After treating the infected canals, we investigated to determine the influence of autogenous dentin chips accidentally applied as filling in the root apexes. Materials were 15 mandibular premolars and molars obtained from healthy, mature dogs. According to standard procedures, after pulp extirpation, canals were temporarily filled with sandarac cotton pellets and left otherwise untreated for 4 weeks. At this time, the degree of infection was ascertained. After root-canal enlargement, root canals were filled with No. 55 gutta-percha points; and a sealer (Neotriozinc Paste, AH26, or Sealapex) was applied. The animals were sacrificed at the conclusion of either a short term (2 weeks) or a long term (16 weeks), and histological studies were performed. Conclusions In short-term specimens, no formation of new hard tissue was observed at the apical dentino-cemental junction; and inflammatory changes in the periapical soft tissue were remarkable. In long-term specimens, hard-tissue formation had resulted in apical closure in 6 out of 8 instances; and periapical inflammatory changes had decreased. Relations between dentin-chip density and histopathological conditions were as follows: In most of the specimens that were evaluated as good, dentin chips were very densely applied. Insufficient numbers of dentin chips produced poor results. In other words, dense applications of dentin chips result in good prognoses. Consequently, when root-canal enlargement has been adequately performed, application of dentin chips to the apex of infected canals stimulates hard-tissue formation resulting in biological apical closure. In infected canals, however, the degree to which dentin chips are infected can be a major factor.