The purpose of this study was to compare clinical results of guided tissue regeneration(GTR) using either a nonresorbable ePTFE membrane or a resorbable membrane made from a synthetic copolymer of glycolide and lactide(PLGA) in the treatment of human class Ⅱ furcation defects. The ePTEE membranes were applied to 16 patients with maxillary molar buccal class Ⅱ furcation defects as Group I, PLGA membranes were applied to 15 patients with maxillary molar buccal class Ⅱ furcation defects as Group Ⅱ, ePTFE membranes were applied to 20 patients with mandibular molar buccal class Ⅱ furcation defects as Group Ⅲ and PLGA membranes were applied to 20 patients with mandibular molar buccal class Ⅱ furcation defects as Group Ⅳ and bone graft materials(DFDBA) were applied in all groups. Probing depth, gingival recession, clinical attachment level, tooth mobility and sulcus bleeding index(SBI) were measured at baseline, 3, 6 and 12months postoperatively. In addition, membrane exposure levels were measured at surgery, 1, 2 and 6weeks postoperatively and postoperative complications were evaluated. The results were as follows: In all groups, there were statistically significant differences in probing depth reduction, gain of clinical attachment and mobility reduction at values of 3, 6 and 12months postoperatively compared to values of baseline, whereas no significant differences in SBI except Group I and gingival recession(p<0.05). Membrane exposure levels were increased at 1, 2 and 6weeks postopratively compared to value of baseline in Group I(p<0.05). There were no statistically significant differences between ePTFE and PLGA membrane in probing depth, clinical attachment level and SBI. There were minimal gingival recession and membrane exposure in Group Ⅳ and pain and swelling were the most common postoperative complications in Group Ⅱ, Ⅲ(p<0.05). In conclusion, this study showed that both nonresorbable membrane and resorbable membrane were effective similarly in the treatment of class Ⅱ furcation defects, without statistical differences in clinical measurements.