Effects of fluoride on bone strength and cortical bone mass remain controversial. We compared 9-month, low-dose sodium fluoride (NaF) treatment with estrogen replacement therapy. Female Wistar rats 4.5 months old were divided into baseline, sham-operated (sham), sham-treated with NaF at 0.5 mg NaF/kg/day in drinking water, and ovariectomy (OVX), OVX treated with NaF and with estrogen. Bone mass was measured by dual X-ray absorptiometry (DXA) in vitro. Dimensions of the first lumbar vertebral body (L1) were determined by radiogrammetry. The right femur was processed undecalcified to obtain a midshaft cross-section to determine cross-sectional moments of inertia (CSMIs). L1 compressive test and left femoral torsional test were performed. OVX induced significant bone loss in L1 and femoral midshaft. Bone mass was increased to a greater extent in NaF-treated rats than in rats receiving estrogen replacement therapy. Femoral CSMIs in OVX rats, both L1 sizes and femoral CSMIs in NaF-treated rats, were significantly increased. Estrogen treatment had the least dimension expansion. OVX significantly decreased L1 compressive variables. There was no statistical difference in compressive parameters between NaF-treated groups and controls. OVX significantly increased femoral torsional strength but NaF treatment did not. Bone fluoride content was significantly increased after treatment with NaF. No significant difference in bone mineralization degree (ash and calcium) was found between treated and control rats. The discrepancy that an increase in bone mass and geometric properties in both trabecular and cortical bones by low-dose, long-term NaF treatment did not increase vertebral strength nor proportionally improve femoral strength indicated that the bone intrinsic biomechanical properties could be changed by NaF treatment.