We examined the effects of nandrolone decanoate (25 mg im every 3 weeks) on bone mass, serum biomarkers, and bone histomorphometric endpoints in 52 female cynomolgus macaques randomized into four treatment groups: (1) sham-ovariectomized (sham); (2) ovariectomized + placebo for 2 years (ovx); (3) ovx + nandrolone decanoate for 2 years (Nan); and (4) ovx + nandrolone decanoate beginning 1 year after ovx (dNan). Serum alkaline phosphatase (ALP), osteocalcin, and tartrate-resistant acid phosphatase (TRAP) were assayed every 3 months, and X-ray densitometry of the lumbar spine was done every 6 months. Fluorochrome-labeled iliac biopsies collected at baseline and 1 year, and lumbar vertebrae and midshaft femur collected at 2 years, were evaluated histomorphometrically. Body weight increased over 50% with administration of nandrolone. After 2 years, ovx animals had lower spinal BMC and BMD than all other groups. Ovx animals also had higher bone turnover rates than all other groups, as indicated by higher levels of the serum and urine biomarkers, and by at least twofold higher label-based bone formation rates in the femur diaphysis and in both cancellous and cortical bone of the ilium and vertebral bodies. Nandrolone-treated animals had similar serum estradiol levels as the sham animals, presumably due to conversion of endogenous or exogenous androgens. The effects of nandrolone on bone in this experiment are consistent with estradiol action and may be attributable to the increased serum estradiol. Despite >50% higher body weight, nandrolone-treated, ovariectomized animals did not have higher bone mass than sham animals.