In order to study bone growth conducting capacities of new biomaterials under standardized conditions, a goat model was developed based on a canine model by Soballe. Titanium alloy implants with and without a hydroxyapatite coating were used as positive and negative controls, and these were implanted with a circumferential gap of one millimeter in the spongious bone of the knee condyles of two groups of four goats. These goats were sacrificed at 6 and 25 weeks. A second experiment was done on two groups of four goats with the same type of titanium alloy and hydroxyapatite-coated implants as controls and with Polyactive 55-45 coated titanium alloy implants for testing. These goats were sacrificed at 9 and 25 weeks, respectively. Qualitative and quantitative differences in gap healing were evaluated through light microscopy, and initiation and direction of bone apposition were determined with fluorescence microscopy. Apposition of bone was seen directly on all hydroxyapatite surfaces and on some of the noncoated titanium alloy surfaces. The difference between the percentage of bone growth on the titanium alloy implants and the hydroxyapatite-coated implants appeared to be divergent in time: the bone growth on the noncoated implants declined after 9 weeks in contrast to the steady increase of bone growth on the hydroxyapatite-coated implants towards the 25 week follow-up time (p = 0.02). No significant difference was found between the first and the second experiment: apposition of bone on the implants differed only 6.6% on a scale of 0% to 100%. Only scarce bone growth was seen on the polyactive-coated implants in this model. The newly tested Polyactive 55-45 coating apparently needs initial bone contact for bone-bonding and therefore showed hardly any direct bone formation on its surface. The clear differences in the reaction of bone to the coated and noncoated implants in this goat study and the reproducibility of these reactions of bone to the different controls indicate the sensitivity of the currently used animal model and its suitability for use as a bioactivity assay.