Mammalian carnivores are rarely incorporated in paleoenvironmental reconstructions, largely because of their rarity within the fossil record. However, multivariate statistical modeling can be successfully used to quantify specific anatomical features as environmental predictors. Here we explore morphological variability of the humerus in a closely related group of predators (Felidae) to investigate the relationship between morphometric descriptors and habitat categories. We analyze linear measurements of the humerus in three different morphometric combinations (log-transformed, size-free, and ratio), and explore four distinct ways of categorizing habitat adaptations. Open, Mixed, and Closed categories are defined according to criteria based on traditional descriptions of species, distributions, and biome occupancy. Extensive exploratory work is presented using linear discriminant analyses and several fossils are included to provide paleoecological reconstructions.

We found no significant differences in the predictive power of distinct morphometric descriptors or habitat criteria, although sample splitting into small and large cat guilds greatly improves the stability of the models. Significant insights emerge for three long-canine cats: Smilodon populator, Paramachairodus orientalis, and Dinofelis sp. from Olduvai Gorge (East Africa). S. populator and P. orientalis are both predicted to have been closed-habitat adapted taxa. The false “sabertooth” Dinofelis sp. from Olduvai Gorge is predicted to be adapted to mixed habitat. The application of felid humerus ecomorphology to the carnivoran record of Olduvai Gorge shows that the older stratigraphic levels (Bed I, 1.99–1.79 Ma) included a broader range of environments than Beds II or V, where there is an abundance of cats adapted to open environments.

Mean adult body mass of mammal taxa is a fundamental ecological variable. Variability in the distributions of body masses of a mammal fauna suggest variability in habitat structure. Mammal remains from the Marmes archaeological site in southeastern Washington State date between 13,200 and 10,400 b.p., during the Pleistocene–Holocene transition (PHT). Known environmental history prompts the expectations that the Marmes PHT mammal remains should represent greater species richness and a larger array of body-mass sizes than modern faunas in the Marmes locale and in open shrub-steppe habitats, and lower species richness and a smaller array of body-mass sizes than modern faunas in closed forest habitats; species richness and the array of body-mass sizes should be similar to that for a mixed habitat of cool shrub-steppe with scattered conifers. The Marmes PHT cenogram meets these expectations. Body-mass clumps displayed by the Marmes PHT mammal fauna fall between those of closed forests and open shrub-steppe habitats in terms of clump richness and breadth, and in terms of gap width. Marmes PHT body-mass clumps are very similar to those for the mixed habitat. Cenograms and body-mass clumps confirm conclusions drawn 40 years ago that the Marmes PHT habitat was much like that of today but cooler and with more plant biomass and greater structural diversity than today.

Numerous environmental and intrinsic biotic factors have been sought to explain patterns in diversity and turnover. Using taxonomically vetted and sampling-standardized data sets of more than 50,000 taxonomic occurrences in the Paleobiology Database (PaleoDB) we tested whether habitat breadth predicts genus durations and diversity dynamics of marine Mesozoic bivalves, and whether this effect is independent of the well-known positive relationship between geographic range and longevity. We defined the habitat breadth of a genus as a function of its realized ranges in water depth, substrate type, and grain size of the substrate. Our analysis showed that mean values of extinction and origination rates are significantly higher for narrowly adapted genera compared to broadly adapted genera, with differences being evident in all analyzed stratigraphic intervals. Linear models showed that both geographic range and habitat breadth have an independent effect on genus durations and on diversity dynamics. These results reaffirm the role of geographic range and furthermore suggest that habitat breadth is an equally important key predictor of extinction risk and origination probability in Mesozoic marine bivalves. Habitat generalists, regardless of their geographic range, are generally less prone to extinction. Conversely, widely distributed genera that are more specialized may be more endangered than one would expect from their geographic range alone. Extinction rates tend to be higher for specialized genera in both background and mass extinctions, suggesting that wide habitat breadth universally buffers against extinction. The trajectories of origination rates through time differ from those of extinction rates. Whereas there is no pronounced ecological selectivity in origination in the Triassic and most of the Jurassic, Cretaceous origination rates are higher for specialized genera. This may best be explained by diversity-dependence. When diversity levels reach a critical point a further increase in diversity is achieved by elevated origination rates of more specialized forms.

In modern ecosystems, regions of topographic heterogeneity, when compared with nearby topographically homogeneous regions, support high species densities of mammals and other groups. This biogeographic pattern could be explained by either greater diversification rates or greater accommodation of species in topographically complex regions. In this context, we assess the hypothesis that changes in landscape history have stimulated diversification in mammals. Landscape history includes tectonic and climatic processes that influence topographic complexity at regional scales. We evaluated the influence of changes in topographic complexity and climate on origination and extinction rates of rodents, the most diverse clade of mammals.

We compared the Neogene records of rodent diversity for three regions in North America. The Columbia Basin of the Pacific Northwest (Region 1) and the northern Rocky Mountains (Region 2) were tectonically active over much of the Cenozoic and are characterized by high topographic complexity today. The northern Great Plains (Region 3) have been tectonically quiescent, with low relief, throughout the Cenozoic. These three regions have distinctive geologic histories and substantial fossil records. All three regions showed significant changes in diversification and faunal composition over the Neogene. In the montane regions, originations and extinctions peaked at the onset and close, respectively, of the Miocene Climatic Optimum (17–14 Ma), with significant changes in faunal composition accompanying these episodes of diversification. In the Great Plains, rodents showed considerable turnover but infrequent diversification. Peak Neogene diversity in the Great Plains occurred during cooling after the Miocene Climatic Optimum. These histories suggest that climatic changes interacting with increasing topographic complexity intensify macroevolutionary processes. In addition, close tracking of diversity and fossil productivity with the stratigraphic record suggests either large-scale sampling biases or the mutual response of diversity and depositional processes to changes in landscape history.

Ontogenetic sequence reconstruction is challenging particularly for extinct taxa because of when, where, and how fossils preserve. Different methods of reconstruction exist, but the effects of preservational bias, the applicability of size-independent methods, and the prevalence of sequence polymorphism (intraspecific variation) remain unexplored for paleontological data. Here I compare five different methods of ontogenetic sequence reconstruction and their effects on the detection of sequence polymorphism, using a large collection of the extinct vertebrates Microbrachis pelikani and Hyloplesion longicostatum. The postcranial ossification sequences presented here for those taxa are the first examples known for extinct lepospondyls. Sequences were reconstructed according to skull length, trunk length, increasing number of ontogenetic events, majority-rule consensus, and Ontogenetic Sequence Analysis (OSA). Results generally were in agreement, demonstrating that paleontological data may be used to robustly reconstruct developmental patterns. When reconstructing sequences based on fossils, size-based methods and OSA are more objective and less dependent on preservational bias than other techniques. Apart from the other methods, OSA also allows for statistical analysis of observed and predicted polymorphism. However, OSA requires a large sample size to yield meaningful results, and size-based methods are justified in paleontological studies when sample size is limited by poor preservation. Different methods of reconstruction detected different patterns of sequence polymorphism, although across all methods the magnitude of sequence variation for M. pelikani and H. longicostatum (1.3−3.4%) was within the lower range of values reported for extant vertebrates. Compared with other extinct and extant tetrapods, all sequence reconstruction methods consistently showed that M. pelikani and H. longicostatum exhibit advanced ossification of the pubis and delayed ossification of the scapula. However, the postcranial ossification sequences of these two taxa largely are congruent with those of other tetrapods, suggesting an underlying conservative ancestral pattern that evolved early in tetrapod history.

The Cretaceous/Tertiary (K/Pg) mass extinction has long been viewed as a pivotal event in mammalian evolutionary history, in which the extinction of non-avian dinosaurs allowed mammals to rapidly expand from small-bodied, generalized insectivores to a wide array of body sizes and ecological specializations. Many studies have used global- or continental-scale taxonomic databases to analyze this event on coarse temporal scales, but few studies have documented morphological diversity of mammalian paleocommunities on fine spatiotemporal scales in order to examine ecomorphological selectivity and ecospace filling across this critical transition. Focusing on well-sampled and temporally well-constrained mammalian faunas across the K/Pg boundary in northeastern Montana, I quantified dental-shape disparity and morphospace occupancy via landmark- and semilandmark-based geometric morphometrics and mean body size, body-size disparity, and body-size structure via body-mass estimates.

My results reveal several key findings: (1) latest Cretaceous mammals, particularly metatherians and multituberculates, had a greater ecomorphological diversity than is generally appreciated, occupying regions of the morphospace that are interpreted as strict carnivory, plant-dominated omnivory, and herbivory; (2) the decline in dental-shape disparity and body-size disparity across the K/Pg boundary shows a pattern of constructive extinction selectivity against larger-bodied dietary specialists, particularly strict carnivores and taxa with plant-based diets, that suggests the kill mechanism was related to depressed primary productivity rather than a globally instantaneous event; (3) the ecomorphological recovery in the earliest Paleocene was fueled by immigrants, namely three multituberculate families (taeniolabidids, microcosmodontids, eucosmodontids) and to a lesser extent archaic ungulates; and (4) despite immediate increases in the taxonomic richness of eutherians, their much-celebrated post-K/Pg ecomorphological expansion had a slower start than is generally perceived and most likely only began 400,000 to 1 million years after the extinction event.

Synapsids dominated the terrestrial realm between the late Pennsylvanian and the Triassic. Their early evolution includes some of the first amniotes to evolve large size, herbivory, and macro-predators. However, little research has focused on the changes in diversity occurring during this early phase in their evolutionary history, with more effort concentrating on later events such the Permo-Triassic extinction. Here we assess synapsid diversity, at both the species and genus levels, between the Carboniferous (Moscovian) and the Middle Permian (Capitanian). A raw, taxic diversity (richness) estimate is generated, and we use two separate methods to correct for sampling biases in this curve. To remove the effect of anthropogenic sampling bias, we apply a recently published modification of the residual diversity method, and then generate a supertree, using matrix representation with parsimony to infer ghost lineages and obtain a phylogenetic diversity estimate. The general diversity pattern reflects the initial diversification of synapsids in the late Pennsylvanian and early Cisuralian, which was followed by an extinction event during the Sakmarian. Diversity recovered during the Artinskian and Kungurian, coinciding with the radiation of Caseidae, although other families begin to decline. A second extinction event occurred across the Kungurian/Roadian boundary, in which Edaphosauridae and Ophiacodontidae died out although Caseidae and Therapsida diversified. The sampling-corrected curves reveal further extinction during the Roadian, although therapsids were again unaffected. Pelycosaurian-grade synapsids survived during the Wordian and Capitanian, but were a minor part of an otherwise therapsid-dominated fauna. Evidence of significant anthropogenic sampling bias calls into question previous diversity studies that have not employed sampling correction.

The deep-sea Cenozoic planktonic microfossil record has the unique characteristics of continuously well-preserved populations of most species, with virtually unlimited sample size, and therefore constitutes, in principle, a major resource for macroevolutionary research. Antarctic Neogene radiolarians in particular, are diverse, abundant and consistently well-preserved and evolved rapidly. This fauna is, in theory, a near-perfect testing ground for paleodiversity reconstructions. In this study we determined the diversity history of these faunas from a new quantitative, taxonomically complete data set from Neogene and Quaternary sections at several Antarctic sites. The pattern retrieved by our whole-fauna data set shows a significant, largely extinctionless ecological change in faunal composition and decrease in the evenness of species' abundances during the late Miocene, followed 3 Myr later, at around 5 Ma, by a significant drop in diversity. We tentatively associate this ecological event with a synchronous, regional change in the composition of the primary producers, but as yet cannot identify any environmental changes associated with the later extinction. Further, our whole-fauna diversity history was compared to diversity computed from much less complete, biostratigraphically oriented studies of species' occurrences, compiled in the Neptune database and reconstructed by using subsampling methodologies. Comparison of our whole-fauna and subsampling-reconstructed diversity patterns shows that the first-order trends are the same in both, suggesting that, to some degree, such literature compilations can be used to explore diversity history of plankton. However, our results also highlight substantial errors and distortions in the reconstructed diversity which make it poorly suited to more-detailed studies (e.g., for comparison of diversity history with paleoenvironmental history). We conclude that detailed studies of plankton diversity, and particularly those attempting to understand the relation between diversity and paleoceanographic change, should be based on taxonomically comprehensive, quantitative data.