In the osteopenic rat model, estrogen deficiency results in increased bone turnover with net bone loss occurring during cancellous modeling. However, estrogen-deficient rats treated with parathyroid hormone (PTH) experience a net gain of bone tissue due to the anabolic effects of PTH. To evaluate the possibility that local insulinlike growth factor I (IGF-I) production modulates the in vivo balance of bone formation and resorption in ovariectomized (OVX) estrogen-deficient rats and in OVX rats treated with PTH, we have studied the expression of IGF-I mRNA in cancellous bone osteoblasts using in situ hybridization techniques. Three-month-old virgin rats were subjected to sham surgery or OVX. Two weeks later, half the OVX rats began treatment with hPTH(1–34), 5 gmg/100 g body weight, 5 days/week for 4 weeks. All animals were killed at the same time, providing three groups: sham surgery alone; OVX alone; and OVX + PTH. Bone histomorphometry performed in undecalcified sections of tibial metaphysis confirmed that OVX rats had significantly (p < 0.05) increased bone surface formation rates (BFR/BS, μm³/μm²/year) with osteopenia while OVX + PTH rats had increased BFRBS with increased bone volumes compared to sham animals (p < 0.05). Decalcified tissue from all three groups contained immunoreactive IGF-I. Similar tissue sections were hybridized with an ³⁵S-labeled IGF-I antisense riboprobe. Evaluation of the specific signal over cancellous osteoblasts allowed a relative estimate of IGF-I mRNA transcript abundance in the three groups by counting silver grains per osteoblast, corrected for background activity. Despite evidence for increased BFR/BS, OVX animals had comparable frequency distributions of grain counts (after correction for background) to sham animals (88% of cells containing ⩽1 grain with a mean of 0.76 ± 0.28 grains/osteoblast vs. 81% containing ⩽1 grain with a mean of 0.61 ± 0.09, NS). However, OVX + PTH animals had over 40% of osteoblasts with ⩾2 grains per cell and twice the IGF-I mRNA abundance that either control group had (ANOVA, p < 0.001; OVX + PTH compared to either sham or OVX, p < 0.05). These data suggest tjat the anabolic effect pf PTH is associated with increased IGF-I mRNA steady state levels; while estrogen deficiency causes increase bone turnover, increased levels of IGF-I mRNA were not observed.