To evaluate systematically the biomechanical properties of 13 popular screw designs, ranging from 0.8 to 2.0 mm in diameter.Screws were characterized in terms of external, core, and drill diameter; cutting flute and head design; material; pitch, thread depth; and height of shank (unthreaded portion) and shank with plate. They were tested in standardized bone specimens (2 x 2.5cm slabs of fresh bovine femur) 1, 2, 3, and 4-mm thick. For each screw-bone thickness combination, 10 trials were conducted to determine push-out force (POF) and another 10 trials to determine insertion (IT) and maximum torque (MT) yielding a total of 1,040 tests.Among the 13 different screws, in 1-mm thick bone, both the lowest (108.5 N) and highest (294.9 N) POF were created by 2-mm screws (P 2 mm), two 2-mm screws showed 30% to 50% lower MT than the other same size screws (P < .001) because their head slots stripped easily. When all screws were considered together for a particular bone thickness, torque was strongly predicted by screw diameter (MT: r = .94, P < .001; IT: r = 0.92, P < .001). Screws with the same diameters varied significantly in IT because of differences in self-tapping cutting flute design.External diameter, unthreaded shank height, head slot, and self-tapping cutting flute design had the greatest impact on screw strength and efficiency in thin cortical bone. Thread depth, core diameter, and metal type did not affect performance significantly. Under these highly standardized in vitro conditions, the ideal 2-mm screw has an unthreaded shank that is as short as possible, and the pitch is about 0.8 mm. Additional aspects of a clinical situation beside holding strength must, however, be considered when choosing a screw.