Purpose: Many researches showed loss of alveolar bone in fresh extraction socket and even in case of immediate implant placement. The aim of this study was to evaluate the effect of non-resorbable barrier membrane on the change of buccal and lingual alveolar bone in immediate implant placement into periapically infected extraction sockets. Materials and methods: Immediate implants were placed into artificially induced periapical lesion of mandibular premolars after complete debridement using buccal bone defect made by a 6mm trephine bur in 4 mongrel dogs. Before flap repositioning, a non-resorbable barrier membrane was placed on the buccal defect in the experimental group. No membrane was placed in the control group. In 12 weeks after placement, the dogs were sacrificed and undecalcified histologic specimens were prepared. The vertical distance from the smooth-rough surface interface(SRI) to gingiva, 1st bone contact and bone crest were measured in buccal and lingual side. The horizontal thicknesses of gingiva and bone at 0, 1, 2 and 3mm below SRI were measured. Results: The buccal bone was resorbed more than lingual bone in both groups and there was statistical significance(p<0.05). The distances from SRI to 1st bone contact were $2.45{pm}2.35;mm$ in experimental group and $4.49{pm}3.10;mm$ in control group. In all vertical level, lingual bone was thicker than buccal bone(p<0.05). Conclusion: Buccal bone was reduced more than lingual bone in immediate implant placement into periapically infected extraction sockets. Placement of non-resorbable barrier membrane reduced the buccal bone resorption. However there was no statistical significance.