Mammalian paleoecology at Minnesota: maps, molecules, and morphology

The degree to which environmental change or biotic interactions control the evolution of ecosystems on long timescales is a central focus in paleobiology. One set of theories holds that ecological interactions within communities are the primary factors that drive community evolution and species evolve by adapting through their interactions with the other species around them, which are also evolving in response to those interactions. The main theory in this camp is the Red Queen hypothesis in reference to the race in Lewis Carroll’s Through the Looking-Glass, about which the Red Queen said to Alice, “Now, here, you see, it takes all the running you can do, to keep in the same place. If you want to get somewhere else, you must run at least twice as fast as that!” Alternative theories propose that community change is driven by species adapting or responding individually to unpredictable changes in non-biological conditions, whether climatic or tectonic. Given their focus on changes in the environment that are effectively random in timing and magnitude, these are often referred to as Court Jester hypotheses in reference to the unpredictability of the clowns employed to entertain royalty. Professor David Fox is a vertebrate paleoecologist and paleobiologist whose research focuses on understanding the connections between changes in climate and habitats through Earth history and changes in the species composition and ecological structure of organismal communities to test the relative roles of the Red Queen and the Court Jester in the vertebrate fossil record. As we await the full extent of anthropogenic climate change over the next few human generations, understanding past connections between environmental change and biotic response gains currency as a means of understanding the potential impact of impending changes on today’s biota.

Research in Fox’s group since he joined the faculty in 2002 has focused on case studies in which independent environmental records of climate and/or habitat change can be developed and compared to records of lineage and community response in terms of ecological organization, morphological evolution, and geographic distributions. The main focus of this work has been on mammal communities in the context of the origin and evolution of the grassland ecosystems of the Great Plains of North America over the last 25 million years, but Fox also has continuing collaborations at fossil sites in eastern Africa that range in age from about 20-15 million years ago and preserve diverse vertebrate faunas, including some of the earliest fossil apes, and a complex history of habitat change that is still being worked out.

A starting point for all of this is mammalian biogeography. Work by Fox and collaborators, including both graduate and undergraduate students, has shown that the ecological diversity of modern North American mammals in terms of numbers of species classified by diet, body mass, and locomotor mode, has strong quantitative relationships to climatic and topographic gradients at the continental scale. This suggests that the ecological structure of mammal communities should be sensitive to climatic change over geological time (i.e., the Court Jester) even if it does not rule out a role for biotic interactions. This work is currently advancing in two directions. First, Nora Loughlin, a PhD student in Fox’s group, is using paleoclimate simulations for key intervals over the last 20,000 years (e.g., the Younger Dryas, the Bølling-Allerød interstadial) to model geographic ranges for all North American mammals, including extinct megafauna, and examine how mammalian functional diversity changed since Last Glacial Maximum across the continent in response to climate change and the megafaunal extinction. Second, Fox and another PhD student, David Birlenbach, are using a published database with species lists for over 2,000 North American Miocene (23-5.3 Ma) fossil sites to test whether and when the Rocky Mountains and the Basin and Range became a barrier to westward dispersal of mammals from the Great Plains, which would have affected species richness of mammals through time at the continental scale.

One key to testing the relative roles of the Red Queen and the Court Jester in shaping the evolution of communities is reconstructing the habitats of past communities through time. A primary way Fox and his students do this is using the ¹³C/¹²C ratio of authigenic carbonates in fossilized soils (paleosols) and the teeth of fossil mammals. Trees, shrubs, and cool growing season grasses use the Calvin cycle or C₃ photosynthetic pathway that evolved originally in cyanobacteria, which are the forebears of the chloroplasts in plant cells. Warm growing season grasses use a modified version of the ancestral photosynthetic system, the Hatch-Slack or C₄ photosynthetic pathway, which evolved multiple times independently from C₃ ancestors in the grasses and flowering plants more generally. The anatomical and biochemical differences in these photosynthetic pathways result in very distinct ¹³C/¹²C ratios in plants that use each pathway, and the strong carbon isotopic difference between C₃ and C₄ plants are transmitted faithfully to authigenic soil carbonate and the tissues of mammalian herbivores. The carbonates provide direction information about the plants that were growing in the soil, and the mammal teeth provide information on the diet of individual species and how communities divided up resources.

Fox’s work in the Great Plains has shown that the modern, C₄-dominated grasslands evolved in a two-stage process. First, forest or woodland habitats were replaced by C₃ grasses and a modest but persistent fraction of C₄ grasses during the Early Miocene (25-20 Ma), then C₄ grasses largely replaced C₃ grasses during the late Miocene to early Pleistocene (7-2 Ma). However, various lines of evidence suggest that habitats were not as homogeneous in the region during the Miocene as they are today. Another of Fox’s PhD students, Clark Ward, is in the early stages of a project to examine habitat partitioning by mammals in the region during the Miocene using ⁸⁷Sr/⁸⁶Sr ratios in fossil mammal teeth, which inherit the Sr isotopic composition of local surface rocks via the plants the mammals eat. Clark will use the ⁸⁷Sr/⁸⁶Sr of paleosol carbonates across the region as a framework to interpret variations among co-occurring species at Miocene fossil sites. In related work, Fox is part of a large international team working in eastern African with Kenyan and Ugandan collaborators that has shown that the habitats of Early Miocene apes were heterogeneous in both time and space across the region and included the earliest documented open habitats with high fractions of C₄ grasses. 

Also key to choosing between Red Queen and Court Jester hypotheses is reconstructing the ecological structure of past communities in terms of diet and body size diversity and morphological change in lineages through time. For large-bodied mammals, the size and shape of the molar teeth have long been used to infer diet and body mass, but for small mammals such as rodents, which can be extremely abundant in the fossil record, estimating details of diet for extinct species has been challenging. Fox and his students have been using 3D surfaces of the teeth, skulls, and jaws of modern rodents with spatial resolutions of 10-50 microns generated in the ESCI XRCT Lab to develop quantitative methods of dietary reconstruction and study ecological and evolutionary responses of rodents to environmental change through time. Rodents have very diverse diets and very diverse tooth morphology, but Fox and Jonathan Keller, a former undergraduate researcher in the Fox group and now a PhD student at the University of New Mexico, have shown for modern North American rodents that a small number of quantitative measures of the 3D shape of the cheek teeth can predict dietary category with about 95% accuracy, and Fox’s PhD student David Birlenbach has shown that other 3D measures can predict body mass with equal accuracy. Fox and a collaborator at the University of Oregon are now applying these methods to fossil rodents from sequences of late Oligocene and Miocene faunas in eastern Oregon and Nebraska to examine how the ecological structure of taxonomically distinct assemblages of rodents responded to similar patterns of local habitat change.

Fox’s PhD students Cessie Socki is studying the evolution of 3D shape of the skulls of the gopher and the pocket mouse lineages (Geomyidae and Heteromyidae, respectively). These closely related groups are endemic radiations in the grasslands and deserts of North America, and both have excellent fossil records. They have specialized diets and locomotor modes, including various modes of burrowing among the gophers and ricochetal hopping like kangaroos among the pocket mice, which likely evolved in relation to the appearance of both grasslands and deserts. Cessie is studying how the shapes of their skulls evolved in relation to their diets and locomotor modes and will relate those patterns to the Miocene to present history of climatic and environmental changes in the Great Plains and montane west. Fox’s newest PhD student, Theo Herring, is using measures of the 3D shape of the lower jaws of modern species in the heteromyid genus Perognathus to develop a method to identify the species from isolated jaws in late Quaternary fossil sites and will test those identifications with ancient DNA sequencing of select fossils. The goal of this project is to be able to track the modern species in space and time back to the Last Glacial Maximum to test geographic hypotheses for the divergences among the species and to test whether the species maintained their current climatic distributions under earlier climate states. Finally, MS student Hayley Orlowski is examining how the proportions of horse teeth evolved over the 56-million-year fossil record of horses in relation to the predictions of a model of mammalian molar tooth development and size (the Inhibitory Cascade Model). Hayley has amassed a large dataset of tooth measurements of fossil horses from museum visits and the literature, and preliminary results suggest that changes in the pattern of horse tooth proportions may coincide with the appearance of grasslands, thus linking developmental and evolutionary processes to patterns of habitat change.

All of these projects focus on looking for links between changes in mammals and their environments through geological time. The various projects use one or more of maps, molecules, and morphology, and the long-term goal is to bring these methods together to better understand if communities are running with the Red Queen or laughing with the Court Jester. Given the contingent nature of history, it is unlikely that only one type of model will apply to all ecosystems, but by using a diversity of approaches to these complex ecological and evolutionary problems, Fox and his group hope to make progress in understanding what drives ecosystem changes on long time scales. This knowledge may help us anticipate how organismal communities will change in the near future in response to anthropogenic climate change.