Researchers hoping to study the brains of ancient animals have until now had to make do with computer-generated reconstructions. A number of groups have been using computed tomography (CT) to generate reconstructions of the brains of extinct animals. Lawrence Witmer and his colleagues at Ohio University, for example, have used CT to reconstruct the brains of dinosaurs and to make inferences about how they might have behaved.

Alan Pradel, of the Museum National d'Histoire Naturelle in Paris, and his colleagues, used the same method to examine four fossilized skulls. The specimens were excavated in Kansas, and later found lying in a drawer in the museum in Paris. They are unique, in that they are the first to be found undamaged - to date, all the other skulls from these particular species were found flattened, and so could not be used for such reconstructions.

The CT scans revealed that one of the skulls, which had been excavated in Kansas, contained "a strikingly brain-shaped structure", and so the researchers decided to examine  it more closely with a more powerful technique. The specimen was taken to the European Synchrotron Radiation Facility (ESRF) at Grenoble, which is one of the largest particle accelerators in the world. There, it was subjected to a technique called absorption-based X-ray holographic tomography, which uses incredibly bright light beams and can produce images with an atomic scale resolution.

The brain of S. denisoni was found to be an elongated, symmetrical structure with an intact midbrain, medulla, and cerebellum. (It is shown in yellow toward the end of the film clip above). The brain is so well preserved that several cranial nerves are also intact. The optic tracts, which would have originated in the retina and functioned to transmit light information, project into large optic lobes from foramina (openings) in the side of the skull, which are positioned where the eyes would have been. The oculomotorius nerves, which would have controlled the movements of the eyes, are also present.

The large optic lobes are consistent with the size of the eyes, which can easily be inferred from the dimensions of its eye sockets. In contrast, the auditory regions of the brain are small, in keeping with previous studies of the inner ear, which have shown that S. denisori and closely related organisms had semicircular canals which were grouped together in the horizontal plane, rather than in three different planes, as in modern mammals. This suggests that S. denisori could detect lateral, but not vertical, water movements. Conspicuous by its absence, however, is the forebrain; instead, there is a vague blade-shaped structure which protrudes from the front of the brain on only one side. 

The brain is tiny in comparison to the skull, measuring just 7mm in length and 1.5 mm wide, but exactly why is unclear. It is possible the organ shrank before the process of fossilization began, but the authors argue that this unlikely, because the cranial nerves are still attached to the corresponding openings in the skull.  To explain the big discrepancy between the size of the brain and that of the skull, they suggest instead that the skull may have continued to grow after the brain was fully developed, as in extant ratfishes.    

Chemical analyses showed that the brain contains calcium phosphate, which may have been deposited by bacteria that covered the surface of the brain before it began to decay.  These microbes, combined with a lack of oxygen inside the skull, and the acidity of the fatty molecules in the tissue itself, may have provided the environment in which the brain could have mineralized and remained so well preserved, from deep time until the present.

Related:

• Dinosaur brain & behaviour
• 2,000 year-old fossilized brain
  

Pradel, A. et al (2009). Skull and brain of a 300-million-year-old chimaeroid fish revealed by synchrotron holotomography. Proc. Nat. Acad. Sci. DOI: 10.1073/pnas.0807047106.