Natural history collections in the digital age

  • For many, a trip to a Natural History museum is a childhood rite of passage – and for some, it inspires an interest that lasts a lifetime. Dr Gregg Gunnell is director of the Division of Fossil Primates at the Duke Lemur Center in Durham, North Carolina. He is passionate on the subject of natural history and his work highlights the importance of maintaining existing collections. Similarly enthusiastic about knowledge sharing, Dr Doug Boyer, a colleague of Dr Gunnell’s in the Department of Evolutionary Anthropology at Duke, has developed MorphoSource, a web archive for storing and accessing 3D models of specimens. This archive makes natural history collections accessible for people across the globe.

    Our planet is over 4.5 billion years old. Life has existed on Earth for about 4 billion of those years, starting out in the form of single-celled organisms and evolving from there to where we are today. The earliest appearance of our species, Homo sapiens, is understood to have been about 200 thousand years ago, and “modern humans” (who are distinct from others of their species by their more complex and “intelligent” behaviours) appeared more recently.

    Fossilisation, the Earth’s natural method of preserving remains of organisms, provides information on species that are now long-extinctQuote_brain

    Despite being relative newcomers to life on Earth, unlike any other species, we have managed to gain an understanding of the planet’s natural history, even from long before our existence. The skeletons, shells and outlines of ancient animals from millions of years in the past have been preserved as fossils. Just as the earth preserves remains of deceased animals for eons, society strives to immortalise collected specimens for eternity in natural history collections. These collections provide us with an insight into past and present biodiversity, acting as snapshots of life on Earth, and can be used to inform future conservation of species and diversity as we know it today.

    A bat skull and jaw from three different angles – even with small specimens, micro-Ct images provide astonishing levels of detail

    Recording the Past
    Fossilisation is the Earth’s natural method of preserving the remains of organisms that once lived on its surface. These fossilised remains provide us with detailed information about species that are now long-extinct. For example, a fossil of an animal can tell us where and when that animal lived, and how it lived by its morphology (its shape): by examining its teeth, we can identify what it ate and by examining its bone structure we can determine how it moved and to which present species it may be related. We can also identify closely related species through examining teeth and other skeletal elements, as recently diverged species share similar dental and body patterns. This can then help us to identify where certain species originated and how they have evolved over time, and aid in providing us with an understanding of the environmental conditions of the past.

    Like the species that inhabit it, planet Earth has evolved and changed tremendously over billions of years, and studying fossils helps us to piece together and understand this journey. For example, if a fossil of a species is found in southern Africa, and another identified to be from the same period is found in South America, this tells us that at that time, these two continents were most probably positioned close together.

    Drawer of vertebrate fossils from La Venta, Colombia properly organised and stored

    Natural History Collections
    The physical and chemical conditions must be just right for the Earth to preserve a deceased animal until it is found by a palaeontologist, so much so that only a vanishing fraction of all organisms become fossils. Similarly, museums have unforgivingly precise protocols for preserving specimens in natural history collections: each type of specimen has a different protocol depending on its composition and what is valuable about it. Such specimens can range from plants to animals to fungi, and contain valuable physical data, including DNA and the species’ physical morphology. This information is vital for understanding various ecosystems and how they have been affected by certain events, which will then help us in conserving and protecting these ecosystems and species, should similar events occur again.

    These physical specimens are extremely valuable to natural history research and are kept in museums and universities around the world, providing the public, students, and researchers alike with an insight into past and present biodiversity.

    Dr Doug Boyer, developer of MorphoSource and Assistant Professor in the Department of Evolutionary Anthropology at Duke University

    Our focal example is the Division of Fossil Primates (DFP) at the Duke Lemur Center, headed up by Dr Gregg Gunnell. The collection contains over 35,000 specimens of primate and other vertebrate fossils, principally from three crucial places and time slices in the world. These places include: 37 to 20 million-year-old samples from North Africa (Egypt) where fossil evidence documents the origin and diversification of anthropoid primates (today’s monkeys, apes and humans); 12 to 13 million-year-old rocks in Colombia where the diversification of New World Monkeys can be documented and studied; and subfossil samples from the island of Madagascar that document the devastating impact of human arrival on the island some 3,000 years ago. This range of specimens is carefully conserved and is available for researchers and students to access. In addition, DFP have an ongoing fieldwork programme in place to collect additional specimens from Egypt, Madagascar, Indonesia and Wyoming.

    Each collection like the one at DFP is unique and therefore critical for our understanding. The catch 22 of uniqueness, is that it is very difficult for a lot of scientists or the public to benefit from the information they contain. Access and use must be restricted and monitored to protect the specimens. Researchers might have to travel thousands of miles to see them. This means we don’t have a good way of knowing how valuable the material ‘would be’ if more people could afford to come and study it.

    MorphoSource is a web archive for storing and accessing 3D models of specimens so they are accessible across the globe, aiding paleontological education and researchQuote_brain

    For many research purposes, there are no easy solutions here: accessing physical specimens is essential for sequencing a copy of the species’ DNA, or documenting its pelage characteristics. However, some research questions can actually be answered more effectively through study of a ‘virtual specimen’: a computer file that emulates the physical specimen in 3D on your computer screen in a way that lets you take measurements of the size and shape of anatomical structures, and even see inside it with microscopic resolution and macroscopic clarity. Though digitisation does not mean we can discard the physical specimens (they serve as the gold-standard of the digital information), it does mean the specimens can become more widely accessible. The push to 3D digitise natural history collections is catalysing a profound change in the methods and rate of study. Frankly, the blockades to access mentioned have been crippling. One upshot is that specimens from different museum collections could rarely, if ever, be directly compared except through photographs and simple measurements.

    A Cambrian trilobite from Australia

    Natural History in the Digital World
    The world is more connected now than it has ever been, with knowledge and information available and able to be shared rapidly and easily. Because of this, digitising and digitally serving natural history collections stands to have a transformative impact on natural sciences. Here, DFP is taking the lead. Dr Doug Boyer, a colleague of Dr Gunnell, is the primary developer of MorphoSource, a web archive that allows for the storage and sharing of both raw x-ray CT-scan data and 3D models of specimens from DFP and elsewhere. Users can study the physical traits of an organism without physical access to the specimen itself. And because most of the specimens are digitised by Micro-Ct scanning technology, not only the surface morphology but also the internal anatomy of specimens can be analysed and studied. In addition to digitising the DFP collection, MorphoSource also allows researchers and other museums to share their own data, including any 3D data representing a museum vouchered natural history object even if it is from a living animal, or an archaeological artifact.

    Such 3D data is already collected in research, but is rarely published or made available; MorphoSource provides a data-sharing platform which allows results from studies to be verified or replicated. It can be thought of as a virtual museum that houses 3D versions of all the world’s most important fossils in a single place. Over 150 different museums from around the world are represented by specimens on MorphoSource.

    Protecting the Future
    Natural history collections are extremely important in preserving and understanding past and present biodiversity; they give us an insight into the structure of past biological communities and how they have changed over time. They allow researchers to identify how communities have responded to past events, giving a valuable insight into possible outcomes for such events in the future and ensuring that preparations can be made to protect biodiversity. Integrating global datasets from collections around the world provides valuable educational material, opens new opportunities to people worldwide and advances our understanding of species and biodiversity. DFP, MorphoSource and other natural history collections are committed to aiding the global research effort in this area.
    But why should we care about biodiversity – what does it matter if species go extinct? Aren’t they going extinct all of the time?

    Biodiversity is really the end-product of a constantly changing and adjusting evolutionary process – some extinction is natural but it happens within the context of the whole community that is evolving as one enormous and complex entity. Our natural history collections and the study of the natural world have begun to reveal the complex interplay between species and the important roles that species play both in regard to each other and as part of the whole. Natural history collections document that this process of inter-dependence has been on-going virtually since life first appeared on the planet. Importantly, though, the unnatural loss of species (i.e., not normal extinction but that caused by outside influences) resets the delicate balance and may have cataclysmic results (witness the Permo-Triassic extinction for example where nearly 95% of all life was wiped out). This reset the direction of future evolutionary development in a much different way than previous to the event. We would not even understand this concept without natural history collections.

    Building on that, study of natural history confirms that by even the most conservative estimates, human activity has accelerated the pace of extinction by 50–1000 fold over ‘background extinction’ (see Ceballos et al. 2015 in PNAS, or Johnson et al. 2017 in Science). As Dr Gunnell succinctly puts it: “Humans rely on the productivity of Earth’s ecosystems and we destroy them at our own peril without reliable knowledge of which losses will trigger trophic collapses.”

  • What are some of the processes you have in place for preserving and maintaining a collection of fossils?
    Preserving fossil collections is a constant battle against attrition – we try to understand the most critical threats to fossils (humidity, heat or cold, physical damage from handling, exposure to oxygen etc.) and then try to minimise those threats. In some cases we use silica gel and dehumidification systems to reduce the threat of humidity. We use non-toxic adhesives to help consolidate specimens. We use archival paper, storage containers and padding to help protect specimens and to preserve documentation of locality and collection information – the paper documentation is also digitised as are locality maps and photographs.

    What are some of the current challenges facing natural history collections?
    The biggest challenge is always funding levels. Unfortunately, many administrators at colleges, universities and museums are charged with minimising expenses and maximising profits and natural history collections (and libraries) are drains on resources because they normally require more than they produce in terms of funding. This puts many at risk of being closed and the collections either being orphaned to another institution if they are lucky or simply being disposed of if they are not. The loss of information in the latter case is catastrophic because in almost every case the collections could never be replaced or reproduced.

    A more positive challenge that we are currently undertaking is integrating and synchronising records with other collections around the world in order to improve their accessibility and research value.

    How important have natural history collections been for your work?
    Crucial – I could not have accomplished much of anything in my career without natural history collections. The fossil record is the one true window we have into the past – if all we knew of our natural world is what lives on planet Earth today, who among us would have ever guessed that animals like dinosaurs once roamed across the globe – yet they dominated life on Earth for ~100 million years. Without their existence the evolutionary story would have played out entirely differently and maybe we wouldn’t be here today discussing the impact of dinosaurs or any other past life. I think it is folly to pretend that we know everything – why wouldn’t we want every available resource at our disposal to try and understand the complexities of this planet we live on?

    What do you believe the future holds for Dr Boyer’s MorphoSource?
    I believe and fully expect that MorphoSource will lead to a revolution in the way organismal scientific investigation is done. GenBank pointed the way, making data and results of molecular studies widely and easily available to all – this legitimised molecular sciences because published results could be tested, replicated and challenged if need be – it made the whole science healthier. MorphoSource will do the same for organismal research – by making data and results easily available the same sort of legitimising of published results can be furthered. For far too long fossil and other natural history collections were jealously guarded by those that had amassed them – it was often impossible to gain access to specimens. MorphoSource makes access to specimens easy and available for everyone with a computer and internet access – that can only be a positive thing.

    I think it’s important to emphasise that while MorphoSource can do for phenomic research what Genbank did for genomics, it potentially has a much broader scope and more fundamental mission. Study of anatomy was a gateway science for naturalistic explanations of biodiversity. Today it is a gateway science for even the youngest school children who recognise dinosaur names before they can finish their ABCs. Tomorrow, study of anatomical variation will be the next great challenge as genomics for its own sake plateaus without being compared against high fidelity quantifications of physical structure (the products of the genome and natural selection).

    Unlike Genbank, MorphoSource can inspire in young minds a fascination with diversity and continuity of Earth’s species. For the first time the public can set its eyes upon the rawest evidence of evolution and come to their own conclusions. They can finally see what Darwin, and Cuvier, and all the natural scientists have seen behind the closed doors of physical museums.

    What are some of the advantages of examining a 3D model over examining a physical specimen?
    There are several advantages – 3D models can be manipulated easily in digital space so that even the tiniest bones can be studied in detail. Because most of the 3D models produced in MorphoSource are digital renderings produced from x-ray slices of the original specimens, not only can precise modelling of the outer surface of a specimen be visualised but all details of the internal structure can be visualised as well without any damage to the original specimens – no more need to physically cut a specimen to see inside. Having 3D models and raw slice data available on MorphoSource means that a researcher in China, for example, can have access to specimens here at DFP without having to spend the thousands of dollars required to travel to Durham, North Carolina. Also, having digital models minimises the number of times that original specimens need to be handled and examined which reduces the chances of breakage or other damage.

    Other benefits include the use of automated algorithmic machine learning approaches for studying patterns of shape in bone and comparing it to data on genomes, ecology, etc. This not only improves the rate and thoroughness of assessment but also allows a much broader community to study bones and anatomy. It literally unlocks these collections for study in a way never before possible.

    On the flip side, when it comes to promoting learning and student engagement, nothing beats a physical specimen. Thanks to advances in printing technology, hundreds of teachers around the world are printing out 3D specimens including DFP specimens and specimens representing human ancestors available on MorphoSource.

  • Research Objectives
    Dr Gunnell’s work focuses on the collection, management and preservation of vertebrate fossils while his research aims to understand the origin and diversification of modern mammalian groups by study of their fossil record. Dr Boyer’s work focusing on the evolution of primates uses the fossil record as a key resource for understanding the interplay between form and function in evolutionary contexts. He founded MorphoSource and manages its development and governance.


    • National Science Foundation (NSF)
    • Duke University

    The Duke University Lemur Center; Dr Anne D. Yoder, DLC Director and research collaborator; Dr Richard F. Kay, Duke Department of Evolutionary Anthropology; Dr Hesham M. Sallam, Monsoura University, Egypt; Ms Catherine Riddle, DFP; Ms Vicki L. Yarborough, DFP; Tim M. McGeary, Digital Technology Services; Seth Kaufman, Whirligig Inc.; Dr Nancy B. Simmons, American Museum of Natural History; Dr Erik R. Seiffert, University of Southern California

    Gregg F. Gunnell received a PhD from the University of Michigan in 1986. He is Director of the Division of Fossil Primates at the Duke Lemur Center in Durham, NC and has edited three books and published over 120 scholarly papers. Doug M. Boyer received a PhD from Stony Brook University in 2009. He is currently an Assistant Professor in the Department of Evolutionary Anthropology at Duke University in Durham, NC, USA.

    Dr Gregg Gunnell
    Director, Division of Fossil Primates,
    Duke Lemur Center
    1013 Broad Street, Durham
    NC 27705, USA
    T: +1 919 684 9820

  • Natural history collections in the digital age

Leave A Reply