X-ray tomography as a non-destructive tool for evaluating the preservation of primary isotope signatures and mineralogy of Mesozoic fossils

The stable isotope compositions of carbonate and phosphate components in fossil teeth and bone are widely used to infer information on paleoclimate and the physiology of extinct organisms. Recently the potential for measuring the body temperatures of extinct vertebrates from analyses of ¹³C-¹⁸O bond ordering in fossil teeth has been demonstrated (Eagle et al. 2010). The interpretation of these isotopic signatures relies on an assessment of the resistance of fossil bioapatite to alteration, as diffusion within, and partial recrystallization, or replacement of the original bioapatite will lead to measured compositions that represent mixtures between primary and secondary phases and/or otherwise inaccurate apparent temperatures. X-ray computed tomography (CT) allows 3-D density maps of teeth to be made at micron-scale resolution. Such density maps have the potential to record textural evidence for alteration, recrystallization, or replacement of enamel. Because it is non-destructive, CT can be used prior to stable isotope analysis to identify potentially problematic samples without consuming or damaging scientifically significant specimens. As a test, we have applied CT to tooth fragments containing both dentin and enamel from Late Jurassic sauropods and a Late Cretaceous theropod that yielded a range of clumped isotope temperatures from anomalously high ∼60^oC to physiologically plausible ≤40^oC. This range of temperatures suggests partial, high-temperature modification of some specimens, but possible preservation of primary signals in others. Three-dimensional CT volumes generated using General Electric Phoenix|x-ray CT instruments were compared with visible light and back-scattered electron images of the same samples. The tube-detector combination used for the CT study consisted of a 180 kV nanofocus transmission tube coupled with a 127 micron pixel pitch detector ( ∼3-12μ m voxel edges), allowing us to clearly map out alteration zones in high contrast, while reducing edge effects and beam hardening artifacts. CT images of these teeth show a range of replacement textures. One tooth -- thought to be the least altered -- shows only localized positive density anomalies near fractures, while a second -- thought to be highly altered -- contains high-density replacement mineralization. A third tooth -- one suspected of possible partial alteration -- shows a network of rectilinear density anomalies in the enamel similar to 2-D transmitted light and back-scattered electron images. This may represent recrystallization or replacement of the primary bioapatite. Unlike 2-D imaging techniques, 3-D volumes can be used to quickly and easily make quantitative measurements of the volumes of altered and unaltered materials: For example, in the enamel of sample 3, we observe a ratio of high density to low density material of ∼ 3:2.