Running is a popular activity of choice for many, and a necessity for athletes and military personnel. The positive physiological adaptations associated with running are well established, and these adaptations can only be exploited if runners remain free from overuse injury. This dissertation utilized a combination of experimentation, musculoskeletal modeling, and a probabilistic model of bone damage, repair, and adaptation to investigate internal structural loading of the lower extremity during running. Specific emphasis was placed on stress fracture development, a common overuse injury that results, in part, from the mechanical fatigue of bone. A series of studies were conducted that addressed the influence of speed on lower-extremity contact forces during running, the relationship between internal femoral loads and stress fracture development, and changes in the probability of tibial stress fracture with practical alterations in kinematics and running mileage. The findings of these studies can be summarized as follows: 1) musculoskeletal models provide meaningful non-invasive estimations of internal structural loads in healthy young adults; 2) joint contact forces increase with speed, 3) stress fractures tend to occur at femoral locations experiencing the largest mechanical loads; 4) the probability of tibial stress fracture increases with stride length and running mileage for a given speed; and 5) the probability of tibial stress fracture increases with running speed for a given mileage. Ultimately this information can be used to develop running regimens that maximize the positive adaptations associated with running and minimize the potential for overuse injury and stress fracture development.