TAPHONOMICALLY INFORMED BODY SIZE ESTIMATION OF WOODRAT (NEOTOMA) FOSSILS FROM THE PAISLEY CAVES, OREGON, USA

Abstract Body size estimation has a long history in paleontology because size correlates with morphological, physiological, and life-history traits. For mammals, teeth, skulls, or limb bones are typically measured and multi-species estimation models constructed. Such models are less robust for small taxa. Moreover, most measurements require intact specimens, limiting their applicability to fragmentary fossils. Using museum collections, we measured features of craniomandibular elements likely to persist on fragmentary fossils based on taphonomic breakage patterns in Quaternary cave settings for three woodrat species that co-occur in Oregon’s Great Basin ecosystem: Neotoma cinerea, N. fuscipes, and N. lepida. Pairing these measurements with field weight and total length, we fitted non-linear regressions. We found that first-molar length, a commonly used body size proxy, was not the best predictor; instead, maxillary diastema length strongly predicted field weight, while the distance from the premaxilla to the distal edge of the frontal best predicted total length. Applying these models to Neotoma fossils from the Paisley Caves, Oregon, allowed us to track size changes over the last ∼ 15,000 years. Neotoma cinerea was smaller and likely more constrained in size in the Holocene compared to the late Pleistocene, consistent with climate warming. However, when the xeric middle Holocene climate ameliorated in the late Holocene, N. cinerea body size did not increase. Two congeneric competitors, N. lepida and N. fuscipes, first appear in the early and middle Holocene, respectively. We thus hypothesize that a shift in biotic interactions may have decoupled N. cinerea’s size from the direct effect of climate change in the late Holocene.