Implications of limb bone scaling, curvature and eccentricity in mammals and non-avian dinosaurs

Current theories of locomotor biomechanics are based largely on observations of extant terrestrial mammals. However, extant mammals are limited with respect to certain aspects of terrestrial locomotion, and this constraint has confounded interpretations of limb design in this group. Dinosaurs include a wide array of large and small bipeds, record several transitions between bipedalism and quadrupedalism, and show a unique covariation of cursoriality and body size. Thus, the wider applications of biomechanical theories may be tested by applying them to the limb bones of dinosaurs. A broad examination of hindlimb and forelimb bone scaling patterns in dinosaurs and mammals reveals several general similarities that provide insight into the general constraints acting on terrestrial locomotion, particularly in animals with parasagittally oriented limbs. Most limb bones scale with negative allometry in these two groups, and larger taxa tend to scale more negatively than smaller forms. However, the strongly negative scaling of large mammals is mostly restricted to ungulates, whose unusually short femora may be the result of constraints of cursoriality at large body sizes. Bipedal and quadrupedal dinosaurs scale very similarly, with most differences apparently resulting from size rather than posture. Bone curvature tends to decrease with increasing body size, while femoral midshaft eccentricity tends Co increase. Femoral midshaft eccentricity is explained as a general adaptation to mediolateral bending on parasagittal limb bones. These trends are more pronounced in dinosaurs than mammals; additional morphological constraints present in the dinosaurian hindlimb may contribute to this distinction.