Evaluating bony predictors of bite force across the order Carnivora

In carnivorans, bite force is a critical and ecologically informative variable that has been correlated with multiple morphological, behavioral, and environmental attributes. Whereas in vivo measures of biting performance are difficult to obtain in many taxa—and impossible in extinct species—numerous osteological proxies exist for estimating masticatory muscle size and force. These proxies include both volumetric approximations of muscle dimensions and direct measurements of muscular attachment sites. In this study, we compare three cranial osteological techniques for estimating muscle size (including 2D‐photographic and 3D‐surface data approaches) against dissection‐derived muscle weights and physiological cross‐sectional area (PCSA) within the jaw adductor musculature of 40 carnivoran taxa spanning eight families, four orders of magnitude in body size, and the full dietary spectrum of the order. Our results indicate that 3D‐approaches provide more accurate estimates of muscle size than do surfaces measured from 2D‐lateral photographs. However, estimates of a muscle's maximum cross‐sectional area are more closely correlated with muscle mass and PCSA than any estimates derived from muscle attachment areas. These findings highlight the importance of accounting for muscle thickness in osteological estimations of the masticatory musculature; as muscles become volumetrically larger, their larger cross‐sectional area does not appear to be associated with a proportional increase in the attachment site area. Though volumetric approaches approximate muscle dimensions well across the order as a whole, caution should be exercised when applying any single method as a predictor across diverse phylogenies. This paper tests three different methods for quantifying osteological proxies of muscle size. (a) 2D muscle attachment areas measured from craniomandibular lateral photograph; (b) 3D muscle attachment areas measured from digitized crania and mandibles; Thomason “Dry skull” technique for estimating temporalis (c) and masseter (d) cross‐sections. Red = surface areas for attachments (a and b) or cross‐section (c) of temporalis; Blue = surface areas for attachments (a and b) or cross‐section (d) of the masseter. The Thomason “dry skull” approach has a greater correlation to dissection‐based muscle masses and cross‐sectional areas confirming that volumetric estimators are more indicative of muscle dimensions than are attachment areas.