EVALUATING TOOTH SIZE AS A METHOD OF IDENTIFICATION AND ASSESSING CHANGES IN BODY SIZE AND MASS DISTRIBUTION IN OLIGO-MIOCENE EQUIDS FROM JOHN DAY FOSSIL BEDS NATIONAL MONUMENT, OREGON, USA

Tooth morphology and size have long been used to differentiate equid taxa in museum collections and scientific literature. However, the accuracy of using tooth size for identification remains largely unexamined. We measured the anterior-posterior length of the molariform lower premolars and molars (p3-m2) of equids identified in the John Day Fossil Beds National Monument collection as Archaeohippus, Miohippus, and Parahippus (n= 57) to evaluate the identification potential of tooth size. An elbow test revealed the optimum number of clusters as three, representing three tooth size classes. A k-means clustering analysis revealed specimens from each genus spread across multiple size classes, with the mid-sized cluster alone containing Miohippus, Archaeohippus, and Parahippus specimens. This suggests that examining tooth size alone is an unreliable method for distinguishing between these genera. However, valuable insights can still be derived from tooth size. Irrespective of taxonomic challenges, changes in mean equid body mass over time and weight distribution within coeval equids can be assessed using lower teeth as a proxy for body size. To quantify change in body mass from the Eocene-Oligocene boundary through the Mid Miocene Climatic Optimum, we performed a Welch Two Sample t-test on specimens from the John Day (n= 58) and Mascall (n= 50) formations to determine if significant changes occurred in mean body size among equids between these formations. The t-test indicated that equids from the John Day Formation were statistically significantly smaller than those from the Mascall. To assess mass distribution within equid populations within the John Day and Mascall formations, an elbow test was used to determine the optimal number of size classes, followed by a k-means cluster analysis to identify the size class means and assess how specimens cluster around the means. In both formations, three size classes were present. In the John Day Formation, horses of the small size class accounted for 48%, medium for 43%, and large for only 9% of specimens. In the Mascall, the small size class constituted 24%, medium 36%, and large 40%. These results suggest that two notable changes occurred between the John Day and Mascall formations: an increase in the mean size of equids and a shift in the prevalence of larger equid size classes.