1. In dogs, redrilling a screw hole after removal of the screw did not improve final repair as judged by both histological examination and biomechanical testing.
2. The screw holes, despite the radiolucency apparent on roentgenograms were rapidly filled with dense woven bone, and then were slowly remodeled into cortical bone.
3. The woven bone filling the screw hole eliminated the stress-concentrating effect of the hole within a surprisingly short time (four weeks).
4. In rabbits' femora the stress concentration induced by holes, holes retaining a screw and holes containing soft silastic plugs were removed by adaptive changes in the bone, and this recovery was complete by about eight weeks.
5. When a screw was removed after strength had been recovered, the bone was again weakened.
6. Holes filled with a soft silastic plug (which could always be freely slid in or out even after eight weeks) did not act as stress concentrators, so the bones must have developed a system for transmitting the forces around the hole.
7. The finding of greatest interest was that bone responded to eliminate stress concentration no matter what the nature of the inclusion.