The purpose of this study was to examine three issues: the relationship between the length of time of skeletal pin insertion and bacterial infection, the connection between skin organisms and the intramedullary infecting agent, and the length of time necessary for the medullary canal to sterilize itself after pin removal. Skeletal pins were inserted in a group of dogs, and a direct correlation was found between the duration of time that the pins remained in the bone and positive intramedullary cultures. The infecting agent in 88% of the medullary cultures was also cultured from the skin. In a second group of dogs, the pins remained in place for 1, 2, 3, or 4 weeks, respectively. The pins were removed, and the pin tracts were permitted to heal for varying periods of time. The medullary canals were then cultured. No medullary canal was infected when the pin tracts were allowed to heal for greater than or equal to 3 weeks. Therefore, if pin tracts are allowed to heal for at least 3 weeks prior to performance of osteosynthesis, the chances of deep infection are probably lessened.