of species diversity within paleocommunities now permits a reevaluation of marine diversity patterns. Data on paleoprovincial patterns are assembled from the literature and evaluated by means of a stochastic computer simulation model. The simulation is based on the statistics of modern patterns of diversity and endemism extrapolated conservatively to the paleoprovincial patterns and on estimates of species duration from the fossil record. The species diversities associated with the paleoprovincial patterns are then corrected for temporal changes in species packing in communities as determined by Bambach (1977) from studies of paleocommunities. The model thus has an empirical basis throughout. Furthermore it is free of biases that can arise due to the differential preservation of taxa in space and time.

The Paleozoic and Mesozoic were characterized by low provinciality and low average species diversity, on the order of 38,000 to 40,000 species, although there were significant fluctuations in standing diversities. In the Cenozoic, provinciality rose markedly, primarily through the appearance of latitudinal provincial chains, and average species diversity rose to about 240,000. Today it stands over 350,000; this is an order of magnitude greater than the Paleozoic average.

Siwalik rocks of Pakistan are a virtually continuous, continental sedimentary sequence, extending in age from 18 to 1 ma b.p. This paper describes taphonomic features of late Miocene mammalian assemblages from a highly fossiliferous interval about 400 m thick, based on field documentation of sedimentary environments at 42 fossil localities and systematic fossil collection of 21 localities.

Within a broadly fluvial system, I recognize four sedimentary environments of bone accumulation, distinguished by lithology, unit-thickness, unit-geometry, contacts, sedimentary structures, and relationship to adjacent units. Each environment corresponds to an association of lithofacies. Facies Association I is interpreted as the persistent, major channel bodies of a meandering fluvial system; Facies Association II as coarse-grained flood deposits, such as crevasse splays, deposited beyond the main channels; Facies Association III as channel margins, including levees and swales; and Facies Association IV as predominantly subaerial floodplains.

Taphonomic features of bone assemblages from each facies association include skeletal-element composition, surface distribution of specimens, degree of articulation, hydraulic equivalence between organic and inorganic sedimentary particles, frequency of juvenile remains, size distribution of fauna, and an estimate of duration of accumulation of individual fossil localities. The distribution of these features among the four facies associations suggests that bone assemblages in Facies Associations I and II accumulated by the action of currents in river channels or floods, whereas bone assemblages in Facies Associations III and IV accumulated through concentration by biological agents and/or attrition at a repeatedly used site of predation.

Inclusion in fluvial accumulations depends on initial availability of skeletal remains and hydraulic characteristics of individual skeletal elements, but not taxonomic identity per se. For biological accumulations, however, taxonomic composition reflects the preferences of the individual agents of accumulation. The probability of preservation of taxa in fluvial accumulations is probably mainly a function of body size, as reflected in the sizes of isolated skeletal elements. Thus, in this Siwalik system, bone assemblages that experienced fluvial transport are better representations of original community composition than bone assemblages created by biological agents or passive accumulation.

The evolutionary rates of Paleozoic Crinoidea were obtained using dynamic survivorship analysis. The stratigraphic ranges of 838 genera were used in the analyses, revealing a mean generic duration of 12.0 m.y. and a mean species duration of 6.7 m.y., values within the range of longevities reported for other taxa.

Further analyses showed differences in evolutionary rates among crinoid taxa: camerate species and genera were shorter-lived than species and genera of flexibles and inadunates. This pattern may result from ecological differences among these taxa: an energy budget equation solved for crinoids with various filter morphologies revealed that crinoids with fine-mesh filters require higher current velocities to supply them with sufficient particulate nutrients than do crinoids with coarse-mesh filters. A hypothesis stipulating that these differences control the distribution of crinoids among different environments is supported by patterns of occurrence of Mississippian crinoids: the pinnulate camerates (fine filter) dominate higher energy settings while the non-pinnulate inadunates and flexibles (coarse filter) are found in all environments. The “specialized” pinnulate crinoids may therefore be more prone to speciation and extinction than the non-pinnulate “generalists,” thus accounting for the observed differences in the evolutionary rates of the three subclasses.

The above hypothesis was tested by comparing evolutionary rates of two morphological groups: fine-filtered crinoids (camerates) and coarse-filtered crinoids (non-pinnulate Paleozoic crinoids). As predicted, fine-filtered taxa had higher extinction and origination rates. A “bootstrapping” technique revealed that the differences in extinction rates were significant at the p < 0.10 level.

We analyze a new compilation of Neogene to Recent (22-0 Ma) Caribbean coral occurrences to determine how ecological and life history traits at the population level affect long-term evolutionary patterns. The compilation consists of occurrences of 175 species and 49 genera in one continuous (> 5 m.y.) sequence and 22 scattered sites across the Caribbean region. Previous study of evolutionary rates using these data has shown that both extinction and origination were accelerated between 4 and 1 Ma, resulting in large-scale faunal turnover. Categories for three morphological and two reproductive variables (colony size, colony shape, and corallite size; and sex, and mode of embryonic development; respectively) are assigned to each species in the compilation. Comparisons of the ecological variables with evolutionary rates using randomization procedures and modified analysis of variance show that only colony size was strongly related to rates of extinction and origination during either normal background times or times of accelerated extinction. Extinction rates were lower in species with large colonies, because species with small massive colonies tend to live in small, short-lived populations with highly fluctuating recruitment rates. During turnover, extinction rates increased disproportionately in species with small colonies. Origination rates are found to be less related to ecological variables, although species with small massive colonies originated at higher rates prior to turnover.

Accelerated turnover may have therefore involved an increase in local population extinction rates that caused increased rates of both species extinction and origination across the entire fauna. Since extinction rates accelerated disproportionately with respect to colony size, the overall result was a relative increase in species with large colonies. After severe disturbance, one might expect that populations of species with large colonies and high rates of fragmentation would be more likely to escape extinction, because of larger population sizes, longer generation times, and more constant rates of population increase. The modern Caribbean reef-coral fauna is therefore structured by large, long-lived colonies that are robust to regional environmental change. Many of the very taxa that allowed reef communities to escape collapse in the past are declining today in response to anthropogenic disturbances, suggesting that Caribbean reef communities may be less resilient in the future in response to ongoing environmental perturbations.

Estimates of phylogenetic relationships among fossil taxa implicitly provide hypotheses about the quality of the fossil record. Phylogenetic inferences also provide hypotheses about character evolution. The likelihood of any hypothesis that makes predictions about two data sets is simply the likelihood of the hypothesis given the first data set times the likelihood of the same hypothesis given the second data set. In this case, data set 1 represents stratigraphy and data set 2 represents morphology. Statistical methods exist for determining the likelihood of hypothesized levels of sampling. The likelihood of a hypothesized amount of character change yielding a particular most-parsimonious solution (i.e, L[hypothesized length | parsimony length] can be evaluated with simulations. A reanalysis of hyaenid phylogeny based on published character and stratigraphic data is presented here, using the maximum likelihood method. Two trees are found, depending on assumptions about ambiguous species, which are 11 and 10 steps longer than the most parsimonious tree (61 or 60 vs. 50 steps). However, the trees invoke far less stratigraphic debt (9 or 12 units vs. 47 units as measured in Mammal Zones). An important feature of the results is that the most likely tree length given hyaenid character data is estimated to be 56 to 62 steps (depending on the model of character evolution) rather than 50 steps. The likelihood tree suggests stronger trends toward bone-crushing specializations than does the parsimony tree and further suggests that high levels of homoplasy caused parsimony to underestimate the true extent of those trends. Simulations based on the character data and fossil record of hyaenids suggest that the maximum likelihood method is better able to estimate correct trees than is parsimony and somewhat better able to do so than previously proposed phylogenetic methods incorporating stratigraphy.

Heterochrony, change in developmental rate and timing, is widely recognized as an agent of evolutionary change. Heterotopy, evolutionary change in spatial patterning of development, is less widely known or understood. Although Haeckel coined the term as a complement to heterochrony in 1866, few studies have detected heterotopy or even considered the possibility that it might play a role in morphological evolution. We here review the roles of heterochrony and heterotopy in evolution and discuss how they can be detected. Heterochrony is of interest in part because it can produce novelties constrained along ancestral ontogenies, and hence result in parallelism between ontogeny and phylogeny. Heterotopy can produce new morphologies along trajectories different from those that generated the forms of ancestors. We argue that the study of heterochrony has been bound to an analytical formalism that virtually precludes the recognition of heterotopy, so we provide a new framework for the construction of ontogenetic trajectories and illustrate their analysis in a phylogenetic context. The study of development of form needs tools that capture not only rates of development but the space in which the changes are manifest. The framework outlined here provides tools applicable to both. When appropriate tools are used and the necessary steps are taken, a more comprehensive interpretation of evolutionary change in development becomes possible. We suspect that there will be very few cases of change solely in developmental rate and timing or change solely in spatial patterning; most ontogenies evolve by changes of spatiotemporal pattern.

The clade Archosauria contains two very different sister groups in terms of diversity (number of species) and disparity (phenotypic variation): Crurotarsi (taxa more closely related to crocodiles than to birds) and Ornithodira (pterosaurs and dinosaurs including birds). The extant species of Crurotarsi may constitute a biased sample of past biodiversity regarding growth patterns and metabolic rates. Bone histological characters can be conserved over hundreds of millions of years in the fossil record and potentially contain information about individual age at death, age at sexual maturity, bone growth rates, and basal metabolic rates of extinct vertebrates. Using a sample of extant amniotes, we have constructed a paleobiological model to estimate bone growth rate from bone histological traits. Cross-validation tests show that this model is reliable. We then used it to estimate bone growth rates in a sample of extinct archosaurs including Crurotarsi and Ornithodira. After testing for phylogenetic signal, optimization of femoral growth rates through squared change parsimony onto a time-calibrated tree of amniotes shows two divergent evolutionary trends: whereas bone growth rates increase from the last common ancestor of Ornithodira to extant birds, they decrease from the last common ancestor of Crurotarsi to extant crocodiles. However, we conclude, on the basis of recent evidence for unidirectional airflow in the lungs of alligators, that crocodiles may have retained the capacity of growing at high rates.

Fine-scale sampling and analysis of fossiliferous sequences have been used in the debate over gradual vs. punctuated evolutionary transitions between species. The time scale and completeness of these studied sequences can be evaluated using criteria based on compilations of sedimentation rates over different time spans. A set of procedures, herein termed “resolution analysis,” provides the means for estimating the time scale and quality of sequences from which evolutionary patterns are distilled. Seven such published studies are evaluated with these procedures. In general, most fossil sequences are too incomplete on fine time scales to show changes operating within a standing population. Short segments of some sequences have the potential to document nearly complete morphological histories on time scales approaching generation-to-generation processes. Resolution analysis is a necessary step in inferences regarding fossil evidence of evolutionary tempo and mode.

Stratigraphic incompleteness necessarily results from the episodic nature of sedimentation. Many stratigraphic gaps result from minor, temporary shifts in sediment distribution, though other, more profound gaps result from changes in habitat conditions that must have had an effect on local biotic distribution. Analyzing paleontological patterns involves not only the positive evidence provided by fossiliferous strata, but also the negative evidence left as gaps by shifting habitat conditions. Thus, incomplete sequences may not be “flawed” records of continuous populations. Incomplete sequences may be a faithful, literal record of separate populations separated in time by local extinctions and re-invasions.

Available data on the stratigraphic ranges of latest Silurian and Devonian vascular plant macro-fossils (sporophytes) and spores provide insights into the tempo and mode of early tracheophyte evolution. Patterns of diversification, origination, and extinction conform in general to the predictions of Sepkoski's kinetic model of diversification. Rates of generic origination and extinction vary not only through time but also between organ systems for a single time interval. This fact, coupled with data on longevity and turnover and comparative morphological observations, can be used to document mosaic evolution in early vascular plant history. Mosaic evolution is an important theme in plant evolution; indeed, what we recognize as macroevolutionary events often correlate with brief periods of pronounced mosaicism. Such evolutionary patterns reflect the developmental biology of tracheophytes in which individual organs often have life spans that are considerably shorter than the life of the whole plant. Under these conditions, individual organs or organ systems can respond to different sets of evolutionary pressures.

The major period of early vascular plant diversification occurred during the late Early and early Middle Devonian Period, 30 Myr or more after the origin of the group. Such lags in diversification are not uncommon in the fossil record. Sometimes they reflect extrinsic controls on diversification, but in other cases they appear to be a consequence of intrinsic rates of origination and extinction.

Rates of phenotypic evolution are central to many issues in paleontology, but traditional rate metrics such as darwins or haldanes are seldom used because of their strong dependence on interval length. In this paper, I argue that rates are usefully thought of as model parameters that relate magnitudes of evolutionary divergence to elapsed time. Starting with models of directional evolution, random walks, and stasis, I derive for each a reasonable rate metric. These metrics can be linked to existing approaches in evolutionary biology, and simulations show that they can be estimated accurately at any temporal resolution via maximum likelihood, but only when that metric's underlying model is true.

The estimation of generational rates of a random walk under realistic paleontological conditions is compared with simulations to that of a prominent alternative approach, Gingerich's LRI (log-rate, log-interval) method. Generational rates are estimated poorly by LRI; they often reflect sampling error more than the actual pace of change. Further simulations show that under some realistic conditions, it is simply not possible to infer generational rates from coarsely sampled populations.

These modeling results indicate a complex dependence between evolutionary mode and the measurement of evolutionary rates, and that there is unlikely to be a rate metric that works well for all traits and time scales. Compilations of paleontological and phylogenetic data indicate that all of the three rate metrics derived here show some relationship with interval length. Although there is no perfect rate metric, at present the most practical choices derive from the parameters of the stasis and random walk models. The latter, called the step variance, is particularly promising as a rate metric in paleontology and comparative biology.

The frequent late Cenozoic glacial ages were accompanied by sea-level falls of 100–150 m amplitude. These falls stranded the complex inner-reef platforms and lagoons of tropical Pacific islands, while outer-reef-slope habitats persisted, although displaced downslope. The effects of glacial regressions on bivalves were studied by examining the zonation of species across reef systems and species composition on tectonically uplifted islands, islands with, effectively, local low sea stands. I show that qualitative habitat loss (the stranding of inner-reef habitats) was responsible for the local extinction of about one-third of the bivalve species that inhabit central Pacific islands during high sea stands, whereas quantitative loss of habitat area and climatic effects were inconsequential. Soft-sediment habitats, and consequently soft-bottom bivalves, were more drastically affected by sea-level fluctuations than were hard-bottom habitats and bivalves.

Although many bivalve species were extirpated in the central Pacific during low sea stands, they survived in the western Pacific, where the different geomorphology of many marine systems provided refugia for lagoonal species. Thus, a large proportion of Pacific bivalve species has very dynamic distributions, undergoing great range reductions and expansions with falls and rises of sea level, and much of the present central Pacific lagoonal fauna is of Holocene age. Several implications of these findings are discussed.

Size and shape are analyzed for Pliocene lineages of the rodent genus Stephanomys Schaub 1938. Previous phylogenetic studies were based mainly on size variation and descriptive comparisons, without any attempt to quantify shape changes. Hence, on the basis of regular size increase, Stephanomys has been considered a prime example of phyletic gradualism. In order to quantify morphological variation within the lineage, a method for analyzing complex outlines, the elliptic Fourier transform, was applied to tooth contour (upper and lower first molars). It was then possible to compare evolution in size, estimated by tooth area, as well as evolution of shape, represented by Fourier coefficients.

While size seems to change gradually through time, morphology gives a rather discontinuous evolutionary pattern for both the upper and lower molar. Such a discrepancy between the evolution of size and shape of a single structure suggests that different genetic determinisms and mechanical constraints may act on size and shape. Hence it may be misleading to infer generalized evolutionary processes from either size or shape alone.

Short-term variations in rates of taxonomic extinction and origination in the fossil record may be the result of true changes in rates of turnover, variable rates of fossil preservation, or some combination of the two. Here, positive extinction and origination rate excursions among Phanerozoic marine animal genera are reexpressed in terms of the amount of normal, background time they represent. In addition to providing a background-adjusted calibration of rate intensities, this reexpression determines the durations of sampling gaps that would be required to explain entirely all observed rate excursions as sampling artifacts. This possibility is explored by analyzing a new compilation of the timing and duration of sedimentary hiatuses in North America.

Hiatuses spanning more than approximately one million years (Myr) in the marine sedimentary rock record have a mean duration of 73 Myr. There are two major hiatus types—those that form in response to long-duration tectonic cycles and that bound the major Sloss-scale sequences (mean duration 200 Myr), and those that form in response to shorter-duration changes in sediment accommodation space and that occur within major Sloss-scale sequences (mean duration less than 23 Myr). The latter are approximately exponentially distributed and have a mean duration that is comparable to the mean duration of intervening sedimentary rock packages.

Although sedimentary hiatuses are generally long enough in duration to accommodate the hypothesis that short-term variations in rates of genus origination and extinction are artifacts of sampling failures at major unconformities (“Unconformity Bias” hypothesis), the observed evolutionary rates are not correlated with hiatus durations. Moreover, hiatuses that follow the major mass extinctions are not long in comparison to most other non-mass extinction intervals. These results do not support the hypothesis that hiatuses at major unconformities alone have artificially clustered genus first and last occurrences, thereby causing many of the documented statistical similarities between the temporal structure of the sedimentary rock record and macroevolutionary patterns. Instead, environmental changes related to the expansion and contraction of marine environments may have been the primary forcers of both biological turnover and the spatio-temporal pattern of sediment accumulation. Fully testing this “Common Cause” hypothesis requires quantifying and overcoming lingering taxonomic, biostratigraphic, facies, and numerous other biases that are both inherent in geologic data and imposed by imperfect knowledge of the fossil record.

Studies of the end-Permian mass extinction have suggested a variety of patterns from a single catastrophic event to multiple phases. But most of these analyses have been based on fossil distributions from single localities. Although single sections may simplify the interpretation of species diversity, they are susceptible to bias from stratigraphic incompleteness and facies control of preservation. Here we use a data set of 1450 species from 18 fossiliferous sections in different paleoenvironmental settings across South China and the northern peri-Gondwanan region, and integrate it with high-precision geochronologic data to evaluate the rapidity of the largest Phanerozoic mass extinction. To reduce the Signor-Lipps effect, we applied constrained optimization (CONOP) to search for an optimal sequence of first and last occurrence datums for all species and generate a composite biodiversity pattern based on multiple sections. This analysis indicates that an abrupt extinction of 62% of species took place within 200 Kyr. The onset of the sudden extinction is around 252.3 Ma, just below Bed 25 at the Meishan section. Taxon turnover and diversification rates suggest a deterioration of the living conditions nearly 1.2 Myr before the sudden extinction. The magnitude of the extinction was such that there was no immediate biotic recovery. Prior suggestions of highly variable, multi-phased extinction patterns reflect the impact of the Signor-Lipps effect and facies-dependent occurrences, and are not supported following appropriate statistical treatment of this larger data set.