Over 1 million bone grafting procedures are performed in the U.S. each year. A significant portion end in delayed bone healing. Autologous bone is frequently used in these procedures, but its use is limited by a finite supply and the extraction procedure may cause additional patient morbidity. This study's purpose is to design a synthetic bone graft that releases functional growth factors in a sustained manner, and serves as a bioactive substrate for bone growth. A material was synthesized through a room temperature sol-gel process that allows functional proteins to be incorporated inside the material. Two xerogel compositions, 100% SiO₂ (S100) and 70% SiO₂-25% CaO-5% P₂O₅ (S70), were evaluated for their ability to release incorporated proteins similar in size to bone growth factors. Both compositions were found to release protein for times greater than 4 weeks. A mathematical model based on diffusion was constructed to predict the release of proteins. Bioactivity was evaluated in vitro by characterizing the kinetics of calcium phosphate (Ca-P) formation when the materials were immersed in a solution similar in electrolyte content to body fluids. S70 was found to form Ca-P within 4 hours, while St00 took several days. The release of active TGF-β1 (a bone growth factor) was quantified. There was less than 1% release of TGF-01 from S200, with release in sufficient quantity to affect osteoblast cell function. There was much less release of TGF-β1 from S70, possibly due to Ca-P formation which might have hindered protein release. The response of stromal marrow ceils (RSMC) cultured on S70 with and without BMP (a protein that initiates bone formation) was evaluated through biochemical assays. RSMC showed greater alkaline phosphatase staining to S70 pre-treated for hydroxyapatite formation and protein attachment. S70 and BMP addition increased greatly the differentiation of RSMC to an osteoblastic phenotype over controls, but decreased cell proliferation. S70 served as a bioactive substrate for osteoblastic cell growth and released active forms of BMP. In vivo testing needs to be done with this material to evaluate its use as a synthetic bone graft.