Being refused the right to examine a sought-after specimen is a common experience in the professional life of a palaeoanthropologist. Too often I have heard in the back rooms of museums that “nobody can find the key to the Neanderthal's cabinet”, “the fossil is away on exhibition” or “it is currently being reconstructed”. Human fossils that make international celebrities of their discoverers are difficult to find in geological strata, but they can become unreachable relics when they are in storage.

This virtual reconstruction of a hominin tooth reveals internal structures. Credit: MATTHEW SKINNER

The spread of micro-computed tomography (microCT) seemed to offer a salve to the frustrations of the field. High-resolution digital avatars of bones and teeth avoid the risks associated with manipulating original fossils. Scans allow endless reconstruction attempts, the exploration of fine internal features and the use of mathematical tools to quantify anatomical variation.

Virtual palaeoanthropology began in the 1980s, but it took decades for microCT scanners to become widely available. Today, researchers routinely bring fossils to synchrotron X-ray facilities. Many large museums have their own scanning equipment, and the transportable devices of the Department of Human Evolution (of which I am director) at the Max Planck Institute for Evolutionary Anthropology (MPI-EVA) in Leipzig, Germany, have made large international CT scanning projects possible. All these efforts have built up digital facsimiles of tangible collections.

Like many others, I believed these advances meant that issues related to consent, scarcity and distance would soon be behind us. Sadly, fluid access to the fossil record, real or virtual, remains a dream. It has become a major bone of contention in palaeoanthropology that digital data, once produced and exploited, are not made available to other researchers.

The release online earlier this week of a large series of palaeoanthropological data — produced by my department from the hominin collection of the Kromdraai B site near Johannesburg, South Africa — is an important new step (http://paleo.eva.mpg.de; see also M. M. Skinner et al. J. Hum. Evol. 64, 434–447; 2013). This collaboration between the Ditsong National Museum of Natural History in Pretoria and the MPI-EVA makes images and three-dimensional surface models of each Kromdraai specimen freely accessible. Highlights include the type specimen of Paranthropus robustus first described in 1938, as well as some never-published specimens. Researchers can also download the microCT data through a password-protected system controlled by curators of the Ditsong Museum. To move forward, the field requires such offerings to become more widespread.

The reasons that in the past constrained access to fossils — sometimes discovered a century ago — still limit the diffusion of digital substitutes today. In most cases, curating institutions retain copyright to all digital data and demand continuous control over their use. Ten years ago, a simple exchange of letters preceded the scanning of a fossil. Now, the same operation depends on the signing of long and sometimes lawyer-proofed memoranda of understanding. These agreements are primarily designed to limit the dissemination of digital data.

Among other issues, museums are wary of the risk of commercial production of printed 3D models. Especially in developing countries, bench fees for the study of specimens and the sale of casts are a significant source of income for museums.

Conflicts of interest

Importantly, many curating institutions want to manage any research that is based on data derived from their fossil collections. In theory, such control avoids overlap in research efforts. However, it also results in clear conflicts of interest, with most of these institutions hosting their own palaeoanthropological research teams. In these situations, the digital data can become a local resource to protect.

After specialists have processed and studied CT scans, one can — in theory — request copies from curators. But obtaining the data remains difficult, partly because ill-funded institutions cannot easily manage large bodies of digital information. Requests succeed sporadically for isolated and less-prestigious specimens or for low-resolution data.

Nevertheless, I am rather optimistic. The breakthroughs of virtual palaeoanthropology create pressure to allow access to numerical data, and some kind of sea change is increasingly perceived as vital for the development of the field. Some institutions already offer open or affordable access to small sets of digital data; these include the Peabody Museum of Archaeology and Ethnology in Cambridge, Massachusetts (go.nature.com/lalyis), the European Research Synchrotron Facility in Grenoble, France (http://paleo.esrf.eu), and the University of Vienna in Austria (go.nature.com/qakehi). The NESPOS initiative (go.nature.com/ehmkxw) is another positive example that provides information on available sources of CT data and some direct downloads.

The Kromdraai initiative is unique in providing direct access to an entire collection, not just a few iconic fossils. This type of collaboration between museums and producers of digital data might serve as a model for other institutions. Among other benefits, such websites give museums that need to generate income the option to implement a fee-based system for data access. Furthermore, because CT scans cannot address all potential research questions, the availability of CT data and digital avatars will probably increase requests to study the original fossils.

Easier access to digital data also depends on journals. Museums' desire to control the raw data derived from their collections will probably lead them to resist pressures from funding agencies and editors. But journals should mandate that secondary products such as surface models of hypothetical 3D reconstructions be made freely available at the time of publication.

Modern palaeoanthropological studies are driven by questions that are testable on series of specimens, not by the description of isolated fossils. Many answers require high-resolution imaging. These virtual representations are integral to scientific inquiry and should facilitate access to the human fossil record.