How a scientist cut his teeth

I was studying the dental structures of primates at the University of Montpellier II at the time. One day my PhD supervisor gave me a box containing beautiful primate jaws and told me that I could cut and polish whatever I wished to carry out my research. Having done so, I was then asked to return the fossils to the collection responsible. To my horror, and due to a misunderstanding between my supervisor and the collector, it turned out that they weren’t meant to be cut. I had destroyed the teeth in 10 almost complete jaws that had been gathered over a period of decades, and vowed to never again use destructive techniques to investigate an important fossil.

After trying other techniques, I heard about the ESRF from a regular ID19 user at INSA in Lyon. At the end of 2000, he agreed to perform a test for me during one of his own experiments to see whether microtomography could resolve the microstructure of dental enamel. Afterwards, José Baruchel, head of the ESRF imaging group, helped me to prepare what was probably the first-ever proposal to scan fossils at a synchrotron, and the subsequent study into the 3D structures of modern and primate teeth was so successful that others followed rapidly. The ID19 team started to give me the beam time that nobody wanted, the isolated nights during weekends and vacations, and it soon became clear that synchrotron imaging held huge potential for palaeontology. I invited colleagues to collaborate or loan samples, and almost every time results exceeded expectations.

Paul Tafforeau focuses in on a fragment of an uncoiled ammonite’s body chamber (Credit I. Montero, ESRF).

Going large

In 2003, a need to image larger fossils brought the lower jaws of fossil apes (hominoid primates) to ID17. Soon afterwards came a proposal to perform the complete imaging of the oldest known hominid skull, Sahelanthropus tchadensis (Toumaï) from Chad aged 7 million years (pictured). Fossil imaging on ID17 began in earnest, although as predicted by my physicist colleagues the initial experiment on the skull failed due to the lack of penetration of X-rays through this very dense fossil.

After three years of regular tests and technical advances on ID19, and two years since almost cancelling the original aim of my PhD, I was finally able to image the microstructure of primate dental enamel without resorting to cutting tools. By that time, at the end of 2003, we had performed 13 experiments on ID19 and ID17 plus a few on BM5, including primate teeth, fossil algae and the first imaging of fossil insects in amber.

It was then that the ESRF’s management decided that synchrotron studies in palaeontology were of major importance, and in 2004 I began a one-year position as junior scientist that was opened in the imaging group. Constant exchanges about tomography development soon revealed the machine’s potential, and new beamline equipment and processing tools installed for material sciences, biology and industry were increasingly tested on fossils. During that year, 3D imaging of fossil inclusions in opaque amber also took off, with a masters student attached to the project followed by a PhD and later a postdoc position. Although the vast majority of amber is studied using photonic devices, the ESRF’s phase contrast technique was so successful in revealing inclusions in opaque amber that it became an international hub for this activity.

Before 2004 was over, a long and important collaborative project was established to study the dental development and dental structures of fossil humans and apes – some images of which have recently been added to the ESRF’s open-access database of fossil scans. At that time, each scan took several hours followed by at least three weeks of manual processing, and the failure level was around 80%. Nowadays, scans still take still several hours but for an imaged volume eight times larger, while processing takes around a week and the failure level has halved.

Going public

In 2005, palaeontology was officially represented at the ESRF panels with the creation of the Environment and Cultural Heritage committee, and activity shifted up a gear. The next phase was driven by imaging large fossils at ID17, where the failure of the first experiment on Toumaï had prompted intensive research to allow high-quality imaging of large fossils. Experimental and mathematical tricks saved the day, allowing the best pictures ever produced of fossils 10–20 cm in diameter with just 0.025% of the beam being transmitted through the sample versus a theoretical requirement of 10%.

The Poitiers palaeoanthropology laboratory then opened a two-year position to allow the complete scanning of the Toumaï skull, during which I was permanently attached to the ESRF. The result was the highest-quality scan of a complete fossil hominid skull and the work also drove new processing tools adapted to meet the high quality control of fossil imaging. Since that time, palaeontologists have developed their own specific methods, many of which have found use for other ID19 and ID17 users.

After the final Toumaï skull data had been reconstructed in late 2007, the ESRF opened a five-year scientific position for palaeontology within the imaging group. That position was made permanent in 2010, and a major refurbishment project of ID19 and ID17, strongly driven by palaeontological activities, was officially started. The construction phase of that project will start in a few months and will push much further the possibilities of non-destructive imaging on fossils, as well as benefiting many other user communities.

X-ray microtomography reveals the jaws and teeth of 65 million-year-old ammonites (Credit: I. Kruta, MNHN).

In just a decade, palaeontology at the ESRF grew from being a basic user study to a regular producer of high-impact publications that today takes up a quarter of ID19’s activities and half of the entire laboratory’s computing resources. There is currently only a single official palaeontologist at the ESRF, but through many collaborative projects with external institutes we now have two postdocs, one PhD researcher and one visiting scientist on ID19, with further positions planned for the near future. Thanks to the audacious decisions of management, along with the support of José Baruchel and many others, the ESRF became the world’s leading institute for high-quality 3D imaging of fossils – all without breaking a single bone.

Matthew Chalmers

This article appeared in ESRFnews, June 2011. 

To register for a free subscription and to rapidly receive the current issue, please go to: