The purpose of this randomized controlled trial was to investigate temporal skeletal responses to short-term high intensity physical activity. Twenty-eight normal active females [age: 20.7 +/- 2.1 yr (mean +/- SD)] were randomized into exercise (EX, n = 15) or control (CN, n = 13) groups. The exercise group trained 6 days/wk for 6 wk, which consisted of maximal isokinetic knee flexion/extension 3 days/wk, combined with 3 days/wk running. The purpose was to expose the tibiae to a period of abruptly increased loading forces. Tibial bending stiffness (EIMRTA), and serum concentrations of biochemical markers of bone formation [osteocalcin (OC)], and bone resorption [n-telopeptide of type I collagen (NTx)] were measured at baseline, 2 wks, 4 wks, and 6 wks. Isokinetic concentric knee extension/flexion peak torque, as well as total body and site-specific bone mineral density (BMD) were measured at baseline and 6 wk. After training, the exercise group significantly increased (p < 0.05) isokinetic concentric peak torque for the dominant (13.6%) and non-dominant (5.7%) quadriceps, as well as dominant (7.7%) and non-dominant (9.5%) hamstrings, compared to the controls. No differences for total body or site-specific BMD were noted. A two-way multivariate repeated measures ANOVA revealed no timeâ ¢group interactions for composite tibial bending stiffness [(EIMRTA); p = 0.57] or the biochemical markers of bone turnover [(OC and NTx); p = 0.15] across the four sampling periods. While there were no main effects for group, a trend for time (p = 0.051) for composite EIMRTA was observed. The exercise group demonstrated a 20% increase in EIMRTA from baseline (74.8 +/- 22.3 Nm2) to 6 wk (89.8 +/- 24 Nm2), compared to controls who demonstrated a 4% increase (Baseline 86.5 +/- 23.8 Nm2; 6 wk 90 +/- 23.7 Nm2). Significant group differences (p = 0.05) were noted for OC, but not NTx. Differences (p < 0.05) for OC were observed at baseline [13.2 +/- 2.4 ng/ml (CN), 15.6 +/- 2.7 ng/ml (EX)], and follow-up ANCOVA revealed no differences for subsequent sampling periods. Main effects for time were found for OC and NTx (p < 0.001). Main effects for time in OC were attributable to changes in the exercise group (p < 0.01) and NTx (p < 0.01), but not the control group.