ENAMEL MINERALIZATION AND THE RECONSTRUCTION OF PALEOENVIRONMENTS: NEW EVIDENCE FROM HIGH RESOLUTION CARBON ISOTOPE PROFILES IN EXPERIMENTAL BOVID ENAMEL

Tooth enamel geochemistry has been used extensively for reconstructing rapid change in animal dietary habits, migration patterns and past climate. Microsamples have typically been collected by drilling or cutting horizontal segments from the tooth along the growth axis, with each sample spanning the totality of the enamel thickness. But amelogenesis is far more complex than what was assumed by pioneer isotopic work, and horizontal segments extracted using traditional methods consist of enamel portions that were mineralized over a large period of time. This discrepancy results in serious slaws in the reconstruction of environmental and climatic variations. Alternative microsampling strategies are required to more accurately sample the complex spatial and temporal pattern of enamel mineralization. We present the results of an experiment on steers (Bos taurus) fed a C3 followed by a C4-dominant diet. Diet was changed in the tenth month of the animal’s life and the switch was recorded in the bovid’s second molar. Six horizontal segments were sliced perpendicular to the growth axis from the top to the mid-height of the crown. Within each segment, high resolution carbon isotope profiles have been generated within the enamel thickness, using micromilling techniques. Three-dimensional mapping of the enamel’s carbon isotope composition reveals a double temporal mineralization gradient: from the upper to the lower part of the crown, but also from the enamel-dentine junction to the outer enamel. Within each profile it is possible to identify a mixture of enamel portions mineralized before, during and after the diet switch. Our results suggest that the theoretical mineralization model presented by Passey and Cerling is applicable, at first order, to bovid teeth even though more complex feeding experiments are probably needed to fine-tune the model. Although it is very unlikely that any microsampling strategy will perfectly isolate discrete time slices, future research should focus on the innermost enamel because this zone is more heavily mineralized during the first stage of mineralization and should therefore optimize the time resolution of the samples.