The purpose of this study was to investigate the reliability of an implantation test using Teflon-simulated root canals for evaluating the pathogenicity of root canal bacteria. Models including suspensions of lyophilized strains (Actinomyces israelii, Streptococcus faecalis, and Porphyromonas asacharolyticus) were implanted in rat subcutaneous tissue for 1 wk, and histological changes were observed. Severe inflammation occurred around the models. Among them, P. asaccharolyticus induced the severest inflammatory response. Further study using P. asaccharolyticus was conducted to compare the implantation test's ability to evaluate pathogenicity with that of an injection test at 1, 2, or 4 wk. Tissue injected with a bacterial suspension showed no clear response through the experimental periods, whereas tissue around implantation sites showed a severe response at I wk. However, the inflammatory response subsided at later stages. Consequently, further improvement is needed to investigate pathogenicity for long periods.