To test the hypothesis that the rate of change of strain to which a bone is subjected is an important determinant to the subsequent functionally adaptive modeling response, the ulnae of growing male rats were subjected to dynamic axial loading in vivo for a short period each day over 2 weeks. Due to the longitudinal curvature of the ulna, such axial loading leads to both compression and bending. The left ulna in three groups of rats was loaded cyclically between 1 and 20 N in a trapezoidal pattern to produce dynamic, longitudinal compressive strains of −0.004 (−4000 microstrain) at the medial midshaft with one of three strain rates:​ low (±0.018 sec⁻¹;​ n = 7);​ moderate (±0.030 sec⁻¹;​ n = 7);​ and high (±0.100 sec⁻¹;​ n = 8). These strain rates span the range recorded from strain gauges bonded to the bone at this site during a variety of normal activities. At the end of the experiment, the loaded ulnae were slightly, but significantly, shorter than their contralateral controls (2.7% to 5.6% mean change in length;​ p < 0.0001). This effect was most marked at lower strain rates, associated with an increased load-bearing time. The pattern of adaptive modeling along the bone shaft was similar for all groups, each showing a reduced rate of periosteal expansion proximally, and increased periosteal new bone production distally. This distal increase was achieved through enhanced periosteal bone formation on the lateral (tension) cortex, and arrest of resorption, with conversion to formation on the medial (compression) surface. The modeling response to axial loading therefore involves complex location-dependent increases and decreases in both formation and resorption. The high-strain-rate group demonstrated a 54% greater osteogenic response than the moderate-strain-rate group, which in turn showed a 13% larger response than the low-strain-rate group. Rate of strain change is therefore a major determinant of the adaptive osteogenic/antiresorptive response to mechanical load. Across the physiological range, a high rate of strain change provides a greater osteogenic stimulus than the same peak strain achieved more slowly.