To characterize the site-specific mechanical and histological properties in fracture repair and to relate these properties to the initial mechanical situation, an experimental fracture model was used in the metatarsus of 42 sheep. The mechanical situation of a transverse osteotomy was described by three gap sizes (1, 2, or 6 mm) and two amounts of strain (7 or 31%). An external fixator that allowed a defined axial movement provided control of these settings. Nine weeks following surgery, the healing area was dissected and tensile and compressive properties were measured in subregions of the fracture gap and the periosteal callus. The central, sagittal section was used for quantitative histology. We found the quality of the tissue along the osteotomy line to be most important for regaining mechanical stability. Increasing the size of osteotomy gaps resulted in poorer mechanical and histological qualities, and the repair process was less complete. Interfragmentary strain did not significantly influence the repair process. The smaller strain levels had already stimulated the secondary repair process, and this stimulatory effect could not be further enhanced by increasing the amount of strain. Our finding that large gaps between bone segments were not as well healed as were smaller gaps suggests that it is advantageous to avoid large gaps in fracture treatment.