The purpose of this study was to evaluate the sealing ability of various canal filling methods. Palatal roots of ninety extracted human maxillary molar teeth were resected at cementoenamel junction. Eighty of them were randomly assigned to four experimental groups, ten were served as positive and negative controls. All canals were prepared to # 40 using Profile. Experimental groups were obturated by lateral condensation technique, Thermafil technique, Continuous Wave of Condensation Technique, and down-pack & back-fill technique of Obtura-II, each with root canal sealer. Control groups were not obturated. Teeth were immersed in resorcinol-formaldehyde resin for 5 days at $4^{circ}C$, and the resin was allowed to polymerize completely for 4 days at room temperature. Teeth were then ground horizontally at 1.5mm(level 1), 2.5mm(level 2), 3.5 mm(level 3) from the anatomical apex and examined with a stereomicroscope at ${ imes}40$ magnification. The gap between the canal wall and the filling material, which was filled with the resin, was measured at each of the three levels. Each ratio of leakage was obtained by calculating the ratio of the area of the resin to the total area of the canal and was analyzed statistically (Rank-sum test). The results were as follows : 1. At the level 1, there was the greatest leakage in the Thermafil group and Obtura-II group, and the difference between the Obtura-II group and Continuous Wave of Condensation Technique group was statistically significant(p<0.05). 2. At the level 2, there was the least leakage in the Continuous Wave of Condensation Technique group, but there was no statistically significant difference between each group (p>0.05). 3. At the level 3, there was no statistically significant difference between each group(p>0.05).