Although animal studies suggest that there may be major differences between the effects of bisphosphonates and ovarian hormones on skeletal metabolism, whether this also holds for their actions in patients is unknown. To address this question, we compared the effects of 12 weeks treatment with HRT on bone turnover markers in osteopenic postmenopausal women with those of an amino-bisphosphonate. Women within 15 yr of the menopause, with a lumbar and/or femoral neck bone mineral density 1 S.D. below the predicted value, received either oestradiol valerate 2 mg and dydrogesterone 5 mg (E/D; n = 16) or aminohexane bisphosphonate 400 mg (AHBP; n = 9). Urine and serum samples were collected on two separate occasions before starting treatment, and 1, 2, 4, 8 and 12 weeks afterwards. To assess bone resorption, we measured the urinary deoxypyridinoline/creatinine ratio (DPD/crea), while serum alkaline phosphatase (ALP), osteocalcin and C-terminal propeptide of type I collagen (CICP) were analysed to assess bone formation. Repeated measures analysis of variance revealed a highly significant decrease in DPD/crea over the treatment period. Furthermore, this pattern of response differed significantly between the two treatment groups, since DPD/crea was maximally suppressed within 2 weeks of starting AHBP, while E/D showed little decrease until 8 weeks. AHBP was also found to suppress ALP, osteocalcin and CICP more rapidly than E/D, the former reducing these markers by 8 weeks, while E/D caused little inhibition even by 12 weeks. We conclude that, in the doses used in this study, AHBP appears to suppress bone turnover significantly more rapidly than E/D, suggesting that clinically important differences may exist in the effects of bisphosphonates and ovarian hormones on bone metabolism.