The fossil record is incomplete in the sense that taxa for which there are no preserved finds are unrepresented. Because the probability that there is at least one preserved find is greater for long-lived taxa, failure to account for this effect will bias estimation of mean taxonomic duration. This paper describes maximum likelihood estimation of mean taxonomic duration, fossil preservation rate, and completeness of the local fossil record under a model in which duration has an exponential distribution and the locations of preserved finds follow a Poisson process. The estimates under this model have simple closed forms. It is also straightforward to construct confidence regions for the model parameters. The method is applied to data on 110 trilobite species from the Upper Cambrian—Lower Ordovician in Oklahoma.

Paleontological completeness and stratigraphic ranges depend on extinction rate, origination rate, preservation rate, and the length of the interval of time over which observations can be made. I derive expressions for completeness and the distribution of durations and ranges as functions of these parameters, considering both continuous- and discrete-time models.

Previous work has shown that, if stratigraphic ranges can be followed indefinitely forward, and if extinction and preservation occur at stochastically constant rates, then extinction rate and preservability can be estimated from (1) discrete (binned) stratigraphic ranges even if data on occurrences within ranges are unknown, and (2) continuous ranges if the number of occurrences within each range is known. I show that, regardless of whether the window of observation is finite or infinite, extinction and preservation rates can also be estimated from (3) continuous ranges when the number of occurrences is not known, and (4) discrete ranges when the number of occurrences is not known. One previous estimation method for binned data involves a sample-size bias. This is circumvented by using maximum likelihood parameter estimation. It is worth exploiting data on occurrences within ranges when these are available, since they allow preservation rate to be estimated with less variance. The various methods yield comparable parameter estimates when applied to Cambro-Ordovician trilobite species and Cenozoic mammal species.

Stratigraphic gaps and variable preservation affect stratigraphic ranges predictably. In many cases, accurate parameter estimation is possible even in the face of these complications. The distribution of stratigraphic ranges can be used to estimate the sizes of gaps if their existence is known.

The ammonoid order Prolecanitida constitutes a relatively small (43 genera, ~250 species) but long-ranging lineage (Lower Carboniferous—Triassic, ~108 m.y.), which narrowly survived the P/Tr extinctions and provided the stock from which were derived all later Mesozoic ammonoids. Prolecanitids were a minority among Late Paleozoic ammonoids, which were dominated by the Goniatitida, and showed many features that set them far apart from their contemporaries, including (1) long-term, gradual changes in shell geometry (W-D-S); (2) the most strongly constrained morphospace of any Paleozoic ammonids examined to date; (3) an eight-fold increase in mean suture complexity (three times that of Pennsylvanian goniatitids); (4) high correlations between shell geometry, shell and septal thickness, and suture complexity; (5) short body chambers and, as a consequence, high aperture orientations; (6) indications that cameral liquid may have been used for buoyancy control; and (7) a genus longevity that averaged 14.7 m.y. compared with 5.7 m.y. in Upper Carboniferous goniatitids, and that appears to have been unrelated to suture complexity. Prolecanitids showed a pervasive tendency to increase suture complexity (in the clade as a whole as well as within subclades and in more than 90 percent of ancestor-descendant genera), thus arguing a case for a driven complexity trend. The uniqueness of the prolecanitids calls into question whether they and their Mesozoic descendants, ceratites and ammonites, were strictly analogous to Paleozoic goniatites.

Disturbances have drastically altered Caribbean coral reefs over the past two decades. Acropora cervicornis (staghorn coral), which predominated at intermediate depths (5—25 m) from the 1950s through the 1970s, has virtually disappeared from most reef environments. Other coral species have declined as well, and the cover of macroalgae has increased. In apparent contrast, fossil reef sequences suggest that the species composition and zonation of coral assemblages did not change during the Pleistocene and Holocene. One interpretation of these observations is that coral species persisted on Caribbean reefs for hundreds of thousands of years as components of tightly integrated communities, and that a rare or unique combination of disturbances led to the synchronous decline of A. cervicornis and other corals throughout the region. The hypotheses of (1) community integration and (2) a unique, recent community transition, were tested by ecological and paleoecological observations in the shelf lagoon of the Belizean Barrier Reef.

The reef growing along the flanks of Channel Cay, a lagoonal shoal, was monitored by point counts along transects over a ten-year period (1986—95). This reef was covered primarily by A. cervicornis at 3—15 m depth until the late 1980s. After 1986, A. cervicornis experienced a mass mortality from White Band Disease, an epizootic of presumed bacterial origin. The cover of A. cervicornis dropped from ~70% in 1986 to nearly 0% in 1993. Agaricia spp. (lettuce corals) responded opportunistically to the availability of free space in the form of A. cervicornis skeletal rubble. Agaricia, which had been a minor constituent of the sessile biota (10% cover in 1986), replaced A. cervicornis as the most common occupant of space on the reef (56% cover in 1995). The percent cover of other coral species and macroalgae remained low throughout the ten-year period. Similar changes were observed on other reefs over an area of at least 250 km2.

The Acropora-to-Agaricia transition left a clear signature in the sedimentary record. Trenches dug into the reef at Channel Cay revealed the accretion of a layer of Agaricia rubble with a mean thickness of 22 cm in the decade after 1986. Due to the unconsolidated, uncompacted nature of the reef sediments, evidence of previous Acropora-to-Agaricia transitions should have been visible in the fossil record as vertical accumulations of A. cervicornis branches interrupted by layers of imbricated Agaricia rubble. Coring studies at Channel Cay revealed that no other Agaricia layers were deposited during at least the past 3800 years; the recent transition was unique on a time scale of millennia. This result supports the contention that excursions from the Acropora-dominated situation are unusual in the history of Channel Cay and nearby reefs. However, the dynamics of the transition do not support the community integration hypothesis for the Channel Cay reef, indicating instead that different coral taxa in this assemblage responded differently, or not at all, to a large-scale biotic disturbance. The community transition also underscores the potential for biological factors in general, and disease in particular, to alter the composition of ecological communities and their sedimentary remains.

Late Eocene foraminiferal extinction shows diverse patterns of selective morphologic and latitudinal extinction. Taxa with discoidal shape, calcareous tests, and narrow and low-latitudinal ranges are at significantly greater risk of extinction. Elevated extinction intensities in calcareous tests are mainly due to the presence of larger benthic foraminifera that evolved in late Paleocene and diversified through the lower to middle Eocene. Selectivity of late Eocene foraminiferal extinction indicates that this extinction event was not a globally uniform event. Although this result does not verify an extraterrestrial impact or any other proposed cause of extinction, it does constrain the causes of late Eocene extinction. Furthermore, the geography of late Eocene foraminiferal extinction, and previously studied Cenomanian/Turonian extinction, demonstrates that mass extinctions exhibit different patterns of selectivity.

Traditional wisdom of the evolution of lungs in fishes is that lungs arose when gill ventilation was hindered by an aquatic habitat that was low in oxygen. This scenario has been buttressed primarily by a proposed correlation between extant air-breathing fishes and hypoxic habitats, as well as by the fact that early vertebrate fossils were found in sediments believed to indicate a semi-arid environment. There are problems with this scenario, yet it retains a dominant influence on how the evolution of aerial respiration is viewed. This paper presents a new hypothesis for lung evolution that is more consistent with the fossil record and physiology of extant animals than the traditional scenario; I propose that lungs evolved to supply the heart with oxygen. The primitive vertebrate heart was spongy in architecture and devoid of coronary support, obtaining oxygen from luminal blood. By supplying oxygen to this tissue, lungs may have been important in ancient fishes for sustaining activity, regardless of environment. Furthermore, this function for lungs may have influenced cardiovascular adaptations of tetrapods because their divided cardiovascular system isolates the right side of the heart from pulmonary oxygen. I propose that three innovations compensate for this isolation: In extant amphibians oxygen-rich blood from cutaneous and buccal respiration enters the right side of the heart; in chelonians and lepidosaurs the intracardiac shunt washes oxygen-rich blood into the right side of the heart; in mammals, birds, and perhaps in crocodilians, support of the heart by coronary vasculature eliminates this problem.

Precise estimates of past temperatures are critical for understanding the evolution of organisms and the physical biosphere, and data from continental areas are an indispensable complement to the marine record of stable isotopes. Climate is considered to be a primary selective force on leaf morphology, and two widely used methods exist for estimating past mean annual temperatures from assemblages of fossil leaves. The first approach, Leaf Margin Analysis, is univariate, based on the positive correlation in modern forests between mean annual temperature and the proportion of species in a flora with untoothed leaf margins. The second approach, known as the Climate-Leaf Analysis Multivariate Program, is based on a modern data set that is multivariate. I argue here that the simpler, univariate approach will give paleotemperature estimates at least as precise as the multivariate method because (1) the temperature signal in the multivariate data set is dominated by the leaf-margin character; (2) the additional characters add minimal statistical precision and in practical use do not appear to improve the quality of the estimate; (3) the predictor samples in the univariate data set contain at least twice as many species as those in the multivariate data set; and (4) the presence of numerous sites in the multivariate data set that are both dry and extremely cold depresses temperature estimates for moist and nonfrigid paleofloras by about 2°C, unless the dry and cold sites are excluded from the predictor set.

New data from Western Hemisphere forests are used to test the univariate and multivariate methods and to compare observed vs. predicted error distributions for temperature estimates as a function of species richness. Leaf Margin Analysis provides excellent estimates of mean annual temperature for nine floral samples. Estimated temperatures given by 16 floral subsamples are very close both to actual temperatures and to the estimates from the samples. Temperature estimates based on the multivariate data set for four of the subsamples were generally less accurate than the estimates from Leaf Margin Analysis. Leaf-margin data from 45 transect collections demonstrate that sampling of low-diversity floras at extremely local scales can result in biased leaf-margin percentages because species abundance patterns are uneven. For climate analysis, both modern and fossil floras should be sampled over an area sufficient to minimize this bias and to maximize recovered species richness within a given climate.