Preservation assessment of Miocene–Pliocene tooth enamel from Tugen Hills (Kenyan Rift Valley) through FTIR,chemical and stable-isotope analyses

We investigated fossil tooth enamel of mammals and crocodiles from two Mio-Pliocene East-African formations (Lukeino and Mabaget Fms) using infrared spectroscopy and chemical and stable-isotope analyses. Infrared spectra indicate that the fossil enamel contains biological apatite (bioapatite), without significant secondary carbonate contaminations. Several empirical infrared indexes were used to analyze the crystal–chemical characteristics of enamel. Fossil enamel has less organic matter, water and structural carbonate of apatite than modern enamel with which it was compared. Fossil apatite has a better crystallinity than bioapatite. The calcium/phosphorus mass ratio and the fluorine content of fossil apatite show intermediate values between bioapatite and geological fluorapatite. The samples also display significant crystal-chemical variations, depending on the vertebrate group (mammals vs. reptiles) and the taphonomic context (Lukeino Fm vs. Mabaget Fm). In spite of these changes, no relationship was observed between the chemical contents (carbonate and fluorine) and the stable-isotope ratios of carbonate (δ¹³C and δ¹⁸O) in fossil enamel. Preservation of the palaeoenvironmental signals is strongly supported by the fact that the stable-isotope composition of the three investigated fossil mammalian taxa (deinotheres, equids and hippos) is consistent with their ecological features. For instance, typical C₃- and C₄-plant isotope compositions are reflected in the deinotheres and equids, respectively, and amphibious hippos display lower δ¹⁸O values than terrestrial herbivores, as expected.     

Method-dependent variations in stable isotope results for structural carbonate in bone bioapatite

Stable carbon and oxygen isotope values (δ¹³C, δ¹⁸O) were obtained for structural carbonate in the bioapatite of archaeological bones from Guatemala and Sudan using several common analytical methods. For the Sudan samples, the different methods produced δ¹³C values within ±0.1‰ and δ¹⁸O values within ±0.7‰, on average. The isotopic results for the Guatemala samples were similar in reproducibility to the Sudan samples when obtained using methods that employed lower reaction temperatures and reactions in sealed vessels. However, many Guatemala samples had highly variable and extremely low δ¹⁸O values when reacted at higher temperatures in vessels that remained open to cryogenic traps. The latter arrangement caused reaction products to be removed immediately upon their production. The anomalously low δ¹⁸O values are related to the production of a contaminant gas that causes the m/z 46/44 ratio to be lowered, either by adding to the m/z 44 peak or subtracting from the m/z 46 peak. That said, potential contaminant materials were not detectable in “anomalous” bones using Fourier transform infrared spectroscopy (FTIR), X-ray diffraction, or inductively-coupled plasma atomic emission spectroscopy. However, subtle structural and chemical differences between “normal” and “anomalous” samples were observed, most notably in the FTIR ν₂ CO₃ domain. We suggest that these changes promote volatilization of an oxyphosphorus compound and oxygen isotope fractionation between PO⁻ derived from this compound and CO₂ derived from bone carbonate. Production of the contaminant gas and the related “anomalous” δ¹⁸O values is reversible if the reaction occurs within a sealed vessel for a sufficient period of time, which allows a “back-reaction” to occur.