A two-dimensional finite element model of the head of a rhesus monkey was built to simulate the head acceleration experiments done by Gennarelli and his colleagues. The purposes of the study were to better understand the mechanisms of traumatic subdural hematoma and to estimate its threshold of occurrence.The brain was treated as an isotropic homogeneous elastic material with and without structural damping and the skull was treated as a rigid shell. To simulate Abel et al.'s (1) experiments, the head was subjected to an enforced forward rotation around the neck. The loading had an initial acceleration phase followed by deceleration. During both acceleration and deceleration phases, high shear stress (and thus strain) occurred at the vertex, where the parasagittal bridging veins are located. The deformation of the bridging vein depended on its orientation relative to the direction of impact. Bridging veins that drain forward into the superior sagittal sinus would be stretched during the acceleration phase and would be compressed during deceleration. Therefore, subdural hematoma may have occurred during the acceleration phase in the primate experiments, in contrast to Gennarelli and Thibault's (2) belief that this phase could be neglected in analyzing the subdural hematoma data.The head motion could be reduced to equivalent rotational and translational components at the head center of mass. With a brain Poisson's ratio of 0.475, both components contributed equally to bridging vein deformation. At a Poisson's ratio of 0.49, rotational acceleration was the dominant factor. Translational acceleration, although less important, did influence deformations and should not be neglected in analyzing the tolerance data of subdural hematoma.The primate subdural hematoma data were replotted in terms of peak angular vs. peak tangential accelerations. The combined effects of tangential and angular accelerations on bridging vein deformation, as determined at the lower levels of test severity from the finite element analysis, were used to estimate tolerance thresholds for subdural hematoma in the experiments.