The aim of this study was to evaluate the effect of deproteinized bovine bone graft material on new bone formation in a guided bone regeneration model system. In 20 rabbits, a periosteal skin flap was raised uncovering the calvaria. A form stable hemispherical dome made of poly-lactic acid (PLA) was placed onto the roughened calvaria. Prior to placement, the dome was either filled with peripheral blood alone (control group, 8 rabbits), or with blood and OsteoGraf/N-300 (test group, 12 rabbits). The wound was closed for primary healing. Morphometric assessment of 1- and 2-month undecalcified histologic specimens revealed better tissue fill in the test domes at 1 month (test 99%, control 55%) (P < 0.05) and 2 months (t, 100%; c, 82%). The fraction of the new bone within the regenerated tissue was higher in the test specimens at 1 month (t, 22%; c, 12%) (P < 0.05) and 2 months (t, 34%; c, 24%). The fraction of the entire space underneath the domes occupied by bone was higher in the test at 1 month, but higher in the controls at 2 months. The fraction of the bone substitute material in contact with bone increased from 1 month (34% +/- 14) to 2 months (45% +/- 5). The surface fraction of osteoblast layers was tendentially higher in the test at 1 month but higher in the control specimens at 2 months. In both test and control, initially woven bone was formed which underwent subsequent remodeling. Cellular degradation of the deproteinized bone graft was frequently detected. It is concluded that deproteinized bovine bone mineral has osteoconductive properties and can initially accelerate new bone formation during guided bone regeneration by increased recruitment of osteoblasts.