TAXONOMIC IMPLICATIONS OF MAMMOTH MOLAR MORPHOLOGY AS MEASURED VIA COMPUTED TOMOGRAPHY (CT)

Two Late Pleistocene species of Mammuthus, M. columbi and M. primigenius, prove difficult to separate on the basis of their third molar (M3) morphology due to the effects of dental wear. A newly-erupted, unworn M3 exhibits drastically different characters than the same tooth would after a lifetime of wear. On a highly-worn molar, the lophs that comprise the chewing surface are more broadly spaced and the enamel ridges are thicker than on an unworn molar. Mammuthus taxonomy depends on the lamellar frequency (# of lophs/decimeter of chewing surface) and enamel thickness of the third molar. Given the effects of wear, it is apparent that these taxonomic characters are variable throughout the tooth’s life. Employing static taxonomic identifications that are based on dynamic attributes is, therefore, a fundamentally flawed practice.

A new taxonomic method is needed to help resolve the relationship between M. columbi and M. primigenius. The ideal method will provide precise character measurements while offering internal inspection of the molars to account for the effects of dental wear. This study explores the utility of computed tomography (CT) in providing such a means of delineating species. Using CT, we digitized a sample of unworn teeth from both species. A model of continual wear was created by removing slices from the chewing surface to the base of the crown. At each time slice, we calculated the lamellar frequency, enamel thickness, and occlusal enamel percentage of the exposed surface of the tooth. We then examined the relationship between relative wear percentage and dental characters to determine if there was a separation between the two species of mammoth with wear.

Our results demonstrate a high degree of intraspecific variation, making a consistent separation of species difficult. In the absence of accompanying cranial morphologies or molecular data, delineation of the North American mammoth species based solely on molar morphology remains challenging, if not impossible, even with the use of CT scanning. This study’s small sample size (n = 6) allows for the possibility that our data lie within the natural variation in dental characters for the genus. Future studies should examine a larger sample size of molars from both species and from numerous populations in order to resolve whether a divergence becomes more apparent.