We compared the abilities of alendronate and risedronate to reduce levels of urinary cross-linked N-telopeptides of type I collagen (NTX) in Japanese postmenopausal women. The patients were randomly divided into two groups (alendronate, 5 mg/day, n = 61; risedronate, 2.5 mg/day, n = 60). All patients had taken all medication prescribed for the first month and at least 90% of that prescribed for each of the following 6 months. Urinary NTX was measured at baseline, as well as at 1 and 6 months after starting treatment. According to the guidelines of the Japan Osteoporosis Society, the minimum significant change (MSC) for urinary NTX is defined as a 35% decrease from baseline and the cutoff level for a high risk of future fracture is 54.3 nmol bone collagen equivalent (BCE)/mmol.Cr. The NTX reduction rates at 1 and 6 months were greater with alendronate than with risedronate, but the difference was not significant. The rate of patients with a reduction in the MSC at 1 month was greater with alendronate than with risedronate, but the difference did not reach significance. Alendronate reduced NTX at 1 month significantly more in patients with a high risk of fracture than risedronate, but the difference was no longer significant at 6 months. The rate of MSC did not significantly differ between the two groups. In conclusion, alendronate decreases bone resorption markers more obviously and rapidly than risedronate, especially in high risk for fracture, but not significantly according to the guidelines of the Japan Osteoporosis Society.