Functional traits are indicators of how species interact with their environment, and changes in community-wide trait composition are often associated with environmental change. This relationship provides a means to track ecological dynamics even in cases when taxonomic resolution is limited. The artiodactyl astragalus is a weight-bearing skeletal element that is key to hindlimb movement, and also preserves morphological signals of body size and cursoriality based on the understanding that relatively short astragali in modern artiodactyls are correlated with open habitats. We examined astragalar morphology in fossil artiodactyls (Antilocapridae and Camelidae) from the Miocene Dove Spring Formation (12.5 to 8.5 Ma) to assess whether locomotor traits responded to a known regional shift from a forest-dominated landscape to a patchwork of woodland and grassland habitats. We found that despite ongoing aridification and local changes in vegetation during the middle to late Miocene, the community-wide astragalar functional ratio remained stable, suggesting that environmental changes were insufficient to drive directional morphological adaptation. Clade-specific analyses showed that camelids possessed consistently higher functional ratios than antilocaprids, reflecting divergent locomotor strategies that likely corresponded to differences in body size, habitat use, or mobility. These results suggest that behavioral plasticity-including range tracking and habitat selection-may buffer large herbivores against extinction or adaptation under regional environmental stress in fragmented, heterogeneous landscapes. This study demonstrates the potential of ecometric analyses to reveal ecological patterns in fossil assemblages with limited taxonomic resolution and emphasizes the utility of linking functional morphology with paleoenvironmental data to reconstruct mammalian response to long-term environmental change. We found that despite ongoing aridification and local changes in vegetation during the middle to late Miocene, the community-wide astragalar functional ratio remained stable, suggesting that environmental changes were insufficient to drive directional morphological adaptation. Clade-specific analyses showed that camelids possessed consistently higher functional ratios than antilocaprids, reflecting divergent locomotor strategies that likely corresponded to differences in body size, habitat use, or mobility.

Canids increased in cursoriality through the Cenozoic, as environments transitioned from closed-canopy forest to open grassland and steppe. Canids have evolved through a series of radiations since their origin in the Eocene, but it is unclear if cursorial adaptations appeared in the earliest of these radiations. In the middle Oligocene, the basal hesperocyonines ecologically diversified, and the coyote-sized Mesocyon coryphaeus exemplified the transition from smaller, omnivorous canids to larger, hypercarnivorous forms. M. coryphaeus is exclusively known from the John Day Formation of North America. Although M. coryphaeus is a relatively common fossil in this formation, first recognized in the late 19th century, no postcranial material from this species has ever been formally described. Here, we present a near-complete skeleton of M. coryphaeus, JODA 3366, which includes a complete cranium, near-complete presacral spine, all long bones, elements of both the manus and pes, and a baculum. The short, robust limbs, mobile elbow joint, and tarsal morphology of M. coryphaeus indicate that this species retained a plantigrade to semidigitigrade posture, similar to the earliest canid Hesperocyon, and lacked the cursorial adaptations found in more derived canids. Based on this morphology, we interpret M. coryphaeus as a terrestrial ambush predator, more similar to large mustelids than extant canids, likely hunting small prey like hypertragulids. Although the habitat of M. coryphaeus would have been cooler and more open than the dense closed-canopy forests of the Eocene, enough vegetation cover was still present in the Oligocene for ambush hunting to remain a successful strategy.

Cities are characterized by elevated temperatures, increased pollution, and high-density human populations which often are accompanied by changes in available resources, like food. These shifts have the potential to drive phenotypic divergence in urban wildlife. Functional morphological traits, like body size, can mediate interactions between wildlife and habitat and are closely tied to life history and fitness. While examples of functional morphological variation associated with urbanization are increasing, variation in such traits as a response to urbanization remains unexplored for most taxa. Here, we investigated morphological divergence between urban and rural populations of house mice (Mus musculus domesticus). House mice are globally distributed in diverse habitats and are a model system with a wealth of phenotypic data, making them useful for the study of the impacts of urbanization on morphology. Using a paired replicate design, we sampled urban and rural populations in three distinct metropolitan regions in the eastern United States. We found that body size was smaller in urban populations. Using 3D geometric morphometrics, we also analyzed variation in cranial shape across habitats. Differences in cranial shape were largely allometric, that is, driven by differences in body size. However, we also uncovered evidence of cranial shape variation between habitats not explained by size. In contrast, we did not find evidence for habitat-driven differences in cranial capacity independent of size. Overall, our results suggest a key role for body size in mediating morphological responses to urbanization and highlight the potential of house mice as a globally-distributed model for urbanization.

Lumbar vertebrae are an important functional unit of the spine in mammals. Although the correlation between lumbar morphology and function in extant mammals is well known, the evolution of disparate lumbar morphology through the fossil record has remained unexplored. I sampled lumbar vertebrae from 100 fossil and extant mammals across a wide taxonomic range and categorized the morphology of these vertebrae with discrete morphological characters. I used marginal ancestral state reconstruction to model the evolutionary history of each character and calculated a retention index to assess homoplasy. I then compared these results to correlations between each character and body size, as a functional metric. Presence of anapophyses, a postzygapophyseal lamina, and xenarthrous articulations all showed high retention, indicating strong phylogenetic signal. Centrum length and transverse process angle showed low retention and strong correlation with body size. These results demonstrate that, although lumbar morphology is strongly controlled by function, phylogeny does have a notable influence. Furthermore, these data support the evolution of lumbar vertebrae along the stem to Theria, with ancestral therian vertebrae having had transverse processes rooted to the neural arch, posteriorly angled neural spines, and flat, horizontal zygapophyses.

The unique morphology of mammalian lumbar vertebrae allows the spine to flex and extend in the sagittal plane during locomotion. This movement increases stride length and allows mammals to efficiently breathe while running with an asymmetric gait. In extant mammals, the amount of flexion that occurs varies across different locomotor styles, with dorsostable runners relying more on movement of long limbs to run and dorsomobile runners incorporating more flexion of the back. Although long limbs and a stabilized lumbar region are commonly associated with each other in extant mammals, many "archaic" placental mammals with short limbs had lumbar vertebrae with revolute zygapophyses. These articulations with an interlocking S-shape are found only in artiodactyls among extant mammals and have been hypothesized to stabilize against flexion of the back. This would suggest that archaic placental mammals may not have incorporated dorsoventral flexion into locomotion to the same extent as extant mammals with similar proportions. We tested the relative mobility of fossil lumbar vertebrae from two early placental mammals, the creodonts Patriofelis and Limnocyon, to see how these vertebrae may have functioned. We compared range of motion (ROM) between the original vertebrae, with revolute morphology and digitally altered vertebrae with a flat morphology. We found that the revolute morphology had relatively little effect on dorsoventral flexion and instead that it likely prevented disarticulation due to shear forces on the spine. These results show that flexion of the spine has been an important part of mammalian locomotion for at least 50 million years.

Carnivores (cats, dogs and kin) are a diverse group of mammals that inhabit a remarkable range of ecological niches. While the relationship between ecology and morphology has long been of interest in carnivorans, the application of quantitative techniques has resulted in a recent explosion of work in the field. Therefore, they provide a case study of how quantitative techniques, such as geometric morphometrics (GMM), have impacted our ability to tease apart complex ecological signals from skeletal anatomy, and the implications for our understanding of the relationships between form, function and ecological specialization. This review provides a synthesis of current research on carnivoran ecomorphology, with the goal of illustrating the complex interaction between ecology and morphology in the skeleton. We explore the ecomorphological diversity across major carnivoran lineages and anatomical systems. We examine cranial elements (skull, sensory systems) and postcranial elements (limbs, vertebral column) to reveal mosaic patterns of adaptation related to feeding and hunting strategies, locomotion and habitat preference. We highlight the crucial role that new approaches have played in advancing our understanding of carnivoran ecomorphology, while addressing challenges that remain in the field, such as ecological classifications, form–function relationships and multi-element analysis, offering new avenues for future research.

After the end-Cretaceous mass extinction, placental mammals rapidly diversified in size and locomotor mode, setting the stage for mammals to move into almost every habitat on Earth. Locomotion in extant mammals includes unique sagittal flexion of the trunk primarily driven by lumbar vertebrae, a ribless region of the spine. Consequently, variation in lumbar morphology is associated with a wide variety of locomotor styles. While the origin of this region in early therian mammals in the Mesozoic has been studied, along the therian stem, the later functional diversification of lumbar morphology in placental mammals, the dominant extant group, has been essentially unstudied. We measured the shape of lumbar vertebrae from early placental mammals to test how body size, locomotor specialization, and phylogeny interacted in the diversification of lumbar function after the end-Cretaceous mass extinction. We used 3D geometric morphometrics to quantify shape and compare variation between these Palaeogene mammals and modern mammals. We found that Palaeogene mammals had high disparity in lumbar morphology and that this variation correlated with size and locomotor style. Surprisingly, several ‘archaic’ placentals, like hyaenodontids, showed highly mobile lumbar morphology. These findings show that lumbar vertebrae formed an important and evolvable functional unit at the beginning of the Cenozoic.

Oxyaenodonta is one of several groups of early carnivorous placental mammals that are now extinct. Patriofelis is a genus of Bridgerian oxyaenodont, one of the last known from North America. We describe two partial skeletons of Patriofelis ulta to reevaluate hypotheses about the locomotion of Patriofelis. Patriofelis ulta had short limbs relative to a longer body, similar in proportion to other oxyaenodonts. The forelimb shows a wide capitulum of the humerus, indicative of the ability to pronate and supinate the hand. Likewise, the hindlimb shows features associated with flexibility such as a shallow acetabulum, and shallowly grooved astragulus. In contrast, the lumbar vertebrae are inflexible with interlocking zygapophyses. We performed a 2D geometric morphometric analysis of lumbar vertebrae from modern mammals and P. ulta to identify a close functional match. We found that while the lumbar shape in anterior view of P. ulta fell closest to carnivorans, the interlocking zygapophyses matched most closely to modern artiodactyls. We confirmed the occurrence of interlocking zygapophyses in other oxyaenodonts, as well as hyaenodonts and mesonychids, and infer that this is a plesiomorphic feature. Based on this information, we rule out the previously proposed semiaquatic hypothesis for locomotion, as well as the scansorial hypothesis based on the large size of P. ulta. We interpret P. ulta, one of the largest carnivores in its ecosystem, as an ambush predator capable of grappling prey with its flexible forearms.

We evaluated the current inventory and monitoring practices at John Day Fossil Beds National Monument (JODA) with an analysis of fossil yield based on locality data. While JODA covers a range of fossiliferous Cenozoic rock, most paleontological fieldwork conducted by the park staff occurs in the highly fossiliferous Oligocene Turtle Cove Member of the John Day Formation. To optimize the number of fossils collected, JODA established a schedule that cycled fieldwork every four years between the five most fossiliferous areas of the park. We digitized over 1000 field collections from 1999-2019 to evaluate whether the four-year return interval allowed enough time for new fossils to be exposed through erosion in an area between visits as intended by the schedule. We found no significant difference in fossil yield between areas that had not been visited for 1, 2, 3, 4, or 5+ years. Based on these results, we infer that the claystones of the Turtle Cove Member are erodible and fossiliferous enough that one winter of erosion will expose enough fossils to fully "recharge" an area that had been thoroughly inventoried the previous year. Because of limited staff and high fossil yield, not every fossil will be collected. Therefore, we use this analysis to shift JODA's paleontological inventory and monitoring practices away from a rigid schedule and propose a flexible new system. The paleontology staff will use a tracker geodatabase of locality data from past years to set priorities each year based on past collection and current available staff and space.