D. Croft Current Projects

Vertebrate paleontology is a broad field. Although most people associate it with fieldwork, this is really just one aspect, albeit a vital one. I like to think of vertebrate paleontology in general, and my research interests in particular, as having a hierarchical structure with studies building on one another.
At the base of the hierarchy is fieldwork: going out and collecting fossils. Fundamentally, all paleontology is based on fossils, and all fossils are the product of fieldwork. If no one had ever bothered to go out and collect fossils, we would know nothing about past life. Although fieldwork is simple in theory, it can be difficult in practice; in addition to the challenges of finding fossils where no one has ever found them before, there are the practical difficulties associated with working in a remote area and/or collecting and transporting fragile specimens. But it isn’t just finding new localities that’s important. It is equally important to revisit places where people have collected before; such repeat visits inevitably produce species that hadn’t before been recorded from that locality as well as new and better specimens of previously recorded species. The bottom line is that new fossils equal new data, and new data can result in new and more refined insights.

     The second level of the hierarchy deals with identifying specimens, describing species, and investigating phylogenetic relationships among species. These pursuits can more or less be grouped together as systematics. Systematics usually isn’t very glamorous work unless you happen to find a new ‘transitional’ species or overturn some long-standing belief of evolutionary relationships. It is, however, extremely important; all 'higher-level' investigations rely on accurate systematic investigations.
     Studying the biology of extinct species is the third level of the hierarchy; this aspect of paleobiology can also be called paleoautecology. Such studies often deal with inferring the diet, body mass, and/or locomotor style of an extinct species, though the sky is the limit in terms of habits that could be investigated. Such studies usually require well-preserved fossil specimens and/or large sample sizes, and therefore are highly dependent on research collections and systematic studies. They usually also incorporate relevant data from modern mammals, which are applied to extinct species by analogy.
     At the top level are studies that incorporate many types of data from many species and/or localities; a major goal is to understand general principles that govern species diversities and distributions. Such studies fall under a variety of headings, but the term 'macroecology' is a good way to sum them up. They are typically synthetic in that they use data generated by many other studies. They might take a ‘broad-brush’ look at changing climates over millions of years based on evolving mammal communities, or they might compare the structure of ecological communities in different continents during a particular interval of time. Large paleontological databases have helped facilitate such studies, and examining the list of references for such databases gives you an idea of the amount of research on which such studies rely.
     One of the things I enjoy about my field is that I’m able to work at all levels of this hierarchy. Below are some plain language summaries of the things I investigate, arranged more or less in the same hierarchical order. It should be kept in mind, however, that most studies and journal articles incorporate aspects of some or all of these levels and therefore can't really be placed in just one. But one area typically is the focus.
     For a more graphic version of this scheme and what I do, you can check out my 2010 poster from Research ShowCase.

Fieldwork

The field research in which I participate is mainly intended to increase geographic and temporal (time interval) sampling within South America. Most fossil-producing localities in South America are in the southern part of the continent (Patagonia), and so we've focused our efforts on the lower latitudes of Bolivia and Chile. Some of these localities have also increased the temporal sampling in South America, representing previously unrecognized or poorly sampled intervals of time.

Bolivia: My fieldwork in Bolivia is primarily undertaken in collaboration with Federico Anaya, Bolivia

although many other investigators will be involved in various aspects of our research. The main goals of our research are to: (1) collect new specimens from early and middle Miocene faunas of Bolivia; (2) describe the species that occur in these faunas based on these new specimens; and (3) compare these faunas to similarly-aged faunas in Argentina and Chile. The relationships among these 'Friasian' faunas have long been a challenge to South American paleomammalogists, and through this NSF-supported research we plan to significantly clarify patterns of mammal diversities and distributions during this interval. Of course, we also hope to discover some new localities along the way, as well as some excellent new specimens that will provide additional information about the habits of these extinct animals!

Quebrada Honda, Bolivia: The middle Miocene locality of Quebrada Honda was first discovered in

the late 1970s and was collected more intensively in the early 1980s. Although a pair of papers were published in 1987 and 1988 describing some of the mammals from there, little detailed research had occurred until a few years ago, when interest resurfaced in the marsupials of Quebrada Honda. Federico Anaya and I subsequently published a paper describing a new hegetotheriid notoungulate from this locality (based on specimens in museum collections), and I reviewed the other notoungulates in another paper published in Palaeontology shortly thereafter. In 2007 we revisited the site, collecting many more specimens, including many excellent rodents; these and other rodent specimens were the focus of the Master's thesis of my graduate student, Jen Chick. More detailed descriptions of the other mammals of Quebrada are currently underway, and will undoubtedly be collaborative efforts.

Cerdas, Bolivia: The small town of Cerdas is located near the eastern edge of the Bolivian Altiplano, Cerdas

about 60 km southeast of Uyuni. Fossiliferous badlands near Cerdas were discovered in 1972 and two fossils, both mesotheriid notoungulates, were collected from there several years later. A team from the University of Florida returned to the site in 1991 and later published a paleomagnetic section with associated radiometric dates indicating that the fossiliferous horizons mostly range from about 16.5 to 15.5 million years old. Our team returned to Cerdas in 2007 for a brief visit, and made a nice collection of fossils. In 2009 we published an overview of the fauna based on these fossils and others in the collections of the Florida Museum of Natural History. We will return to Cerdas in the next year or two to collect more fossils and, hopefully, to document more species.

Chile

Chile: Our research group (including John Flynn, Andy Wyss, and Reynaldo Charrier, among others) has been working in Chile since the late 1980's, long before I had even begun graduate school. I joined the team during my graduate studies at The University of Chicago, and have now become one of the regular participants in expeditions to Chile. Like most paleontological research, our efforts are supported by the National Science Foundation.

Tinguiririca: Perhaps the most important of our Chilean localities is Tinguiririca, which is located in central Chile near the town of Termas del Flaco. In the transition between the warm, equitable climate of the Eocene Epoch (54.8 to 33.7 million years ago) Abanico Formation

and the cooler climate of the Oligocene Epoch (33.7 to 23.8 mya), numerous 'archaic' mammal groups went extinct while newer lineages of mammals, many with extant representatives, became more diverse. Before the discovery of the Tinguiririca Fauna, a significant gap spanning this period of time was present in the South American fossil mammal record. Understanding this critical time in the evolution of mammals worldwide (known as the Eocene-Oligocene transition) is essential to understanding the historic factors that influence the distribution of animals today. Our publication on the Tinguiririca Fauna was the first to describe the evolutionary effects of the Eocene-Oligocene Transition on South American mammals and habitats, and it reinforces the idea that 'open' (i.e., non-forested) habitats appeared earlier in South America than anywhere else in the world. A more recent publication, incorporated an ecological diversity analysis of Tinguiririca and taxonomic comparisons with other faunas of similar age.

Chucal: Another Chilean locality that has been the subject of much research is Chucal, located in the Altiplano (high plateau) of northern Chile. At an elevation of some 4,500 m, it is (as far as we know) the highest vertebrate fossil locality in the western hemisphere (as well as one of the highest in the world). The Chucal Nesodon, mesothere

area was at a much lower elevation when the fossils were deposited there, and at that time the region supported a diverse and abundant mammal fauna (at least 18 different species have been collected). Comparisons with well-known faunas of potentially the same age from southern South America, however, illustrate that quite different mammal groups are represented at Chucal, thus documenting marked provinciality within the continent at the time. (Provinciality is the term used to describe the presence of different mammals in different places.) For example, the mesotheriid notoungulates are common and diverse at Chucal but are unknown in comparable faunas from southern Chile and Argentina. Similarly, interatheriids (another family of small notoungulates) and ground sloths are common in Patagonia but so far have not been recorded at Chucal. By combining the information from Chucal with that from other Chilean and Bolivian faunas, we will better understand when and how such provinciality originated and what effects it has had on modern faunas. The ungulates of Chucal were described a few years ago and we more recently tackled the cingulates (armadillos and glyptodonts).

Laguna del Laja: A recent area of focus in terms of fieldwork has been the fantastic sequence of Antuco

strata in the Laguna del Laja area of Chile. These rocks, which are set against a backdrop of a beautiful lake and a picturesque volcano, preserve a diverse array of mammal fossils. We have collected several hundred specimens from the area during our four years of fieldwork there, many of them excellently preserved. We recently published an overview of this fauna, and now are in the process of describing various components in greater detail. Based on our provisional identifications and radiometric dates, this remarkable region records at least five different Miocene time slices between about 20 and 10 million years ago.

Other central Chilean localities: Tinguiririca, noted above, is one of many fossil localities that we have discovered in central Chile. All of these apparently derive from the Abanico Formation, a group of Andy surveying outcrops

rocks that was deposited mainly between about 40 and 15 million years ago. This particular formation often preserves specimens in detail, but the matrix (the rock surrounding the specimen) is typically very hard. This means it usually takes weeks to prepare (i.e., clean) a single specimen. This obviously slows down our research quite a bit, since the fossils need to be prepared before they can be described. We have published papers on particular specimens from some of these other faunas, but most are awaiting preparation and/or further study. A current focus of our group is compiling provisional species lists for these faunas in order to make preliminary age and biostratigraphic interpretations. This is particularly important for this area, since different parts of the Abanico Formation were deposited at different times. We plan to supplement these age estimates based on fossils with radiometric dates wherever layers suitable for dating are present. Together, these faunas and their associated dates will be very useful for: (1) better understanding the tectonic and uplift histories of these part of the Andes; (2) increasing the number of 'time slices' represented by fossil mammals in Chile; and (3) expanding the geographic sampling of fossil mammal localities in South America.

Systematics

Systematics is the study of the evolutionary relationships among organisms. It is intricately linked with the identification and naming of organisms (taxonomy). Most systematists specialize Cladogram

in one group (which can include many or only a few species), and my specialty is Notoungulata. I am particularly interested in notoungulates because they represent a completely separate radiation of mammals into the herbivore niche. As such, they demonstrate adaptations for herbivory that are sometimes unique and sometimes convergent compared to those exhibited by herbivorous mammals living today. They also are very common in most fossil localities, which means there usually are plenty of specimens to study. Among notoungulates, I've mostly focused on typotheres, which are the generally smaller-bodied half of the notoungulate family tree. My systematic investigations aren't limited to notoungulates, however. Like most field paleontologists, I describe what I find, and so I've also worked on a variety of other mammals.

Archaeohyracids: Archaeohyracids are a group of small (typothere) notoungulates known from the Archaeohyrax patagonicus type

Casamayoran through Deseadan South American Land Mammal 'Ages' (SALMAs). They are generally rare animals in the faunas in which they are found, although they are unusually common in the Deseadan fauna of Salla, Bolivia, as well as the Tinguiririca Fauna of central Chile. My collaborators and I have published two papers in Fieldiana that describe new archaeohyracids from the Tinguiririca Fauna, revise a significant portion of archaeohyracid taxonomy, and include a phylogenetic analysis. These new taxa were included in the most recent summary of the Tinguiririca Fauna, in which we formally named the Tinguirirican SALMA. More recently, I published a paper with Argentine colleagues that describes a new, very small species from northwest Argentina.

Hegetotheriids: Hegetotheriids are another group of small notoungulates that are closely related to archaeohyracids. Unlike archaeohyracids, they did not go extinct at the end of the Oligocene, Hemihegetotherium

but rather persisted all the way through to the Pleistocene. The specific phylogenetic relationships between archaeohyracids and hegetotheriids are a key focus of my research, as are the relationships among hegetotheriids themselves. And since they are relatively common in many Oligo-Miocene faunas, I often run across them while doing fieldwork and therefore usually have lots of specimens awaiting description. I recently published a paper (along with Federico Anaya) describing a new species of hegetotheriid (pictured at left, illustrated by Velizar Simeonovski) from the middle Miocene (Laventan SALMA) of Quebrada Honda, Bolivia; this paper also included a phylogenetic analysis of the family. This species is the first hegetotheriid known from this particular period of time. I didn't happen to find these specimens in the field, however, but rather in the drawers of a museum in La Paz, Bolivia. We have collected some beautiful hegetotheriid specimens from Chile, but most have yet to be described.

Mesotheriids: Mesotheriids are typothere notoungulates that are related to archaeohyracids and hegetotheriids; they span a size range greater than that of archaeohyracids, with larger members generally described as being Eutypotherium superans holotype

sheep-sized. They first appear in the Deseadan (late Oligocene) and persist into the Pleistocene. The last revision of the group (actually, of one of the two subfamilies) was a study by Francis in 1965. Since then, new specimens have been collected from Argentina, Bolivia, and Chile, and several new taxa have been described. In our three field seasons at Chucal, we have collected a large number of mesotheriid specimens. We presented a preliminary review of these new mesotheriines in a short JVP article, and named three new species from Chucal in a subsequent Fieldiana paper. We later described another new species from younger strata to the south. Based on these investigations, it appears as though mesotheriines underwent a significant radiation in the middle latitudes of South America in the early Miocene. I continue to be very interested in this peculiar group of notoungulates. Most recently, Beth Townsend and I published a paper analyzing the mesotheres from Cerdas.

Relationships among notoungulates: A big question in the evolution of South American mammal faunas is how the different groups (families) of notoungulates are related to each other. Although various CT Scan

authors (myself included) have looked at particular parts of the notoungulate family tree, comprehensive analyses based on many taxa and many characters are currently lacking. One of our research priorities is undertaking a comprehensive examination of notoungulate ear and basicranial anatomy using high resolution computed tomography (CT), a technology that allows us to look not only at the external structure of the skull, but also at the internal anatomy. The base of the skull and the ear region are anatomically complex areas of the mammalian skull that have been valuable in phylogenetic studies of many groups of mammals. Although several early studies of notoungulate skulls have highlighted important distinctions in basicranial anatomy among various groups, these findings have been underutilized in investigations of notoungulate systematics. Our investigations should help clarify the evolutionary history of the Notoungulata and may also shed light on functional aspects of their ear region and other parts of the skull.

Paleobiology

As noted above, paleobiology (specifically paleoautecology) focuses on reconstructing the characteristics and habits of extinct animals. This can be especially challenging in the case of notoungulates; they have no living representatives, and it isn't always obvious which living mammals might be the most appropriate models. We generally try to deal with such issues by comparing notoungulates to a wide variety of living mammals (rather than just one group) and/or by using 'taxon-free' methods of reconstructing habits. 'Taxon-free' methods are those that are based on physical laws or other such principles and are therefore thought to be applicable across all mammals. A good example might be the height of a mammal's cheek teeth. In many different groups of mammal, such an increase in height (termed hypsodonty) is correlated with feeding on abrasive vegetation in open habitats. It is thus believed to be useful for reconstructing the habits of extinct species, even if they are only distantly related to living mammals.

Dietary Inference: Pretty much everyone agrees that notoungulates were herbivores (i.e., eating leaves, fruits, twigs, etc.) but it is less certain how important various types of plants were in the diets of different species. Because diet is a very important aspect of a mammal's biology and because dietary data can sometimes Microwear

be used to infer habitat, I am quite interested in figuring out what sorts of plants notoungulates and other extinct South American mammals were eating. One way Beth Townsend and I have been investigating this topic is by studying 'enamel microwear.' Microwear is the term applied to very small marks - such as scratches, pits, and gouges - that are left on a mammal's teeth by the food it eats. It is called microwear because the marks left on the tooth can only be seen well using a microscope. For several decades, scientists have been studying the correlation between microwear and diet in living (extant) mammals and have found that certain types of food usually leave characteristic marks. Based on analogy, if you can 'read' the microwear on the teeth of a fossil specimen, you can then get an idea of what that animal was probably eating. (There are lots of other factors involved, but that's the short version of the process.) Beth and I were the first to apply this method to notoungulates and were also the first to examine microwear in living caviomorph rodents. Our notoungulate studies have been particularly Tooth cusps

interesting because they haven't supported the traditional dietary interpretation that these animals were eating abrasive vegetation like grasses, at least in the early Miocene of Argentina. Future studies using mesowear (a different type of tooth wear) should help test the accuracy of our results. A student of mine and I worked together on a project that applied the mesowear technique to Oligocene notoungulates from Bolivia, demonstrating its efficacy for endemic South American ungulates (another first such study of its kind). Beth and I have received funding from the National Science Foundation to further explore these methods of dietary inference for endemic ungulates and caviomorph rodents so that we can better characterize the ways that ancient plant-eaters were partitioning the available resources.

Locomotion and Body Mass: Having no living representative also makes it difficult to figure out how Humeri

large or small notoungulates were and how they moved, two other important aspect of an animal's biology. So far we've taken a close look at these attributes in two groups of notoungulates. With Bruce Shockey and Federico Anaya, I examined the structure of the postcranial skeleton in mesotheriids. Our study was published in 2007 in Paleobiology. In sum, it looks as if mesotheres were habitual diggers. We don't necessarily think they were burrowers, but it seems likely that digging was used in some aspect of their lifestyle, perhaps in finding food. A similar sort of study was published by a student of mine and me in 2008, except that it focused on a single genus of interatheriids, Protypotherium. In contrast to mesotheriids, this animal was more generalized but was tending toward cursoriality. In other words, it probably was a speedier sort of animal, whose skeleton was adapted for moving quickly or relatively efficiently over long distances.

Macroecology

Macroecology means different things to different people, but for me it is a good term to describe many of the more synthetic studies I undertake. These studies try to answer very large questions; in my estimation, these are some of the most interesting ones in paleontology. More often than not, of course, Australia

these studies raise as many questions as they answer. But bit by bit they help refine our understanding of past communities. One of the things that makes South America a great place to study these sort of questions is that it acted like a giant island for much of the past 65 million years. Due to their isolated nature, islands have long been known as places for evolutionary innovation; more than 150 years ago, careful study of the faunas of differently-sized islands in various locations contributed significantly to the ideas of both Charles Darwin and Alfred Wallace, the co-founders of the idea of evolution via natural selection. Although continents are seldom thought of as islands, Australia (today) and South America (until a few million years ago) both demonstrate unique faunas, typical of isolated land masses. In contrast to other continents – where immigration and dispersal have had significant roles in shaping and re-shaping the mammal fauna over the past 65 million years – in South America and Australia, autochthonous origination has been the norm. These island continents are therefore particularly well-suited for testing macroecological models based exclusively (or almost exclusively) on data from other continents. By testing these models with data from the fossil record, I hope to use paleocommunities to better understand the development of modern communities and patterns of species diversity.

Paleoenvironments: Reconstructing past habitats sort of falls under the category of macroecology, especially when the data used to reconstruct such habitats are the mammals themselves. One type of such macroecological model is a 'cenogram' analysis. Using data from modern faunas, a cenogram analysis correlates characteristics of faunal body size distributions with habitat and environmental attributes such as rainfall and vegetational structure. Cenogram analyses have been

widely used for interpreting habitats of northern hemisphere paleofaunas, but a study I published a few years ago was the first to apply this method to South American paleofaunas. One of the most interesting results of this study was that the habitat interpretations based on cenogram analyses were in conflict with those based on other lines of evidence (e.g., craniodental morphology of the herbivores) for several South American localities. This suggests that these types of macroecological models may not apply to endemic faunas in places such as South America, and that previous studies using these models in other areas may or may not have reached accurate ecological conclusions. Of course, I'd now like to figure out why they do or do not work under certain circumstances. In collaboration with Beth Townsend, I am expanding our repertoire of techniques for inferring ancient habitats by using ecological diversity analysis (EDA). In 2005 I presented preliminary results of our study of the Santa Cruz Fauna of southern Argentina using EDA and at some point we plan to put together the manuscript. Additionally, a student and I have been working to expand our modern comparative dataset and to apply it to other South American paleofaunas, to better characterize the structure of mammalian paleocommunities.

Predator-Prey Diversity: A common attribute of Australia and South America throughout most of the Cenozoic is low mammalian predator diversity, consisting solely of marsupial predators. Although this Thylacosmilus

relative lack of diversity is well documented, little research has investigated what effects this might have had on prey diversity and community dynamics. At a Society of Vertebrate Paleontology meeting, I presented preliminary results from a new study examining the relationship between mammal predator and prey species diversity in modern habitats. The final version of this study was published in Evolutionary Ecology Research back in 2006. Relative diversity data from both South America and Australia strongly support the qualitative observation that marsupial predators - as a group - are not as successful as placental mammals in terms of species diversity. Abundance data suggest that marsupial predators were also rarer in their respective faunas than typical placental predators. These trends do not appear to result from differences in continental area or habitat diversity, nor do they appear to be attributable to effects of fossilization. In future investigations, I'd like to examine how such low levels of predator diversity and abundance might have affected the structure of fossil ecosystems (which is related to the issues noted above). Together with a student of mine and Gina Wesley-Hunt, I'm examining morphological disparity in these extinct marsupials and comparing them to members of the same guild in North America. This should provide additional insights into how these animals might have been partitioning dietary resources among themselves and also with the so-called 'terror-birds' (Phorusrhacids).

Island Dwarfism: One of the best examples of an evolutionary phenomenon that has occurred repeatedly through geologic time is that of island dwarfing: when a large mammal becomes smaller through evolutionary time after becoming isolated on an island. Examples of dwarf mammoths, hippos, deer, and other animals abound, but this phenomenon had never previously been reported in the cattle group (Bovinae in technical terms). Thanks to my connections at The Field Museum, I became involved in Cenogram

a collaboration with the unique opportunity to describe some bones of an extinct species of dwarf water buffalo from the Philippines (see reconstruction at right by Velizar Simeonovski). These bones had been discovered almost 50 years ago by Michael Armas, a mining engineer, in a phosphate mine on Cebu Island, Philippines. He collected them and kept them safe for nearly four decades until they came to the attention of Dr. Hamilcar Intengan, a physician. Dr. Intengan recognized the importance of these fossils and in 1995 brought them to the Field Museum for scientific study. After comparing the bones with those of the modern water buffalo, the tamaraw (an endangered Philippine water buffalo), and an anoa (a more distantly related buffalo from Sulawesi), we verified the distinctiveness of the bones and estimated that this tiny buffalo would have stood only 2.5' high at the shoulder and probably would have weighed a mere 350 lbs. The full report was published in the Journal of Mammalogy and was featured as an open access article by BioOne. Because the Philippines has a relatively poor fossil record, we are very interested in finding additional mammal fossils there.

This page was last updated on April 17, 2010.