OVX monkeys treated for 18 months with 1 or 5 microg/kg/d teriparatide [PTH (1-34)] had significantly stronger proximal femora relative to ovariectomized controls. Teriparatide enhancement of cortical area, cortical width, and trabecular bone volume seemed to more than compensate for the dose-dependent increase in cortical porosity. Beneficial effects of teriparatide treatment on the proximal femur persisted beyond the treatment period and may extend to the marrow. We conducted a detailed quantitative analysis of the effects of teriparatide on the proximal femur of ovariectomized monkeys. Teriparatide increased bone mass, enhanced structural architecture, and strengthened the hip, despite increasing cortical porosity. Monkeys were treated with vehicle (sham or OVX controls), 1 microg/kg/day teriparatide [parathyroid hormone (1-34); PTH1], or 5 microg/kg/day teriparatide (PTH5) for 18 months or for 12 months followed by 6 months of treatment withdrawal (PTH1W and PTH5W, respectively). Excised proximal femora were analyzed by microCT, conventional histomorphometry, and biomechanics. The femoral neck showed significant reduction in trabecular bone volume (BV/TV) for OVX compared with sham, whereas PTH1 BV/TV was restored to sham levels and PTH5 BV/TV was greater than sham and OVX. The withdrawal groups had BV/TVs intermediate between sham and OVX. PTH1 had trabecular number (Tb.N) greater than OVX, and PTH5 Tb.N was greater than sham and OVX. The withdrawal groups had Tb.Ns intermediate between sham and OVX. No differences between groups were observed for trabecular orientation or trabecular thickness. Teriparatide dose-dependently increased bone formation rate and activation frequency in the femoral neck. Cellular composition analyses suggested a tendency of ovariectomy to increase adiposity of marrow by 100%, whereas PTH tended to reduce adipocyte number and increase osteoblast number compared with OVX. Analyses of the cortex showed dose-dependent elevation of cortical porosity, which was consistent with enhanced bone turnover with treatment. Cortical porosity was reduced after withdrawal of teriparatide, because PTH1W cortical porosity was lower than OVX, whereas PTH5W cortical porosity was intermediate between sham and OVX. Increased cortical porosity did not weaken the proximal femora. Biomechanics showed that ovariectomy weakened proximal femora compared with sham, but PTH1, PTH5, and PTH1W were stronger than OVX and not different from sham. PTH5W strength was intermediate between sham and OVX. Therefore, teriparatide had beneficial effects on the proximal femur, despite increasing cortical porosity. Cortical porosity did not adversely affect the mechanical integrity of the proximal femora, because enhanced cortical area and trabecular bone volume more than compensated for the porosity. Much of the beneficial effects of teriparatide were retained after 6 months withdrawal from treatment. PTH effects on the femoral neck were not limited to bone but may include inhibition of OVX-stimulated adiposity of the marrow.